Aerobic fitness profile of youth soccer players : effects of chronological age and playing position Perfil aeróbio de jovens jogadores de futebol : efeitos da idade cronológica e da posição tática

To establish a profile of the aerobic fitness in young soccer players, it is critical to consider different intervenient factors such as maturity status, chronological age and playing position. The aim of this study was to identify the biological maturation, chronological age, and playing position effects on physical and physiological characteristics of young soccer players. Two hundred and one soccer players of 11-19 years old were divided into groups relative to their maturity status, chronological age and playing position. A maximal exercise test was performed to determine peak oxygen uptake (VO2peak) and ventilatory thresholds (VT1 and VT2) parameters in a treadmill. Biological maturation showed no significant effect on relative values (mL.kg-1.min-1) of VO2peak, VT2 and VT1 (0.004 < h2 < 0.039), but showed large positive effect on maximal aerobic speed (MAS) and speed at VT2 (VT2speed). Chronological age showed a medium positive effect on relative values of VO2peak, VT2 and VT1 (0.095 < h2 < 0.137) and a large positive effect on MAS and VT2speed. Relative values of VO2peak and VT1 showed no significant differences among groups for playing position (P>0.05; 0.044 < h2 < 0.051). However, goalkeepers showed significant lower relative values for VT2 and VT2speed than other playing positions and a medium positive effect was observed (P<0.05; 0.077 < h2 < 0.119). Chronological age showed a medium to large positive effect on aerobic fitness parameters, while biological maturation showed a positive effect only on MAS and VT2speed. Playing position showed a medium positive effect on VT2 and VT2speed.


INTRODUCTION
Soccer is one of the most popular sports in the world and it is characterized by a combination of physical, technical and tactical factors 1,2 .Soccer is an intermittent team sport involving low to high-intensity efforts 3 and showing running as the main physical activity of the game 1,3 .During a match, the total running distance covered by a young soccer player ranges from 4 to 8.5 km 1,3,4 while by a professional soccer player ranges from 8 to 12km 1 .In terms of the physical performance during a soccer match, the aerobic fitness level has been considered as most powerful determinant of match running performance 3 .In addition, the distance run at high-intensity has been used to rank physical performance between players 4,5 .
In this context, the aerobic metabolism is important to soccer performance mainly because it is the predominant source of energy to supply physical demands 1 .For this reason, to determine aerobic fitness parameters such as peak oxygen uptake (VO 2peak ), ventilatory thresholds (VT 1 and VT 2 ), maximal aerobic speed (MAS), speed at VT 2 (VT 2speed ) and VT 1 (VT 1speed ) are important to understand soccer performance and to identify intensity training zones and cardiorespiratory adaptations related to soccer training [1][2][3][4][5][6] .VO 2peak , MAS and VT 2speed have showed a positive correlation with the total distance covered and high-intensity efforts performed by young soccer players during a match 2,7 .For example, physically fit players perform more high-intensity running during a match compared to theirs less-fit counterparts 7 .Improvements of 10% in VO 2peak have been associated with a 20% increase in the total distance covered and twice the number of sprints during the game 2 .Additionally, VT 2 have been considered an important variable to soccer performance mainly because it is very close to the intensity at which the soccer matches are played 1,2 .VT 2speed and VT 1speed have been used to delimit the high-intensity and low to moderate-intensity zones during match analysis, respectively 5 .
The interpretation of aerobic fitness features may be problematic from a talent identification/development perspective due to the biological maturational, chronological age, relative age and playing position effects.Some studies indicate that biological maturation can exert a positive effect on VO 2peak [8][9][10] and ventilatory thresholds 11 .Consequently, biological maturation may be a confounding factor during the talent development and/or talent identification process.Athletes more advanced in biological maturation process were described with higher VO 2peak , power, speed, strength, sport-specific skills, and aerobic performance 8,9 .Consequently, late maturing boys are systematically excluded while early maturing boys are generally selected 9,12,13 .
Youth soccer competitions are generally organized into annual agegroups using specific cut-off values.This organization model was designed to give players equal participation and training opportunities, including fair opportunities for talent identification and/or development process 13,14 .However, soccer players from the same selection year can be older com-pared to their teammates.This can result in maturational and physical performance differences in favor of players born early in the birth year 13,15,16 .This phenomenon has been defined as relative age and its consequences are known as the relative age effect (RAE) [13][14][15] .
Players who are born early in the selection year (e.g.first birth quarter) take advantage of this subtle chronological lead and are more likely to be selected compared to their peers born later in the selection year (e.g.fourth birth quarter) 13 .In this scenario, chronological age plays similar role as biological maturation favoring those athletes born in the first quarter and excluding those born in the fourth birth quarter.Therefore, these factors should be considered by soccer clubs to not bias the athlete's selection process towards the early maturing players due to the influences of RAE 17,18 .
In addition to biological maturation and chronological age, to establish an aerobic fitness profile in young soccer players, it is critical to consider other intervenient factors, such as playing position 3 .Playing position has been described as an important feature related to soccer performance in young soccer player due to its effects on aerobic fitness 3,[19][20][21] .Currently literature lacks in studies about the effects of biological maturation, RAE and playing position on VO 2peak and ventilatory thresholds parameters in high-level youth soccer players.So far, only few studies have investigated the relationship among aerobic fitness, biological maturation, RAE, and playing position in high-level young soccer players simultaneously.Therefore, the present study aimed to describe the physical and physiological profile of high-level young soccer players and to identify whether chronological age, maturity status, and playing position exert effects on VO 2peak and ventilatory thresholds parameters.

Participants
This cross-sectional sample includes 201 high-level youth male soccer players aged 11-19 years old who were recruited from a professional club academy competing at the Brazilian national level.The participants were categorized into groups according to their maturity status, chronological age and playing position.All participants were engaged in formal training of 5 to 8 training sessions per week (60-120 minutes) and completed approximately one local game per week during the last competitive season.Inclusion criteria included at least 1 year of experience in formal soccer training.Participants were not taking any medications at the time of participation, reported no medical conditions including no history of surgery or musculoskeletal injury, and were not diagnosed with any chronic diseases.The study approval was obtained from the University Research Ethics Board (Project ID: 2008082) and was conducted in compliance with the standards set by the Declaration of Helsinki.Participants and their legal guardians were informed of the experimental protocol and potential risks and provided written informed assent and consent prior to participation.

Study Overview
Participants visited the laboratory on a single session to complete several physical and physiological examinations, which included anthropometrics (height and body mass), determining maturity status, and completing an incremental exercise test to assess VO 2peak and ventilatory thresholds parameters.They also completed a questionnaire about their health status, use of medication, birth date and years of training.Similar to previous studies 20,22,23 , all participants were separated into groups according to their self-reported best position in the field: goalkeeper (GK); full-back (FB); central-back (CB); defensive midfielder (DM); offensive midfielder (OM); and attacker (AT).Participants were familiar with the procedures before data collection commenced.All measurements were performed during the preseason period.

Maturity Status
Biological maturation was assessed by a trained healthcare professional using the criteria described by Tanner 24 .Prepubertal status was defined as stage I (n = 26), pubertal as stage II (n = 13), III (n = 39), and IV (n = 38), and as postpubertal as stage V (n = 85).

Aerobic Fitness
VO 2peak and ventilatory thresholds parameters were assessed by open circuit spirometry (breath by breath) during the performance of a maximum treadmill running test.Following the manufacturer's instructions, the ergospirometry device (MedGraphics Cardiorespiratory Diagnostic Systems -CPX-D, St Paul, Minnesota) was manually calibrated using known gas concentrations.The participants were given time to adapt to the treadmill (Quinton Instruments-Seattle, USA) and ergospirometry device.Participants then performed a single progressive maximum effort test, which consisted of running for three minutes at 8 km .hr -1 followed by increases of 0.5 km .hr -1 every 30 seconds until exhaustion 21 .Participants were verbally encouraged during the test to achieve their maximal performance.To verify an exhaustive effort, each participant had to satisfy at least two of the following criteria upon termination of the treadmill test due to volitional exhaustion: 1) heart rate at the end of the test ≥ 95% of the predicted maximal heart rate, 2) a plateau of VO 2 that was defined as an increase in VO 2 of less than 2.1 mL .kg -1.min -1 with a corresponding increase in exercise intensity; 3) respiratory exchange ratio (RER) ≥ 1.0 21 .Heart rate was measured using a heart rate monitor (Polar S610, USA).According to these criteria, all participants showed a valid VO 2peak .
VO 2peak was defined as the highest value recorded during the last minute of exercise test.When a plateau in VO 2 could be detected, the higher value during plateau period was used to represent VO 2peak .Ventilatory thresholds were determined in plots of the ventilation (VE), ventilatory equivalents (VE/VO 2 and VE/VCO 2 ), partial pressure of end-tidal carbon dioxide (PETCO 2 ), and partial pressure of end-tidal oxygen (PETO 2 ) as a func-tion of oxygen uptake.Criteria to determine ventilatory thresholds were as follows: VT 1 was determined when there was a rapid increase in the VĖ/VȮ 2 and PETO 2 with no concomitant increase in VĖ/VĊO 2 and PETCO 2 .VT 2 was determined as the point at which a rapid rise in VĖ/ VĊO 2 and a fall in PETCO 2 were observed.VT 1 and VT 2 were defined as the work rates associated with a first and a second nonlinear increase of VE and VCO 2 6,21 .Three independent reviewers blindly determined ventilatory thresholds following the criteria previously described.

Statistical analysis
The normality of the distributions was investigated using the Kolmogorov-Smirnov test, and the homogeneity of the variables was assessed using Levene's test.The mean and standard deviation values were used in the descriptive analysis.Differences among maturity status, chronological age, and playing position groups for physical and physiological traits were tested with analysis of variance (One-way ANOVA).Bonferroni adjustments were done for post-hoc comparisons.For each ANOVA, partial eta-squared (h 2 ) was calculated as a measure of effect size (ES).To interpret the magnitude of the effect size the following were adopted < 0.01 as small, 0.06 to 0.15 as moderate and > 0.15 as large effect 25 .Differences between the observed and expected birth-date distribution were tested with the Chi-square statistic.Expected birth-date distributions were estimated to be 25% for each quarter of the year.The level of statistical significance was set at P ≤ 0.05.Statistical analyses were conducted using the statistical programs SPSS (version 19.0, SPSS, Inc., IBM Company; NY, USA).

RESULTS
Physical and physiological characteristics considering the entire group of high-level youth soccer players are presented in Table 1.
Table 2 depicts the physical and physiological characteristics in accordance to participant's biological maturation.As expected, postpubescent soccer players were significantly heavier and taller than prepubescent and pubescent soccer players (P < 0.05).Biological maturation showed no significant effects on relative to body size values (mL .kg -1.min -1 ) of VO 2peak , VT 2 and VT 1 (0.004 < h 2 < 0.039); and no significant differences among maturational groups were observed (P > 0.05).
Table 3 shows the physical and physiological characteristics in accordance to maturity status (Tanner stages).No significant differences were observed among maturational groups for relative to body mass values (mL .kg -1.min -1 ) of VO 2peak , VT 2 , and VT 1 (P > 0.05).Exception was observed on the comparison between stages III and V for VT 1 values (P < 0.05).Tanner stages showed no significant effects relative to body mass values of VO 2peak and VT 2 (0.047 < h 2 < 0.058).However, a medium positive effect was found for relative to body mass values of VT 1 (h 2 = 0.067).
The groups comparisons based on their chronological age are presented in Table 4. Significant differences were found for relative to body mass values of VO 2peak , VT 2 and VT 1 manly in favor of older groups (13-14, 15-16, and > 17 years old) in comparison to the younger group (11-12 years old).Chronological age showed medium positive effect on relative to body mass values of VO 2peak , VT 2 and VT 1 (0.095 < h 2 < 0.137).
Table 5 shows the physical and physiological characteristics according to playing position.Relative values (mL .kg -1.min -1 ) of VO 2peak and VT 1 showed no significant differences among groups for playing position (P > 0.05; 0.044 < h 2 < 0.051).However compared to the GK group, the FB, DM and OM groups showed significant differences for relative values of VT 2 , with a medium positive effect (P < 0.05; h 2 = 0.077).
Interestingly, large positive effects of biological maturation status and chronological age on MAS and VT 2speed were observed, while playing position showed a medium positive effect on VT 2speed (Tables 2, 3, 4 and 5).
RAE was identified due to overrepresentation of players born during the first semester of the year (P < 0.05).Birth-date distribution by month quartile was 49.8%, 26.4,15.9% and 8% for BQ 1 , BQ 2 , BQ 3, and BQ 4 , respectively.

DISCUSSION
This study presented a profile of physical and physiological features for highlevel young soccer players and identified the effects of biological maturation, chronological age, RAE and playing position on aerobic fitness parameters (VO 2peak , VT 2, VT 1, MAS, VT 2speed and VT 1speed ).These variables have been described as important intervenient factors for the talent development and/ or talent identification process as well for soccer performance.
To our knowledge, this was the first study to report the effects of biological maturation, chronological age, RAE and playing position on aerobic fitness profile directly assessed by ergospirometry with a large sample and age range of high-level young soccer player's simultaneously.The main findings from the present study were as follows: 1) biological maturation status showed no significant effects on VO 2peak , VT 2 , and VT 1 relative to body size values (mL .kg -1.min -1 ) in young soccer players; 2) chronological age showed a medium positive effect on VO 2peak , VT 2 , and VT 1 (mL .kg -1.min -1 ) relative values; 3) RAE was confirmed since there was an overrepresentation (76.2%) of players born in the first semester of the year; 4) playing position showed no significant differences among groups for VO 2peak and VT 1 (mL .kg -1.min -1 ) relative values, but showed medium positive effect on VT 2 (mL .kg -1.min -1 ); 5) biological maturation and chronological age showed a large positive effect on MAS and VT 2speed, while playing position showed a medium positive effect only VT 2speed.
Programs developed for detection or development of players with potential talent can be biased by the intervenient effects of biological maturation, relative age, and playing position.As a result, potentially talented players can systematically drop out the sport at an early age, and not reach higher levels of competition 12,26 .In the present study, biological maturation status did not show a significant positive effect on aerobic fitness parameters (VO 2peak , VT 2 , VT 1 express in mL .kg -1.min -1 ) in young soccer players.Some studies also did not find positive effects of biological maturation on aerobic fitness parameters of young soccer players 6,[16][17][18]21,27 . Howevr, other studies demonstrated positive effects of biological maturation status on aerobic fitness parameters of young soccer player [9][10][11] .The contrasting findings regarding effects of biological maturation on aerobic fitness parameters may be attributed to differences in the sample size experimental design (i.e.different methods used to assess aerobic fitness), scaling variable to normalize aerobic fitness (i.e.absolute values, relative to body mass, relative to fat-free mass), and methods for estimating biological maturation (i.e. Tannr stages, peak high velocity) 6,10,18,21 .The effect of biological maturation status on aerobic fitness parameters seems inconclusive for young soccer players.It is important to highlight that body mass (kg) is a crucial variable that has a large impact on athletic performance.The normalization of aerobic fitness parameters for body mass has been broadly used over time due to the high correlation with VO 2peak .For this reason, absolute VO 2peak , VT 2 , and VT 1 (mL.min-1)values are not considered appropriated to compare young soccer players with diverse biological maturation status, chronological age and/or body size 6,11,21 .
Chronological age showed a medium positive effect on aerobic fitness parameters.Positive effects were observed for VO 2peak, VT 2 , and VT 1 (mL .kg -1.min -1 ) manly due significant differences among the 13─14, 15─16 and > 17 years-old groups compared to the 11─12 years-old group.This result corroborate with previous finding 26,27 .These studies showed that aerobic fitness parameters increase progressively from 8 to 18 years of age.However, Cunha et al. 6 using multiple linear regression analysis showed that chronological age did not show positive effect on VO 2peak , VT 2 and VT 1 (mL .kg -1.min -1 ) in young soccer player.
In the present study, an overrepresentation of players born during the first semester of the year was observed, confirming the phenomenon known as RAE.Birth-date distribution by month quartile was 49.8%, 26.4%, 15.9% and 8% for BQ 1 , BQ 2 , BQ 3, and BQ 4 , respectively.Similar findings were shown in early studies 13,16,28 .It has been well accepted that players born in the first semester of the year are usually taller, heavier and have greater aerobic power, muscular strength, speed, perception of competence, and self-efficacy compared with their relatively younger peers (born in the last semester of the year) 14,16,23,28 .RAE has been associated with success in youth soccer players.There were positive correlations among RAE and rank, points scored, and goals against in U-17 Germany soccer teams 29 .In addition, Carling et al. 23 showed that out of a total of 75, 49, 26, and 10 players born in the BQ 1 , BQ 2 , BQ 3 , and BQ 4 periods, 45.6%, 38.8%, 43.3% and 70% achieved professional status, respectively.Interestingly, a meta-analytical review identified consistent prevalence of RAE in some sports (ice-rocky, volleyball, basketball, football, and soccer), but with small effect size 14 .
Playing position can also to exert effect on aerobic fitness parameters in young soccer players 3,4,19,20 .A large association has been observed between aerobic fitness and total distance covered during a match in youth soccer players 3,7 .More specifically, relationship between match running performance and aerobic fitness parameters are more position-dependent 3,4 .In general, midfielders and attackers show higher values of aerobic fitness and covered more distance in highest intensity zone compared to defensive players 3,4 .In the present study, playing position showed no significant effects on VO 2peak (mL .kg -1.min -1 ) among GK, FB, CB, DM, OM and AT groups.In additional, other studies showed that aerobic endurance (m) was different among playing positions.Indeed goalkeepers showed lower values compared to defenders, midfielders and attackers 19,20 .However, Coelho & Silva et al. 22 also did not find effects of playing position on aerobic endurance (m) in young Portuguese soccer players from local and regional levels.The discrepancies among these results might be partly explained by the use of different methods to determine aerobic fitness parameters (direct vs. indirect tests), by the complexity nature of physical demands during a game, and also by the playing position-specific running pattern (such as the link between defense and attack, technical actions, change positions, covered distance in low, moderate or high-intensity, acceleration for create space and counter-attack) 3,4,19,20,22 .
Few studies have assessed the effects of playing position on ventilatory thresholds in young soccer player.In the present study, GK showed VT 2 (mL .kg -1.min -1 ) values significantly lower compared to FB, DM and OF groups.However, VT 1 was similar among groups.It can be hypothesized that the observed results can be partially attributed to differences in physical and physiological demands of game among different playing positions (e.g. total distance covered during the match in low, moderate and high-intensity), as well as the position-specific training for goalkeepers, defenders, midfielders and attackers.
There is a gap in the literature about the effects of biological maturation, chronological age and playing position on MAS, VT 2speed and VT 1speed .
MAS has been considered the most powerful determinant of physical game performance in young soccer players 3 .In the present study, biological maturation and chronological age showed a large positive effect while playing position no showed effect on MAS.With respect to MAS, similar results have been reported 3,4,30 .In general, older and more mature players (15 to 18 years old) are faster according to the estimated MAS values (15 to 18 km .hr -1 ) in comparison to young and less mature players (12 to 15 years old) counterparts (13 to 15 km .hr -1 ) 3,4,30 .In addition, playing position showed no significant effect on MAS 3,4,30 .It is noteworthy that our results directly assessed of MAS are approximately 1 km .hr -1 higher than those indirectly values previously reported in function of biological maturation and chronological age.
Based in the common belief that older players are more mature and exhibit better performance at the selection time, coaches tend to prefer those players that achieve success in highly competitive youth sports 29 .However, from the long-term development talent perspective (i.e.development program from youth athlete to elite professional athlete) this selection model might be a concern.A recent longitudinal study 12 showed that late maturing soccer players are more likely to achieve success in the top level game (60.1%), compared with their normal maturing (38.1%) and early maturing counterparts (11.8%).Additionally, athletes born in the BQ 4 have reached the professional level in a greater proportion than athletes born in BQ 1 , BQ 2 and BQ 3 23 .From a practical point of view, coaches, sport scientists and sport administrators should take into account the biological maturation status, RAE and playing position effects on aerobic fitness profile during talent development and/or talent identification programs, in order to avoid dropout of the potential successful young players.Since MAS and VT 2speed have been associated with match running performance in soccer, these variables could too be used to help talent identification/development process and to design specific training programs focused on position-dependent highintensity traits.In addition, caution must be taken when interpreting the effects of biological maturation, chronological age and playing position on aerobic fitness parameters in young soccer players on talent identification/ development perspective.We cannot to assume that important factors of success in professional players automatically can be extrapolated to young soccer players.

CONCLUSION
In the present study, biological maturation showed no positive effects on aerobic fitness parameters except for MAS and VT 2speed .On the other hand, chronological age showed positive effects on aerobic fitness parameters and an overrepresentation of players born during the first semester of year was found.Finally, playing position showed a positive effect on VT 2 and VT 2speed .

Table 1 .
Physical and physiological characteristics of youth elite soccer players (n=201) Data expressed as mean and standard deviation (mean ± SD).MAS = maximal aerobic speed; VT 2speed = speed at VT 2 ; VT 1speed = speed at VT 1 ; HR= heart rate.

Table 2 .
Physical and physiological profile of young elite soccer players according to biological maturation status

Table 3 .
Physical and physiological profile of young elite soccer players according to Tanner stage Data were expressed as mean and standard deviation (mean ± SD), where * = significantly different from group I; # = significantly different from group II; † = significantly different from group III; § = significantly different from group IV; MAS = maximal aerobic speed; VT 2speed = speed at VT 2 ; VT 1speed = speed at VT 1 ; HR= heart rate; h 2 = partial eta square; ES q = qualitative effect size.Significance (P<0.05).

Table 4 .
Physical and physiological profile of young elite soccer players according to chronological age

Table 5 .
Physical and physiological profile of young elite soccer players in according to playing position Data expressed as mean and standard deviation (mean ± SD), where * = significantly different from goalkeeper group; # = significantly different from full-back group; † = significantly different central-back group; § = significantly different from defensive midfielder group; GK= goalkeeper; FB = full-back; CB = central-back; DM= defensive midfielder; OM =offensive midfielder; AT= attacker; MAS = maximal aerobic speed; VT 2speed = speed at VT 2 ; VT 1speed = speed at VT 1; HR= heart rate; h 2 = partial eta square; ES q = qualitative effect size; Significance (P<0.05).