Discussion summary

Journal of Science and Medicine in Sport (2009) 12, 79—84

ORIGINAL PAPER

Heart rate and blood lactate correlates of perceived exertion during small-sided soccer games

Aaron J. Coutts a,∗, Ermanno Rampinini b, Samuele M. Marcora c, Carlo Castagna d, Franco M. Impellizzeri b

a School of Leisure, Sport and Tourism, University of Technology, Sydney, Australia b Human Performance Laboratory, Mapei Sport Research Center, Castellanza (VA), Italy c School of Sport, Health, and Exercise Sciences, University of Wales-Bangor, Bangor, United Kingdom d School of Sport and Exercise Sciences, Faculty of Medicine and Surgery, University of Rome Tor Vergata, Rome, Italy

Received 13 September 2006 ; received in revised form 10 August 2007; accepted 13 August 2007

KEYWORDS Training intensity; Heart rate; Blood lactate; Rating of perceived exertion; Soccer

Summary The rating of perceived exertion (RPE) could be a practical measure of global exercise intensity in team sports. The purpose of this study was to examine the relationship between heart rate (%HRpeak) and blood lactate ([BLa−]) measures of exercise intensity with each player’s RPE during soccer-specific aerobic exercises. Mean individual %HRpeak, [BLa−] and RPE (Borg’s CR 10-scale) were recorded from 20 amateur soccer players from 67 soccer-specific small-sided games training sessions over an entire competitive season. The small-sided games were performed in three 4 min bouts separated with 3 min recovery on various sized pitches and involved 3-, 4-, 5-, or 6-players on each side. A stepwise linear multiple regression was used to determine a predictive equation to estimate global RPE for small-sided games from [BLa−] and %HRpeak. Partial correlation coefficients were also calculated to assess the relationship between RPE, [BLa−] and %HRpeak. Stepwise multiple regression analysis revealed that 43.1% of the adjusted variance in RPE could be explained by HR alone. The addition of [BLa−] data to the prediction equation allowed for 57.8% of the adjusted variance in RPE to be predicted (Y = −9.49 − 0.152 %HRpeak + 1.82 [BLa−], p < 0.001). These results show that the combination of [BLa−] and %HRpeak

measures during small-sided games is better related to RPE than either %HRpeak or [BLa−] measures alone. These results provide further support the use of RPE as a measure of global exercise intensity in soccer. © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

∗ Corresponding author. E-mail address: [email protected] (A.J. Coutts).

1440-2440/$ — see front matter © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jsams.2007.08.005

t R c m c [ f b R s

M

T m a s p t h a a p t a

m w b t f i s e a c o w J p d i o a r d G m S r

80

Introduction

The ability to monitor exercise intensity during soccer training can be used to provide important feedback to the coach regarding the training stim- ulus applied to the players. It is now common for top level professional soccer teams to moni- tor training intensity using technical devices such as heart rate monitoring systems1 and player tracking devices.2 These systems can be used to provide use- ful information on the external (i.e. distance) and internal (i.e. heart rate) training load experienced by the players. Unfortunately, time constraints associated with using these devices such as analy- sis and interpretation of data from multiple players can limit their usefulness in the practical setting. We have recently shown that a player’s rating of perceived exertion (RPE) provides an alternatively valid and time effective method for quantifying training intensity during an entire soccer training session (consisting of small-sided games, techni- cal, speed, aerobic conditioning and plyometric training).3 However, to the authors’ knowledge, no studies have specifically examined the validity of RPE as an indicator of exercise intensity solely dur- ing soccer-specific small-sided games.

Rating of perceived exertion has been suggested to be a more appropriate measure of exercise intensity than individual physiological variables6

and is thought to be representative of the com- bination of many factors affecting the internal load of exercise such as an athlete’s psycholog- ical state,7,8 training status8,9 and the external training load.10 Indeed, RPE has been shown to be a simple and valid method for quantifying whole training session intensity for both steady-state11,12

and intermittent exercise.3,12 Moreover, RPE has been correlated with many physiological measures of exercise intensity such as oxygen consumption (V̇ O2), ventilation, respiratory rate, blood lactate concentration ([BLa−]), heart rate (HR) and elec- tromyographic activity during a variety of exercise protocols.13—15 Taken together, these factors sug- gest that RPE may be a valid marker of global training intensity in athletes who undertake high- intensity, intermittent exercise.

We have previously demonstrated that session- RPE be a good general indicator for evaluating global session training intensity in soccer players on the basis of moderate correlations (r = 0.50—0.85) between HR and RPE measures of training inten- sity during soccer-specific training.3 In our previous

research RPE measures were referred to the global perception of effort for the entire training session rather than the perception of exercise intensity during each training session. However, we suggest

2 s o w T

A.J. Coutts et al.

hat a better understanding of the validity of using PE for monitoring exercise intensity during soc- er training could be gained by comparing RPE easures taken during soccer training with other

onventional markers of exercise intensity such as BLa−]. The aim of the present study was to there- ore examine the relationship between RPE with oth HR and [BLa−] to further validate the use of PE for measuring global exercise intensity during occer-specific small-sided games.

aterials

wenty soccer players from the same team (body ass: 73.0 ± 9.0 kg, height: 178.8 ± 5.2 cm, and

ge: 25 ± 5 years) volunteered to participate in the tudy. In order to be included in the study, partici- ants were required to gain medical clearance from he team physician to ensure they were in good ealth. Informed consent was obtained after verbal nd written explanation of the experimental design nd potential risks of the study, and the partici- ants were aware that they could withdraw from he study at any time. The study was approved by n Independent Institutional Review Board.

The amateur soccer team trained for approxi- ately 120 min, two to three times each week. Data ere collected two times a week from Septem- er to June during 67 team training sessions. For his study, HR, [BLa−] and RPE data were collected rom the small-sided games component of train- ng that consisted of 3 min × 4 min soccer-specific, mall-sided games play with 3 min of active recov- ry. Each small-sided game session was conducted s part of the normal training regime for the soc- er team and all games were completed outdoors n the same grass soccer pitch. The data collection as suspended in the winter period (December and anuary) to avoid the colder weather and to exclude ossible influences of extreme environmental con- itions on the results. The small-sided games nvestigated were 3-, 4-, 5-, and 6-a-side, with- ut goalkeepers, using small goals, free touches nd with a second ball always available for prompt eplacement when it left the playing area (for more etail of the formats used in these games see).4

oals were considered valid only when all team ates were in the opponents half of the pitch.

mall-sided games were performed on various sized ectangular pitches with playing areas ranging from

2 2

40 m (12 m × 20 m) to 2208 m (46 m × 48 m). A tandard warm up procedure consisting of 20 min f low intensity running, striding and stretching as completed by all players before each session. hese small-sided soccer game formats were cho-

M

s t a a f

i i c a S F c w p i y r v d i t

p m f ( H m H P r t

s P m a d s t a g e f o

a o p a a t a t d

g q i b t a

s s w s r s s a p f t

S

D ( t K i t l I

d o a a R a c c m o v i c v a m t a c O (

onitoring soccer training intensity

en for this study as they are commonly used in raining by soccer teams to develop both physical nd technical-tactical qualities and also to provide n ecologically valid range of exercise intensities or soccer training.4,5

In order to obtain the reference each player’s ndividual peak heart rate (HRpeak) at regular ntervals during the study period, soccer players ompleted both a yo-yo endurance test (level 2) nd a yo-yo intermittent recovery test (level 1) in eptember (beginning of the competitive season), ebruary (mid-season) and May/June (end of the ompetitive season). These tests were used as they ere normal part of each players physiological and erformance testing regime and conducted accord- ng to previously described methods.16,17 Both the o-yo endurance test16 and the yo-yo intermittent ecovery test17 have been shown to elicit HRpeak alues that are very close to actual HRmax (99 ± 1%) etermined in a laboratory. All players were famil- ar with the field-testing procedures being part of heir usual fitness assessment program.

In June (prior to the play-off phase) the soccer layers also completed an incremental tread- ill (RunRace, Technogym, Gambettola, Italy) test

or the determination of maximal oxygen uptake V̇ O2 max) using previously described methods.

4

eart rate was recorded throughout the incre- ental treadmill test using a portable recordable R monitor (VantageNV, S710 and Xtrainer models, olar Electro, Kempele, Finland). The highest HR eached during the laboratory or the field tests was aken as the HRpeak.

Heart rate was recorded every 5 s during each mall-sided game training session using individual olar HR monitors (VantageNV, S710 and Xtrainer odels, Polar Electro, Kempele, Finland). Immedi-

tely after every training session, the investigators ownloaded the HR data to a portable PC using the pecific software (Polar AdvantageTM, Polar Elec- ro, Kempele, Finland) and subsequently exported nd analysed using the Excel XP software pro- ram (Microsoft Corporation, USA). The mean HR xpressed relative to each players HRpeak (%HRpeak) or the entire three 4 min small-sided game section f each training session was used for analysis.

Blood lactate samples were taken within one min fter the completion of the third 4 min interval f the small-sided game. Capillary blood sam- les (5 �L) were collected from the ear lobe nd immediately analysed using several portable

mperometric microvolume lactate analysers (Lac- atePro, Arkray, Japan). Before each test, the nalysers were calibrated following the manufac- urers recommendations. To limit the influence of iet on [BLa−], all players were asked to follow a

H m c f m

81

eneric weekly nutritional plan to ensure an ade- uate carbohydrate intake (50—60% of total energy ntake). However, a food diary was not recorded y the athletes. During the small-sided games and raining sessions all players were permitted to drink d libitum.

Rating of perceived exertion (RPE, Borg’s CR-10 cale)18 was also used as a measure of inten- ity for the small-sided game. Each player’s RPE as collected at the end of each soccer-specific

mall-game to ensure that the perceived effort was eferred to the small-game training only. In this tudy, a printed Italian translation of the CR-10 cale modified from Foster et al.11 was used to ssist the players in making their responses. All layers who participated in this study had been amiliarized with this modified scale for RPE before he commencement of this study.

tatistical analyses

ata are presented as means ± standard deviation S.D.). Prior to parametric statistical procedures, he assumption of normality was verified using the olmogorov—Smirnov test and Lillefors probabil- ties. If this assumption was violated a Box-Cox ransformation was completed with the optimal ambda being determined by MINITAB 14.1 (Minitab nc., PA, USA).

A stepwise multiple regression was used to etermine a predictive equation to estimate RPE f small-sided soccer games training from [BLa−] nd %HRpeak. Partial correlation coefficients were lso calculated to assess the relationship between PE with [BLa−] and %HRpeak. Collinearity toler- nce statistics were calculated to determine the orrelation between the predictor variables. The ollinearity tolerance statistics are used to deter- ine when a predictor is too highly correlated with

ne or more of the other predictors. If the predictor ariables are highly correlated with each other, the nfluence of one variable on the response variable ould not be separated from the other predictor ariable. Therefore any variable that had a toler- nce level of less than 0.10 was not included in the odel. Standard statistical methods were used for

he calculation of means, standard deviation (S.D.) nd Pearson’s product moment correlation coeffi- ients. Statistical significance was set at p < 0.05. ne-way repeated measures analysis of variance

ANOVA) was used to examine for differences in

Rmax and distance covered during the yo-yo inter- ittent recovery tests performed throughout the

ompetitive season. Where a significant F-value was ound, post-hoc Bonferroni’s test was applied. The ultiple regression, ANOVA and collinearity statis-

82 A.J. Coutts et al.

Table 1 Partial correlations, standardized coeffi- cients and level of significance for predictors of rating of perceived exertion.

%HRpeak [BLa−]

Partial correlations 0.519 0.508 Standardized coefficient (ˇ) 0.449 0.436 Significance of standardized p < 0.001 p < 0.001

Figure 1 Changes in (A) HRmax and (B) Yo-Yo intermit- t ( (

t b l t f

e R i e H F d o r e d a v a m m s r H c w

coefficients

All correlations significant (p < 0.05).

tics and linear regression analysis were conducted using SPSS statistical software package (SPSS Inc. Version 12, Chicago, USA).

Results

The mean HR, [BLa−] and RPE from the 851 indi- vidual training sessions were 87.9 ± 3.8 %HRpeak, 5.59 ± 1.78 mmol L−1 and 7.0 ± 1.3, respectively. The RPE measures were significantly corre- lated with [BLa−] (r = 0.63, p < 0.05) and %HRpeak (r = 0.60, p < 0.05). The stepwise multiple regres- sion analysis revealed that 43.1% of the adjusted variance in RPE could be explained by exercise intensity measured by HR alone. The addi- tion of [BLa−] data to the prediction equation allowed for 57.8% of the adjusted variance (57.9% unadjusted) in RPE to be predicted (Y = −9.49 − 0.152 %HRpeak + 1.82 [BLa−]) [Adjusted R2 = 0.58; F2,849 = 582.01, p < 0.001]. Partial corre- lations, standardized coefficients and the level of significance of predictors of RPE are shown in Table 1. The collinearity statistic for this multiple regression was acceptable with tolerance levels at 0.820.

Fig. 1A shows that HRpeak did not change dur- ing the season (p > 0.05). The mean HRpeak values obtained during the laboratory and field tests were not significantly different to each other (p > 0.05). Additionally, the total distance covered by the soc- cer players during the yo-yo intermittent recovery test significantly (p < 0.01) increased between the three test sessions (Fig. 1B). The V̇ O2max of soccer players measured during the incremental treadmill test in the laboratory was 56.3 ± 4.8 ml kg−1 min−1.

Discussion

The main finding of the present study was that the combination of %HRpeak and [BLa

−] predicts RPE following soccer small-games training better than %HRpeak or [BLa

−] measures alone. In addi-

t n s t

ent recovery test performance during the study period mean ± S.D.). aSignificantly different to September p < 0.05); bsignificantly different to February (p < 0.05).

ion, the present results also demonstrated that oth %HRpeak and [BLa

−] were moderately corre- ated to RPE. These results therefore demonstrate he validity of RPE as indicator of training intensity or intermittent aerobic soccer-specific exercises.

Correlation analysis showed that %HRpeak xplained approximately 43% of the variance in PE following soccer-specific small-games train-

ng. These results are similar to previous studies xamining the relationship between RPE and R measures during intermittent exercise.13,19

or example, in a meta-analysis Chen et al.13

emonstrated that the 95% confidence interval f validity coefficients between HR and RPE was = 0.397—0.617 during progressive intermittent xercise. Likewise, Green et al.19 also recently emonstrated a moderate correlation between HR nd RPE (r = 0.63) during 5 min × 2 min cycling inter- als with 3 min of active recovery in 12 physically ctive males. The present results provide confir- ation that RPE is not a valid substitute for HR easures during high-intensity, non-steady-state

occer-specific exercises. However, since stronger elationships have been reported between RPE and R measures during steady-state endurance exer- ise (95% confidence interval: r = 0.583—0.643),13

e suggest factors other than HR contribute to

he perception of fatigue following high-intensity, on-steady-state training. The results of this tudy also revealed that the [BLa−] taken after he small-sided soccer games were moderately

M

c a t i [ p m t

t t n i t t o t a f S c a e r t r a t f e f t t [ c

R i t d t u l s h t b s d a h

r i f t

s t b f i l d v t w b t

R

onitoring soccer training intensity

orrelated with RPE taken at the same time. In greement, Green et al.19 previously demonstrated hat RPE taken following each bout of 5 × 2 min nterval cycling was moderately correlated to BLa−] (r = 0.43). Taken together, these results rovide further support to the validity of RPE as easure of global exercise intensity during interval

raining. The major finding of this study was that 57.8% of

he variance in RPE during soccer-specific aerobic raining sessions was accounted for by the combi- ation of the %HRpeak and [BLa

−] measures. It is nteresting that the addition of [BLa−] measures to he %HRpeak data in the multiple regression equa- ion resulted in an additional 14.7% of the variance f RPE being explained. Furthermore, the addi- ion of [BLa−] to the multiple regression equation lso reduced the standard error of the estimate rom 0.98 to 0.87 units on the Borg CR 10-scale. ince the present results show 42.2% of the RPE ould not be explained by [BLa−] and %HRpeak, it ppears that other factors may contribute to a play- rs’ RPE during small-sided games training. Other esearchers have suggested psychobiological fac- ors such as metabolic acidosis, ventilatory drive, espiratory gases, catecholamines, �-endorphins nd body temperature are also related to percep- ion of effort,8 however, the relationship of these actors to RPE during high-intensity, intermittent xercise is yet to be determined. Although these actors were not measured in this study, it is likely hat these could also account for some of the addi- ional variance in RPE not explained by HR and BLa−] given that these variables are also signifi- antly changed during soccer-specific exercise.20

The results of this study validate the use of PE as a marker of training intensity during high-

ntensity intermittent exercise and further support he use of RPE for quantifying training intensity uring small-sided games in soccer. A limitation of his study is that only a single [BLa−] measure was sed as the representative measure of the blood actate response to the entire small-sided games ection. It is possible that the present results may ave been altered if multiple [BLa−] measures were aken during each bout. However, it has previously een reported that the [BLa−] measured during occer activities may not represent the lactate pro- uction immediately before sampling, but rather n accumulated/balanced response to various prior igh-intensity activities.1

In summary, the present data extend earlier esearch suggesting RPE as a good indicator of train- ng intensity during soccer training. In this study, we ound that both HR and [BLa−] independently relate o the RPE measures during soccer-specific, small-

83

ided games. We also demonstrated that most of he variation in RPE measures might be explained y the combination of %HRpeak and [BLa

−], which urther supports the validity of RPE as indicator of ntensity during intermittent exercise. Since regu- ar assessment of HR and [BLa−] can be logistically ifficult and expensive, we suggest that RPE pro- ides an alternative and valid method for coaches o monitor soccer training intensity. Nonetheless, e suggest that small-sided soccer game training is est monitored through the combination of each of hese measures.

Practical implications

• Rating of perceived exertion correlates well with traditional markers of exercise intensity during soccer-specific small-sided games train- ing.

• Player’s ratings of perceived exertion may be used within a soccer training session to mon- itor global exercise intensity and help the coach control the training stimulus.

eferences

1. Bangsbo J, Mohr M, Krustrup P. Physical and metabolic demands of training and match-play in the elite football player. J Sports Sci 2006;24(7):665—74.

2. Edgecomb SJ, Norton KI. Comparison of global positioning and computer-based tracking systems for measuring player movement distance during Australian football. J Sci Med Sport 2006;9(1/2):25—32.

3. Impellizzeri FM, Rampinini E, Coutts AJ, Sassi A, Marcora SM. The use of RPE-based training load in soccer. Med Sci Sports Exerc 2004;36(6):1042—7.

4. Rampinini E, Marcora S, Castagna C, Abt GA, Chamari K, Sassi A, et al. Factors influencing physiological responses to small-sided soccer play. J Sports Sci 2007;25(6):659—66.

5. Little T, Williams AG. Measures of exercise intensity dur- ing soccer training drills with professional soccer players. J Strength Cond Res 2007;21(2):367—71.

6. Borg GA. Perceived exertion: a note on ‘‘history’’ and meth- ods. Med Sci Sports 1973;5(2):90—3.

7. Morgan WP. Psychological factors influencing perceived exertion. Med Sci Sports Exerc 1973;5(2):97—103.

8. Robertson RJ, Noble BJ. Perception of physical exertion: methods, mediators, and applications. Exerc Sport Sci Rev 1997;25:407—52.

9. Martin DT, Andersen MB, Gates W. Using profile of mood states (POMS) to monitor high-intensity training in cyclists: group versus case studies. Sport Psy 2000;14:138—56.

10. Impellizzeri FM, Rampinini E, Marcora SM. Physiological

assessment of aerobic training in soccer. J Sports Sci 2005;23(6):583—92.

11. Foster C, Hector LL, Welsh R, Schrager M, Green MA, Sny- der AC. Effects of specific versus cross-training on running performance. Eur J Appl Physiol 1995;70(4):367—72.

84

12. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, et al. A new approach to monitoring exercise training. J Strength Cond Res 2001;15(1):109—15.

13. Chen MJ, Fan X, Moe ST. Criterion-related validity of the Borg ratings of perceived exertion scale in healthy indi- viduals: a meta-analysis. J Sports Sci 2002;20(11):873— 99.

14. Robertson RJ, Goss FL, Dube J, Rutkowski J, Dupain M, Bren- nan C, et al. Validation of the adult OMNI scale of perceived exertion for cycle ergometer exercise. Med Sci Sports Exerc 2004;36(1):102—8.

15. Lagally KM, Robertson RJ, Gallagher KI, Goss FL, Jakicic JM,

Lephart SM, et al. Perceived exertion, electromyography, and blood lactate during acute bouts of resistance exercise. Med Sci Sports Exerc 2002;34(3):552—9, discussion 60.

16. Metaxas TI, Koutlianos NA, Kouidi EJ, Deligiannis AP. Comparative study of field and laboratory tests for the eval-

Available online at www.

A.J. Coutts et al.

uation of aerobic capacity in soccer players. J Strength Cond Res 2005;19(1):79—84.

17. Krustrup P, Mohr M, Amstrup T, Rysgaard T, Johansen J, Steensberg A, et al. The Yo-Yo intermittent recovery test: physiological response, reliability, and validity. Med Sci Sports Exerc 2003;35(4):697—705.

18. Borg G, Ljunggren G, Ceci R. The increase of perceived exertion, aches and pain in the legs, heart rate and blood lactate during exercise on a bicycle ergometer. Eur J Appl Physiol 1985;54(4):343—9.

19. Green JM, McLester JR, Crews TR, Wickwire PJ, Pritchett RC, Lomax RG. RPE association with lactate and heart rate

during high-intensity interval cycling. Med Sci Sports Exerc 2006;38(1):167—72.

20. Bangsbo J. The physiology of soccer with special refer- ence to intense intermittent exercise. Acta Physiol Scand 1994;151(Suppl. 619):1—156.

sciencedirect.com