Official Journal of NASPEM and the

Pediatric Exercise Science, 2013, 25, 186-197

© 2013 Human Kinetics, Inc. European Group of PWP
www.PES-Journal.com
ORIGINAL RESEARCH

Role of Free Testosterone in Interpreting

Physical Performance in Elite Young
Brazilian Soccer Players
Alexandre Moreira, Arnaldo Mortatti, Marcelo Aoki,
Ademir Arruda, Camila Freitas, and Christopher Carling

This study investigated the contribution of salivary testosterone (sT) concentration,

years from peak height velocity (YPHV) and height by body mass interaction on
jumping performance (Countermovement jump; CMJ) and aerobic fitness (Yo-Yo
intermittent endurance test, level 1) in young elite soccer players. Forty-five par-
ticipants (age: 12.5 ± 0.5y; body mass: 48.6 ± 10.2kg, height: 155.7 ± 10.0cm)
belonging to a top level Brazilian soccer club were evaluated at four time points
across a single semester. None of the assessed players had reached PHV. The data
from the four evaluations were averaged and multiple linear regression analysis
conducted. For CMJ, the model explained 42.88% of the variance (R2 = 42.88; p
< .000); sT concentration was the primary contributor (R2 = 32.84) and the YPHV
contributed 9.95% of the variance. The model explained 28.50% (p < .000) of the
variance in Yo-Yo. The sT was the primary and single significant contributor (R2
= 21.32). A significant difference was noted between high and low testosterone
groups divided a posteriori to CMJ performance (t = 3.35; p = .001). These results
suggest an important role for hormonal status in interpreting physical performance
in preadolescent soccer players.

Measures of performance in tests of functional capacities such as endurance,

speed, power and strength are habitually used by coaches and researchers to investi-
gate physical attributes in young soccer players (16). Several authors have reported
large variations across individuals in physical performance measures as well as in
morphological variables, soccer-specific skill tasks, and in motor proficiency during
puberty with ability closely associated to maturity status (24,32). Indeed, in youth
soccer, advanced physical development is considered an advantage and less devel-
oped players (considered late maturers) are frequently disadvantaged throughout
the talent identification and selection process (15). Some talented young soccer
players might be overlooked for selection into elite academies simply because they
are behind in biological maturity and less developed physically.

Moreira, Arruda, and Freitas are with School of Physical Education and Sport, University of São Paulo,
São Paulo, Brazil. Mortatti is with Physical Education and Sports Institute, Ceará Federal University,
Fortaleza, Ceará, Brazil. Aoki is with School of Arts, Sciences and Humanities, University of São
Paulo, São Paulo, Brazil. Carling is with EMSP, LOSC Lille Métropole Football Club, Camphin-en-
Pévèle, France.

186
 Testosterone and Performance in Soccer   187

The endocrine system influences anthropometric characteristics and body

composition (6,17) which in turn can affect neuromuscular capacities and athletic
performance (5,10). During adolescence the endocrine system is subject to major
alterations and many hormones such as growth, thyroid and adrenal hormones are
involved in the maturational process (30). Hypothalamic-pituitary-gonadal axis
activation is the main factor that induces progressive secretion of testicular sex
hormones which ultimately are responsible for the changes associated with biologi-
cal, morphological and psychological dimensions that occur during puberty (14).
Moreover, as highlighted by Marshall and Tanner (26), the timing of the adoles-
cent spurt, associated with genital and pubic hair development is in line with the
assumption that androgens from the testes play a key role in accelerating growth.
In recognition that testosterone helps to indicate acute and chronic changes in
neuromuscular capacity for explosive performance (8), it is reasonable to suggest
that in youth soccer players, interindividual differences in free testosterone levels
are associated with variations in physical performance. Indeed, in young players
participating in the same age category, there are significant interindividual differ-
ences in puberty-related hormonal patterns (9). Late physical development could
therefore be associated with lower endocrine hormonal levels. In the practical set-
ting, the issue of endocrine hormone levels, notably testosterone, might therefore be
taken into consideration by medical, fitness and coaching practitioners. Observed
differences in physical performance among individuals could be explained, at least
in part, by androgen circulating levels, which in turn might aid in informing the
strength and conditioning process in young players.
Unfortunately, there is a lack of information regarding the relationship between
pubertal development, endocrine hormone concentrations and functional performance
measures in the formal training processes of young preadolescent soccer players.
Biological maturation associated with hormone status could be considered concomi-
tantly with functional performance measures when selecting and training young soccer
players (21). Indeed, knowledge of individual hormonal status in conjunction with
maturational status measures across different periods during preadolescence and ado-
lescence might aid decision-making during long-term development training programs.
Therefore, the aim of this study was to assess the contribution of salivary tes-
tosterone (sT) concentration, in addition to maturity status (years from peak height
velocity; YPHV) and anthropometric measures to variations in jump height and
aerobic performance in elite young Brazilian soccer players. The authors suggest
that a model integrating these measures could help to explain potential variations
in physical performance.

Methods
Participants
Forty five young soccer players belonging to one of the major professional soccer clubs
in Brazil voluntarily agreed to participate in the study. After ethics approval by the local
Institutional Review Board the protocols were explained in detail. Written informed
consent was obtained from each participant and their respective parents or guardians.
At senior level, the present club has previously won five National Championships
and has participated in the FIFA Club World Championship. It has several young teams
188  Moreira et al.

distributed into age-categories with the aim of developing professional soccer players.
Players are divided into categories according to chronological age. Two of these cat-
egories were selected for investigation: under 12 (U12; N = 19) and under 13 (U13; N
= 26). The two age categories were chosen as these are considered essential stages in
the long-term development training process. Indeed, in these two age groups, players
are either selected or not to continue their development in the present professional club.
The 45 players were characterized as a single group for all analyses as no sig-
nificant differences were observed between U12 and U13 categories in testosterone
concentration, years from PHV and performance in functional tests. None of the
participants reported any injury or illness over the course of the study. All had either
been engaged in the club’s regular soccer training program for at least thirty months
after recruitment during an annual selection scheme or had been recruited by coaches
and staff members from another team. The players participated in three training
sessions (90–120 min) and one game per week, the latter taking place at weekends.

Experimental Approach to the Problem

Resting saliva sampling to determine testosterone levels and anthropometric and
performance measurements were performed on four occasions by the same researcher
and at the same time on each testing day. Testing was conducted in the same afternoon
as regular training sessions and completed before training commenced. All partici-
pants were familiar with all the test procedures which are habitually performed by
the club’s coaches during the season. Regarding saliva collection, the participants
were familiarized with the procedures before the start of the investigation.
The investigation period was structured as follows: a 12-week preparatory
training phase (preseason training), a 7-week competitive and a 2-week detraining
phase. All measurements were performed on four occasions for each participant;
before (T1) and after (T2) the preseason phase, and after the competitive (T3) and
detraining phases (T4). The competitive phase included 7 official matches. All
measurements were performed in the afternoon (14:00), at least 90 min postlunch.
Anthropometric measures included height, body mass, sitting height. Height was
measured using a digital platform scale (Welmy©, São Paulo, Brazil), calibrated,
graduated from zero to 150 kg; level of precision of 0.1 kg; sitting height using
a vertical stadiometer, 210 cm long and 0.1 cm precision (Welmy©, São Paulo,
Brazil). The thickness of five skinfolds thickness (triceps, subscapular, suprailiac,
abdominal, and medial calf) was measured using a Harpenden© skinfold caliper (St.
Albans, UK) by the same investigator to the nearest 0.2 mm, with the participants
in the reference standing position. All skinfold measurements were taken on the
right side of the body. Body fatness was subsequently estimated using the skinfold
equation proposed by Slaughter et al. (35).

Procedures
Maturity Status
Maturity status was assessed in accordance with methods described by Mirwald,
Baxter-Jones, Bailey, & Beunen (28). This approach has been proposed as a somatic
maturity indicator and predicts the time before or after PHV from measures of age,
height, body mass, sitting height, and leg length. It has been used in adolescent
 Testosterone and Performance in Soccer   189

male soccer players (23), swimmers (34), gymnasts (22,31), figure skaters and
ballet dancers (29) and schoolchildren (20) among others. The somatic maturational
assessment by means of years from peak height velocity (YPHV) using anthropo-
metric variables, is predicted in males using the following equation (23,28): [- 9.236
+ 0.0002708·Leg Length and sitting height interaction—0.001663·age and leg
length interaction + 0.007216·age and sitting height interaction + 0.02292·weight
by height ratio]. By using the age of PHV as a maturational indicator and describing
the measurement as years from PHV, the subtraction of the age of PHV from the
chronological age at the time of the measurement allows the differences in years
to be defined as a value of maturity offset (28).

Countermovement Jump Height (CMJ)

CMJ was performed on a contact mat. The participants lowered themselves from an
initial standing position to a self-selected squat position and performed a vertical
jump keeping their hands placed on their hips as quickly as possible with maximal
effort. While no restrictions were placed on the knee angle attained during the
eccentric phase of the jump, participants were instructed to maintain straight legs
during the flight. The jump mat provides valid measures of jump height compared
with a criterion system (r = .967; 18). Pilot testing indicated that the jump mat
system also provides reliable measures (CV’s < 2.0%).

Yo-Yo Intermittent Endurance Test

The Yo-Yo intermittent endurance test, level 1 (3), was performed in line with the
procedures adopted by Figueiredo et al. (11) in their investigation in young male
soccer players. Groups of eight to ten participants were evaluated each time. Four
assistants (researchers and staff members) were each responsible for two or three
participants. This test involves 5–20 s intervals of running interspersed by regular
short rest periods of 5 s. Briefly, participants ran outdoors on a grass flat surface
between 2 markers set at a 20 m distance apart. In addition to these two markers,
a third marker is positioned 2.5 m behind of the start marker. Starting at the first
marker, the athletes ran to the 20 m marker on an audio cue from a CD player,
before turning and running back toward the first marker, which they had to reach by
the next signal. When the start marker was reached, the players continued forward,
jogging (lower tempo runs), until they reached the third marker, and then ran back
to the start marker. They then stopped and waited for the next signal. The start-
ing speed at level 1 was 8.0 km·h-1, and throughout the test, the speed increased
progressively. The test was finished when the athletes were not able to maintain
the indicated speed for two trials. A warning was given when they were unable
to reach the start marker, and on a second warning the athletes were instructed to
stop. The total distance covered was recorded in meters.

Saliva Collection and Hormone Assessments

The players provided saliva samples approximately 15–20 min before the anthro-
pometric and performance measures. The participants abstained from food and
caffeine products for at least 1.5 hr before saliva collection. In a seated position
and with the head tilted slightly forward, unstimulated saliva samples were col-
lected by passive drool into sterile 15-ml centrifuge tubes over a 5 min period. The
190  Moreira et al.

saliva samples were then stored at -80 °C until assayed for testosterone. Salivary
testosterone concentrations were measured in duplicate using an enzyme-linked
immunosorbent assay (ELISA, Salimetrics testosterone expanded range kit) in
accordance with the manufacturer instructions. Briefly, a microtitre plate is coated
with rabbit antibodies to testosterone. Testosterone in standards and unknowns com-
petes with testosterone linked to horseradish peroxidase for the antibody binding
sites. After incubation, unbound components are washed away. Bound testosterone
peroxidase is measured by the reaction of the peroxidase enzyme on the substrate
tetramethylbenzidine (TMB). This reaction produces a blue color. A yellow color
is formed after stopping the reaction with sulfuric acid. Optical density is read on
a standard plate reader at 450 nm. The amount of testosterone peroxidase detected,
as measured by the intensity of color, is inversely proportional to the amount of
testosterone present. The average intra-assay coefficient of variation for testosterone
assay verified in the present analysis was 3.2%. The minimum detection limit for
the T assay, in accordance with the manufacturer, was 21 pmol/L.

Statistical Analysis
SPSS 17.0 was used for all statistical analyses. The distribution of the data were
analyzed by the Shapiro-Wilk test and Mauchly’s Test of Sphericity performed.
Descriptive statistics were calculated for the total sample (n = 45) and expressed
as mean ± SD. An ANOVA with repeated measures was used to compare the four
time points (T1, T2, T3, and T4) for salivary testosterone, CMJ, Yo-Yo intermittent
endurance test, body mass, PHV, and height. Tukey’s HSD post hoc test was used
if necessary. In the case of violation of the assumption of sphericity, the signifi-
cance was established by utilizing the Greenhouse-Geisser correction. As results
from ANOVA showed no significant differences between moments, data on each
variable from the four distinct periods were averaged and a multiple linear regres-
sion analysis was conducted to estimate the relative contribution of testosterone
levels, YPHV, and height by body mass interaction to variation in Yo-Yo and CMJ
performance. Since height and body mass are highly related (24,25), residuals
(individual values minus the mean) were used in the regression. A height by body
mass interaction was calculated as the product of the residuals for height and body
mass as proposed by Malina et al. (24,25). This method reduces the collinearity
among the independent variables, thus these are considered more stable predictors
of the dependent variables (24).The stepwise regression protocol was used.
Finally, as we were interested in determining the effect of testosterone con-
centration on physical performance, participants were divided into low and high
testosterone groups according to the median-split technique. An independent t test
was used to compare CMJ and aerobic performance between groups. The level of
significance for all statistical analyses was set at p ≤ .05.

Results
Descriptive statistics from the four individual assessments are presented in Table
1. No significant differences were observed for any assessed variables across the
4 assessments.
 Testosterone and Performance in Soccer   191

The results from multiple linear regression analysis showed that relative to
CMJ performance, the model explained 42% of the variance (R2 = 42.88). Salivary
testosterone (sT) concentration was the primary contributor (R2 = 32.84) and the
PHV added 9.95% of the variance (Table 2).
Regarding the aerobic performance test, the model explained 28.50% of the
variance. Running performance in the Yo-Yo test was explained solely by sT con-
centration (R2 = 21.32). This was the only variable which attained a significant F
to enter into the model (Table 2). A significant difference was observed to CMJ (t
= 3.35; p = .001) between low and high testosterone groups, divided according to
the median-split technique. However, the groups did not differ in Yo-Yo perfor-
mance (Table 3).

Table 1  Descriptive Statistics (mean± SD; SD) and Results

From ANOVA Across the Four Assessed Time Points
F
Variable T1 T2 T3 T4 (p-value)
Testosterone 53.9 ± 44.3 50.6 ± 30.0 49.2 ± 30.5 58.6 ± 46.7 0.20 (0.67)
(pg·ml-1)
YPHV (years) -2.2 ± 0.8 -2.1 ± 0.8 -1.9 ± 0.9 -1.9 ± 0.8 1.10 (0.34)
Height (cm) 154.9 ± 10 155.9 ± 10 157.7 ± 10 158.2 ± 10 1.10 (0.87)
Body mass 47.5 ± 9.9 48.7 ± 10.2 49.6 ± 10.6 49.6 ± 10.7 0.39 (0.75)
(kg)
Body fatness 15.9 ± 6.3 15.7 ± 6.6 16.2 ± 6.3 16.1 ± 6.3 0.15 (0.92)
(%)
CMJ (cm) 35.4 ± 4.9 35.1 ± 5.0 35.9 ± 5.3 34.8 ± 5.2 0.35 (0.78)
Yo-Yo (m) 1626 ± 382 1752 ± 306 1747 ± 302 1658 ± 317 1.66 (0.17)
YPHV = years from peak height velocity; Yo-Yo = Yo-Yo intermittent endurance test—level 1; CMJ
= Countermovement vertical jump.

Table 2  The Main Predictors of Performance on CMJ and Yo-Yo

Functional R2
test Predictor R2 change β p r
CMJ Salivary 32.84% 32.84% 0.41 < 0.00 0.57 (CMJ
testosterone (sT) and sT)
sT, YHPV 42.88% 9.95% 0.35 < 0.00 0.54 (CMJ
and YPHV)
Yo-Yo Salivary 21.32% 21.32% 0.38 < 0.00 0.46 (Yo-Yo
testosterone (sT) and sT)
CMJ = Countermovement vertical jump; YoYo = Yo-Yo intermittent endurance test—level 1
YPHV = years from peak height velocity; r = Pearson correlations between predictors and CMJ and
Yo-Yo.
192  Moreira et al.

Table 3  CMJ and Yo-Yo Performance to Low and High Testosterone

Groups Divided According to the Median-Split Technique (Mean± SD)
Low-testosterone High-testosterone
group group t value (p-value)
CMJ (cm) 33.0 ± 4.7 37.4 ± 4.0 3.35 (p = 0.001)
Yo-Yo (m) 1659 ± 345 1731 ± 185 0.86 (p = 0.38)
CMJ = Countermovement vertical jump; YoYo = Yo-Yo intermittent endurance test—level 1

Discussion
The aim of the current study was to assess the contribution of salivary testosterone
(sT) concentration, years from peak height velocity (YPHV) and anthropometric
measures to variation in vertical height (countermovement vertical jump; CMJ) and
aerobic performance (Yo-Yo intermittent endurance test; Yo-Yo) in elite preadoles-
cent Brazilian soccer players. The major findings were that salivary testosterone
concentration emerged as the main contributor of the variance for both aerobic and
vertical jump performance. However, its role was more prominent in vertical jump
performance which is an indicator of ability to perform explosive movements in the
vertical plane. In the countermovement jump test (CMJ) the high-testosterone group
performed significantly better than the low-testosterone group. Finally, the stabil-
ity observed in salivary testosterone concentration across the investigation period
revealed that both seasonality and training content did not affect sT concentration.
The role of hormonal status in influencing youth athletic performances is
receiving increasing attention by the sport science community (4,9). Baldari et al.
(2) highlighted its importance in youth sports, particularly in talent identification and
long-term training processes. Di Luigi et al. (9) reported that chronological age may
not correspond to biological age, and consequently large interindividual variability in
many parameters such as pubertal stage, testosterone concentrations, among others,
could be observed within the same category of athletes habitually selected according
to chronological age. Di Luigi et al. (9) reported a significant relationship (r = .33; p
< .01) between baseline salivary testosterone and chronological age in young soccer
players (9–17 years, n = 110). However, the low coefficient of determination observed
might indicate that variables other than chronological age require consideration
thereby aiding in explaining the correlation with salivary testosterone.
The influence of testosterone concentration on physical performance may be
associated with its role in mediating the development of the neuromuscular system
as well as with short-term changes in neuromuscular capacity, in particular for
explosive performance (8,38). The development of the nervous system seems to be
strongly associated with performance in endurance and power (37) being mediated
by hormonal changes. Testosterone also promotes neural growth, myelination and
axonal conduction velocity (36). Indeed, as pointed out by Styne and Grumbach
(14), the progressive secretion of the testicular sex hormones are responsible for the
changes associated with biological, morphological and psychological dimensions
occurring during puberty. Therefore, it seems that functional performance depends
on the development status of the neuromuscular system, which in turn, might be
 Testosterone and Performance in Soccer   193

related to the development of other systems (i.e., metabolic, cardiorespiratory)

while being mediated by hormonal status.
Testosterone exerts anabolic actions upon muscle tissue, therefore contribut-
ing to muscle growth, and improving strength-related performance (7). Among its
indirect anabolic actions, testosterone stimulates the secretion of other hormones
such as growth hormone. As salivary testosterone is suggested to provide a better
indication of bioavailable testosterone, that is potentially available to tissue in com-
parison with plasma testosterone concentration (1), the influence of sT on changes
in functional performance verified in the current study might be explained by its
key role in mediating neuromuscular performance.
The present results add novel information to the literature concerning the role
of the sT concentration on physical performance in preadolescent soccer players.
At early gonadarche, the pulsatile release of luteinizing hormone (LH) and follicle-
stimulating hormone (FSH) leads to increases in testosterone concentration (33).
In future studies, the assessment of these hormones at different time points during
a day, could improve knowledge about their influence on testosterone release as
well as biological maturation that could ultimately affect the physical performance
of young soccer players.
The present study demonstrated a lack of seasonal differences for testosterone
and that variations in training content (volume, intensity, and frequency) did not
affect sT concentration. Despite the inclusion of a period of high-volume training
(12-week preparatory training phase; preseason training), a 7-week competitive and
also a 2-week detraining phase, we did not observe any significant alterations in
sT levels. The possible explanation for the lack of fluctuation in sT concentrations
across the investigated period despite changes in training content may be related to
incomplete sexual maturation in the assessed soccer players. This explanation was
proposed by Matchokc et al. (27) who also noted the absence of seasonal fluctua-
tions of testosterone in a sample of boys and girls. Moreover, the lack of fluctuation
in sT concentration in both the current study (athletic population) and the study
of Matchokc and coworkers (nonathletic population) reinforces the notion that
sT measures are relatively stable and thus are pertinent for monitoring hormonal
status in this population.
In addition, the observations of Granger et al. (12) that sT is stable over time,
are also in line with the present findings. The authors (12) stated that such stability
over time influences the probability of detecting longitudinal associations between
testosterone and outcome variables. However, we cannot rule out that interpartici-
pant variation could partly explain the stability of the testosterone concentration.
Large variability could reduce the likelihood to observe a statistical significance.
Therefore, this issue should be addressed in future studies to corroborate or refute
the hypothesis that the stability observed is actually due to incomplete sexual matu-
ration in the assessed population and not an artifact of intraparticipant variability.
The role of free testosterone on physical performance in young soccer players
has also been examined elsewhere. Despite lower amount of data in young athletes
compared with adult populations regarding the role of free testosterone concentra-
tion in influencing athletic performance (4,9,30), a related study has demonstrated
similar associations in young soccer players to those identified here. In a group of
66 Spanish soccer players aged between 10 and 14 years old, Gravina et al. (13),
reported that salivary testosterone concentrations at the beginning of the season
194  Moreira et al.

were positively correlated with improvements in tests of countermovement jump and

drop jump performance as well as an increase in relative VO2max during the season.
A significant relationship (r = .48) between sT concentration at the beginning of
the study and changes in CMJ was presented. The authors also demonstrated that
the relationship between initial sT values and increments in relative VO2max (r =
.32) was lower than verified to CMJ.
These results are partially in agreement with present findings. Following
division of the groups according to low or high testosterone concentrations, we
identified significant differences between groups regarding CMJ but not Yo-Yo
performance. Moreover, the contribution of testosterone to explained variance in
Yo-Yo for the whole group was lower than verified in CMJ. One could argue that
the higher contribution from testosterone levels to CMJ performance can be related
to it playing a more important role in strength/power-related performance compared
with intermittent endurance type activities.
In contrast to the present findings and those observed by Gravina et al. (13),
work by Baldari et al. (2) demonstrated no significant correlation between salivary
testosterone concentrations (r = .12) and standing long jump performance in fifty-
one players (10–14 years old) from two Italian soccer clubs. However, standing
long jump length was positively and significantly correlated with salivary dehy-
droepiandrosterone (DHEA) concentrations (r = .38). Baldari et al. (2) speculated
that the contrasting relationship between salivary DHEA and salivary testosterone
concentration in relation to standing long jump performance could be explained, at
least in part, by different concentrations in circulation during growth. The authors
also suggested that their players could have had high DHEAS concentrations
whereas high testosterone concentrations had not yet been reached.
In summary, the present observations show that hormonal status, notably
testosterone levels, has a key role in determining the functional performance in
young soccer players. Alterations in testosterone levels (increments) associated
with pubertal development appear to be a reasonable predictor of variations in
performance notably in power-strength related activities. These results could sug-
gest that preadolescent players, in whom hormonal (testosterone) concentrations
are elevated, could be favored in identification and selection schemes. Thus there is
potential for sports science staff members to introduce systematic saliva sampling
in training monitoring throughout the long-term talent selection and development
process. The information could aid coaches in interpreting physical performance in
elite preadolescent and adolescent soccer players. It is also important to highlight
that in contrast to blood sampling, saliva sampling offers many benefits for assessing
androgen hormones in young athletes. In contrast to the former, it is less-invasive
and stress-free and thus repeated measurements can be performed (19).
The absence of seasonal fluctuations (6 months) of testosterone in preadolescent
soccer players may indicate that monitoring sT could help researchers, support staff
and coaching practitioners to understand and interpret appropriately the findings
from studies in young athletes’ as well as changes in performance throughout the
training process. As testosterone is very stable in preadolescent players, any sig-
nificant changes in its concentration across the competitive season could predict
variations in physical performance. Finally, it is important to highlight that the
lack of changes in testosterone concentration across the four time points, might
have contributed to the absence of any significant improvements in both CMJ and
 Testosterone and Performance in Soccer   195

Yo-Yo performance. These findings are partially in agreement with those in a recent
study by Williams et al. (39). Despite observed changes in countermovement jump
performance above the smallest worthwhile effect of 1.8% over a 6-month-interval,
the authors did not demonstrate significant statistical differences across 6 months-
intervals in performance in 12–13 year-old soccer players. They also demonstrated
that there were no consistent differences between the rates of change observed
during and outside the playing season in jump performance, and attributed these
results to the prolonged development of motor coordination during childhood.

Practical Applications
In the current study, findings indicate a key role for hormonal status in interpreting
physical performance in elite preadolescent soccer players. Boys who are advanced
in terms of hormonal status might be at an advantage physically compared with
peers. Sport science staff members and coaches could include salivary testosterone
monitoring to track the hormonal status in young soccer players across the long-
term training process. Greater knowledge about the relationship among hormonal
status and physical performance could help to understand and interpret appropriately
potential changes in performance throughout the training process. This approach
based on the use in conjunction of testosterone concentration, YPHV, and anthropo-
metric measures to interpret performance results may be useful in practical settings
and could help to minimize the likelihood that talented young soccer players who
are behind in biological maturity and less developed physically are not overlooked.

Acknowledgments
We would like to thank the FAPESP (Fundação de Amparo à Pesquisa do Estado de São
Paulo; São Paulo Research Foundation, process, 2010/12489-6) for funding this research.
We also wish to acknowledge all soccer players, their parents and team staff, in particular
Wanderlei Brilhante Junior, Daniel Leite Portella and Sandro Sargentim, involved in this
study for their committed participation.

References
1. Arregger, A.L., L.N. Contreras, O.R. Tumilasci, D.R. Aquilano, and E.M. Cardoso.
Salivary testosterone: a reliable approach to the diagnosis of male hypogonadism. Clin.
Endocrinol. (Oxf.). 67:656–662, 2007. PubMed doi:10.1111/j.1365-2265.2007.02937.x
2. Baldari, C., L. Di Luigi, G.P. Emerenziani, M.C. Gallotta, P. Sgro`, and L. Guidetti. Is
explosive performance influenced by androgen concentrations in young male soccer
players? Br. J. Sports Med. 43:191–194, 2009. PubMed doi:10.1136/bjsm.2007.040386
3. Bangsbo, J. Yo-Yo Tests. Copenhagen: August Krogh Institute, 1996.
4. Boisseau, N., and P. Delamarche. Metabolic and hormonal responses to exercise in chil-
dren and adolescents. Sports Med. 30:405–422, 2000. PubMed doi:10.2165/00007256-
200030060-00003
5. Bunc, V., and R. Psotta. Physiological profile of very young soccer players. J Sports
Med Phys Fitness. 41:337–341, 2001. PubMed
6. Campbell, B., M.T. O’Rourke, and S.F. Lipson. Salivary testosterone and body com-
position in Arial males. Am. J. Hum. Biol. 15:697–708, 2003. PubMed doi:10.1002/
ajhb.10203
196  Moreira et al.

7. Crewther, B.T., J. Keogh, J. Cronin, and C. Cook. Possible stimuli for strength and
power adaptation: acute hormonal responses. Sports Med. 36:215–238, 2006. PubMed
doi:10.2165/00007256-200636030-00004
8. Crewther, B.T., C. Cook, M. Cardinale, R.P. Weatherby, and T. Lowe. Two emerging
concepts for elite athletes: the short-term effects of testosterone and cortisol on the neu-
romuscular system and the dose-response training role of these endogenous hormones.
Sports Med. 41:103–123, 2011. PubMed doi:10.2165/11539170-000000000-00000
9. Di Luigi, L., C. Baldari, M.C. Gallotta, et al. Salivary Steroids at Rest and After a Training
Load in Young Male Athletes: Relationship with Chronological Age and Pubertal Devel-
opment. Int. J. Sports Med. 27:709–717, 2006. PubMed doi:10.1055/s-2005-872931
10. Eisenmann, J.C., and R.M. Malina. Age- and sex-associated variation in neuromuscular
capacities of adolescent distance runners. J Sports Sci. 21:551–557, 2003. PubMed
doi:10.1080/0264041031000101845
11. Figueiredo, A.J., C.E. Gonçalves, M.J. Coelho e Silva, and R.M. Malina. Characteristics
of youth soccer players who drop out, persist or move up. J Sports Sci. 27:883–891,
2009. PubMed doi:10.1080/02640410902946469
12. Granger, D.A., E.A. Shirtcliff, A. Booth, K.T. Kivlighan, and E.B. Schwartz. The
“trouble” with salivary testosterone. Psychoneuroendocrinology. 29:1229–1240, 2004.
PubMed doi:10.1016/j.psyneuen.2004.02.005
13. Gravina, L., S.M. Gil, F. Ruiz, J. Zubero, J. Gil, and J. Irazusta. Anthropometric and
physiological differences between first team and reserve soccer players aged 10-14
years at the beginning and end of the season. J. Strength Cond. Res. 22:1308–1314,
2008. PubMed doi:10.1519/JSC.0b013e31816a5c8e
14. Styne, D.M., and M.M. Grumbach. Puberty: ontogeny, neuroendocrinology, physiology
and disorders. In: Kronenberg, H.M., S. Melmed, K.S. Polonsky, P. Reed Larsen, edi-
tors. Williams textbook of endocrinology. Philadelphia: Saunders, 2008, pp. 969-1166.
15. Helsen, W.F., J.L. Starkes, and J. Van Winckel. Effect of a change in selection year
on success in male soccer players. Am. J. Hum. Biol. 12:729–735, 2000. PubMed
doi:10.1002/1520-6300(200011/12)12:6<729::AID-AJHB2>3.0.CO;2-7
16. Hoff, J. Training and testing physical capacities for elite soccer players. J Sports Sci.
23:573–582, 2005. PubMed doi:10.1080/02640410400021252
17. Kiess, W., A. Meidert, R.A. Dressendorfer, et al. Salivary cortisol levels throughout
childhood and adolescence: relation with age, pubertal stage, and weight. Pediatr. Res.
37:502–506, 1995. PubMed doi:10.1203/00006450-199504000-00020
18. Leard, J.S., M.A. Cirillo, E. Katsnelson, et al. Validity of two alternative systems for
measuring vertical jump height. J. Strength Cond. Res. 21:1296–1299, 2007. PubMed
19. Lewis, J.G. Steroid analysis in saliva: An overview. Clin. Biochem. Rev. 27:139–146,
2006. PubMed
20. Lloyd, R.S., J.L. Oliver, M.G. Hughes, and C.A. Williams. The influence of chronologi-
cal age on periods of accelerated adaptation of stretch-shortening cycle performance
in pre and postpubescent boys. J. Strength Cond. Res. 25:1889–1897, 2011. PubMed
doi:10.1519/JSC.0b013e3181e7faa8
21. Malina, R.M. Growth and maturity status of young soccer (football) players. In: Reilly,
T and A. M. Williams, editors. Science and Soccer. London: Routledge, 2003, pp. 287-
306.
22. Malina, R.M., A.L. Claessens, K. Van Aken, et al. Maturity Offset in Gymnasts: Appli-
cation of a Prediction Equation. Med. Sci. Sports Exerc. 38:1342–1347, 2006. PubMed
doi:10.1249/01.mss.0000227321.61964.09
23. Malina, R.M., M.J. Coelho e Silva, A.J. Figueiredo, C. Carling, and G. Beunen. Inter-
relationships among invasive and non-invasive indicators of biological maturation in
adolescent male soccer players. J. Sports Sci. ahead of print, 2012.
24. Malina, R.M., S.P. Cumming, A.P. Kontos, J.C. Eisenmann, B. Ribeiro, and J. Aroso.
Maturity-associated variation in sport-specific skills of youth soccer players aged 13-15
years. J. Sport Sci. 23:515–522, 2005. PubMed doi:10.1080/02640410410001729928
 Testosterone and Performance in Soccer   197

25. Malina, R.M., J.C. Eisenmann, S.P. Cumming, B. Ribeiro, and J. Aroso. Maturity-asso-
ciated variation in the growth and functional capacities of elite youth football (soccer)
players 13-15 years. Eur. J. Appl. Physiol. 91:555–562, 2004. PubMed doi:10.1007/
s00421-003-0995-z
26. Marshall, W.A., and J.M. Tanner. Variations in the pattern of pubertal changes in boys.
Arch. Dis. Child. 45:13–23, 1970. PubMed doi:10.1136/adc.45.239.13
27. Matchock, R.L., L.D. Dorn, and E.J. Susman. Diurnal and seasonal cortisol, testosterone,
and DHEA rhythms in boys and girls during puberty. Chronobiol. Int. 24:969–990,
2007. PubMed doi:10.1080/07420520701649471
28. Mirwald, R.L., A.D.G. Baxter-Jones, D.A. Bailey, and G.P. Beunen. An assessment
of maturity from anthropometric measurements. Med. Sci. Sports Exerc. 34:689–694,
2002. PubMed doi:10.1097/00005768-200204000-00020
29. Monsma, E.V., K.A. Pfeiffer, R. Harvey, R. Ross, S. Brown, and R.M. Malina. Maturity-
offset, age at menarche, and social physique anxiety among female participants in
aesthetic activities. J Sport Exer Psychol. 27:S109, 2005.
30. Naughton, G., N.J. Farpour-Lambert, J. Carlson, M. Bradney, and E. Van Praagh.
Physiological issues surrounding the performance of adolescent athletes. Sports Med.
30:309–325, 2000. PubMed doi:10.2165/00007256-200030050-00001
31. Nurmi-Lawton, J.A., A.D. Baxter-Jones, R.L. Mirwald, et al. Evidence of sustained
skeletal benefits from impact-loading exercise in young females: a 3-year longitudinal
study. J. Bone Miner. Res. 19:314–322, 2004. PubMed doi:10.1359/JBMR.0301222
32. Philippaerts, R.M., R. Vaeyens, M. Janssens, et al. The relationship between peak height
velocity and physical performance in youth soccer players. J Sports Sci. 24:221–230,
2006. PubMed doi:10.1080/02640410500189371
33. Rilling, J.K., C.M. Worthman, B.C. Campbell, J.F. Stallings, and M. Mbizva. Ratios of
plasma and salivary testosterone throughout puberty: production versus bioavailability.
Steroids. 61:374–378, 1996. PubMed doi:10.1016/0039-128X(96)00043-8
34. Simmons, S.E., J.C. White, and J.M. Stager. Maturity assessment in competitive swim-
mers. Med. Sci. Sports Exerc. 36:S103, 2004.
35. Slaughter, M.H., T.G. Lohman, R.A. Boileau, et al. Skinfold equations for estimation
of body fatness in children and youth. Hum. Biol. 60:709–723, 1988. PubMed
36. Stocker, S., H.R. Gittinger, and G. Herth. Exogenous testosterone differentially affects
myelination and neurone soma sizes in the brain of canaries. Neuroreport. 5:1449–1452,
1994. PubMed doi:10.1097/00001756-199407000-00010
37. Vänttinen, T., M. Blomqvist, K. Nyman, and K. Häkkinen. Changes in body composi-
tion, hormonal status and physical fitness in 11, 13 and 15 year old Finnish regional
youth soccer players during a two year follow-up. J. Strength Cond. Res. 25:3342–3351,
2011. PubMed doi:10.1519/JSC.0b013e318236d0c2
38. Viru, A., and M. Viru. Preconditioning of the performance in power events by endog-
enous testosterone: in memory of Professor Carmelo Bosco. J. Strength Cond. Res.
19:6–8, 2005. PubMed
39. Williams, C.A., J.L. Olivwer, and J. Falkner. Seasonal monitoring of sprint and jump
performance in a soccer youth academy. Int J Sports Physiol Perform. 6:264–275, 2011.
PubMed