Science & Sports (2012) 27, e31—e37

Disponible en ligne sur

www.sciencedirect.com

ORIGINAL ARTICLE

Acceleration time, peak torque and time to peaktorque in elite karate athletesLe temps d’accélération, le pic de force et le temps pour le pic de forcechez les athlètes d’élite en karaté

R. Scattone-Silva ∗, G.C. Lessi, D.F.M. Lobato, F.V. Serrão

Department of Physical Therapy, Laboratory of Intervention and Assessment in Orthopaedics and Traumatology,Federal University of São Carlos (UFSCar), Rodovia Washington Luís, Km 235, 13565-905, São Carlos, São Paulo, Brazil

Received 10 March 2011; accepted 24 August 2011Available online 25 November 2011

KEYWORDSMartial arts;Muscle strength;Athletic injuries;Knee;Elbow

SummaryAim. — To assess the bilateral differences in peak torque normalized by the body mass (PT/BM),acceleration time (AcT) and time to peak torque of the knee and elbow muscles of elite karateathletes, in order to identify risk factors for injury.Methods. — Seven male elite karate competitive athletes were evaluated in an isokineticdynamometer at 60◦/s and 360◦/s.Results. — No bilateral difference was found in any of the variables on the knee flexion andextension assessment (P > 0.05). The elbow assessment revealed higher values of PT/BM inelbow flexion (P = 0.02) and a smaller AcT in elbow extension (P = 0.01) at the 60◦/s speed onthe dominant limb when compared to the non-dominant limb. At the 360◦/s speed, the non-dominant elbow presented a smaller AcT in elbow extension when compared to the dominantlimb (P = 0.05). Moreover, the data regarding the flexor/extensor ratio on both joints revealedvalues that have been related to an increased risk of joint injury in young athletes.Conclusion. — Our results indicate that the functional demands of regular competitive karatetraining are not necessarily bilateral strength differences induced-factors in male karate eliteathletes, especially considering that similar bilateral differences were found previously in

healthy non-athletes. However, the competitive regular training of this martial art could pro-duce agonist-antagonist muscle asymmetries that could predispose these athletes to injuries inthe elbow and knee joints.© 2011 Elsevier Masson SAS. All rights reserved.

∗ Corresponding author.E-mail address: rodrigo scattone@hotmail.com (R. Scattone-Silva).

0765-1597/$ – see front matter © 2011 Elsevier Masson SAS. All rights reserved.doi:10.1016/j.scispo.2011.08.005

e32 R. Scattone-Silva et al.

MOTS CLÉSArts martiaux ;Force musculaire ;Lésions sportives ;Genou ;Coude

RésuméObjectifs. — Évaluer chez les athlètes d’élite en karaté, les différences du pic de force normalisépar le poids corporel (PF/PC) des muscles extenseurs-fléchisseurs de la jambe et du coude, del’accélération angulaire de ces muscles (TAc), et de leur temps de contraction, entre les mem-bres gauches et droits. Cette étude se propose d’identifier des facteurs de risque d’accidentsliés aux réponses à l’entraînement.Méthodes. — Sept athlètes d’élite de karaté du sexe masculin ont été évalués sur undynamomètre isocinétique, en utilisant les vitesses angulaires de 60◦/s et 360◦/s.Résultats. — On n’a pas pu trouver de différences bilatérales, pour aucune des variables étudiéesen extension ou en flexion du genou (p > 0,05). L’évaluation fonctionnelle des muscles du coudepermet de montrer que les muscles fléchisseurs du membre dominant ont des valeurs de PF/PCsupérieures au côté opposé, alors que les muscles extenseurs ont un TAc plus bas du côtédominant à la vitesse de 60◦/s (p = 0,01) par rapport au membre non dominant. À la vitesse de360◦/s, le coude non dominant présente des valeurs inférieures de TAc pour les extenseurs parrapport au membre dominant (p = 0,05). Les valeurs du rapport fléchisseur/extenseur pour lesdeux articulations suggèrent un risque potentiel accru de lésions chez ces jeunes athlètes.Conclusion. — Les résultats indiquent que les exigences fonctionnelles de l’entraînement pourla compétition n’induisent pas nécessairement des différences bilatérales chez les athlètesd’élite en karaté, d’autant que des différences similaires ont été trouvées dans d’autres étudeschez des sujets non athlètes. Toutefois, l’entraînement régulier en karaté peut produire desasymétries entre les muscles agonistes et antagonistes qui pourraient prédisposer ces athlètesà des lésions des muscles mobilisateurs du genou et du coude.

. Tou

1

AhlmaEtamtcip

aotttmtHdhlsetraamot

anfa

p[nftweTtlstaoaeioc

spsehiid

© 2011 Elsevier Masson SAS

. Introduction

symmetries between dominant and non-dominant limbsave been related to an increased risk of injury in ath-etes [1,2]. Imbalances between the agonist-antagonistuscles of the knee and elbow have also been linked to

greater susceptibility to injury in these joints [2—4].lite athletes may develop significant muscular asymme-ries in response to daily high demand sports training,nd the specificity of the sport gestures may lead touscle imbalances that could predispose these athletes

o injuries [5—7]. Therefore, the assessment of the mus-ular function of elite athletes is relevant in order todentify muscle imbalances and establish injury preventionrograms.

The isokinetic dynamometer has been widely utilized tossess the muscle performance of athletes [2—4,6—15] inrder to identify specific adaptations related to sports prac-ice and important muscle imbalances that could be relatedo injury predisposition. In isokinetic testing, the peakorque variable is considered the ‘‘gold standard’’ measure-ent [16,17] and, therefore, it is an important parameter

o be assessed, especially in the athletic population.owever, it has been stated that strength measurementso not reflect muscle performance characteristics compre-ensively and muscle balance considerations should not beimited to analysis of strength parameters [18]. Parametersuch as acceleration time and time to peak torque have beenstablished in the literature as muscle recruitment variableshat provide valuable information regarding neuromusculareadiness to produce maximal contractions [17,19,20]. Thebility to produce torque quickly is an important skill in mostthletic endeavors and the assessment of muscle recruit-

ent patterns in athletes may provide better indications

f the functional performance than the evaluation of peakorque alone [17]. Through a more comprehensive isokinetic

ka[

s droits réservés.

ssessment, muscle recruitment issues may be identified andeuromuscular control interventions may be implementedor injury prevention or in the rehabilitation of injuredthletes.

Karate is currently considered one of the most widelyracticed system of Japanese martial arts in the world21]. Its practice requires high technical skill accompa-ied by a great ability to perform strikes and kicks asast as possible [15,22,23] and the practitioners are con-inuously challenged in performing very complex actionsith precision and high velocity to adequately executeffective attack and defense techniques in combat [15,24].herefore, the musculature involved in these actions needso be quickly recruited in maximal effort, which high-ights the importance of assessing isokinetic parametersuch as acceleration time, peak torque and time to peakorque in this population. Rehabilitation specialists shouldlso be aware that karate athletes spend a great dealf time training in positions that could place significantmount of stress on joints such as the hips, ankles andspecially the knees [11,22]. Thus, a better understand-ng of the muscle performance of these athletes may bef great usefulness in injury rehabilitation and preventionenters.

Muscle imbalances around the knee joint have beentudied in several populations and it has been pro-osed that decreased hamstrings relative to quadricepstrength is implicated as a potential mechanism for lowerxtremity injuries [2,3,8,12,13,25]. Deficits in relativeamstrings strength and recruitment may contribute toncreased anterior cruciate ligament (ACL) injury riskn athletes, considering that hamstrings activation canecrease the load on the passive restraints of the

nee [26], increase the knee joint compression force,nd stabilize the knee from external varus/valgus loads27].

s

tkci

(uawicmoriltdtste

pswamtco

tafitil(wtth[

m6wc3aflarpm

Acceleration time and time to peak torque in karate athlete

Injuries in the elbow are less common than injuries in thelower limb joints, but have also been related to chronic com-plaints in martial artists [28]. Valkering et al. [28] recentlyreported a condition they called the boxer’s elbow syn-drome. The authors describe this condition as an injuryof posterolateral impingement of the elbow that occursbecause of hyperextension traumas in the elbow with theforearm in pronation due to punches that have missed thetarget [28]. Muscle imbalances between the flexors andextensors of the elbow might be an underlying contribut-ing factor to the development of this injury and must beassessed for prevention purposes. Because of the similar-ities between the techniques of punching in boxing andkarate (elbow extension with forearm pronation), it is pos-sible that karate athletes may be prone to developing theboxer’s elbow syndrome, although there is no report in theliterature of such injuries in this population.

Therefore, the purpose of this study was to assess themuscle performance of the dominant and non-dominantlimbs regarding strength (peak torque) and recruitment(acceleration time and time to peak torque) of the knee andelbow muscles of elite competitive karate athletes. It washypothesized that the specific sport adaptations related tothe daily practice of this martial art would produce muscleimbalances that could predispose these athletes to injuriesin these joints.

2. Methods

2.1. Subjects

Seven male competitive elite Shotokan karate athletes(24.5 ± 4.8 years; 171.5 ± 6.5 cm of height; 76.0 ± 17.7 kgof body mass) volunteered to this study. They were out-standing athletes who had recently obtained prominence instate, national and international tournaments. Inclusion cri-teria for the study were: 18 to 30 years of age; black-beltdegree (1st Dan or greater); current participation in orga-nized competitions; and daily training frequency of at least2 hours. Exclusion criteria were: previous injury or currentinstability in the elbow and/or knee joints; regular partic-ipation in other sports activities; and presence of reportedcardiovascular disease.

The athletes were approached at renowned academies inthe state of São Paulo (Brazil), and all the athletes who vol-unteered for the study were instructed about the proceduresand signed an informed consent. The study was approved bythe ethics committee of the university.

2.2. Procedures

Subjects were submitted to a physical examination by asingle examiner for verification of the presence of instabil-ities in the elbow and/or knee joints and for the collectionof anthropometric data. Before all testing, the upper limbmuscles were warmed up in an upper limb cycloergometer(Saratoga Cycle, Rand-Scot, Colorado, USA) for 5 minutes

without load [29]. The lower limbs warm up was performedin a cycloergometer (Ergo 167 Cycle, Ergo-Fit, Pirmasens,Germany) for 5 minutes at a 75 W load [20] and at a 20 km/hspeed. Madsen [30] suggests that warm up before isokinetic

ebwe

e33

est does not alter significantly the peak torque values ofnee and elbow assessments, therefore this procedure wasonducted exclusively with the purpose of avoiding musclenjuries during the isokinetic assessments.

An isokinetic dynamometer Biodex Multi-joint System IIBiodex Medical Systems, Inc, Shirley, New York, USA) wassed for assessment of the muscle performance on the kneend elbow flexion/extension of the athletes. The deviceas calibrated before each assessment and all procedures,

ncluding gravity correction of the torque measures, wereonducted according to the specifications of the equip-ent’s instructions manual [31]. Determination of the order

f the tests of the dominant and non-dominant limbs wasandom in an attempt to minimize possible effects of learn-ng that could affect the results [9]. The dominant upperimb was determined as the limb used in the majority ofhe daily life activities and the dominant lower limb wasetermined through the question: ‘‘which leg would you useo kick a ball as far as possible?’’. All assessments weretandardized in such a way that the upper limbs evalua-ions were always performed previously to the lower limbsvaluation.

For elbow flexion-extension testing, the subjects wereositioned sitting on the dynamometer chair stabilized bytraps across the chest and pelvis. The glenohumeral jointas positioned in 45◦ of abduction, 30◦ of flexion andpproximately 30◦ of lateral rotation, and the forearm wasaintained in supination. The mechanical rotation axis of

he dynamometer was aligned to the lateral humeral epi-ondyle and the established range of motion was 20◦ to 110◦

f elbow flexion (0◦ = full extension).For the knee joint testing, subjects were stabilized in

he dynamometer chair with straps across the chest, pelvisnd thigh and were oriented to maintain the arms crossed inront of the chest during the procedure [32]. The mechan-cal axis of rotation of the dynamometer was aligned tohe femoral lateral epicondyle. The resistance was appliedmmediately above the medial malleoli [33] and the estab-ished range of motion was 20◦ to 90◦ of knee flexion0◦ = full extension). The final degrees of knee extensionere avoided in the assessment because it is believed that

he combination of high muscle forces associated to theibial rotation, inherent to the tibiofemoral locking (screwome), might be potentially hazardous to the knee joint8].

All tests were performed in the reciprocal concentricode and each test was composed of five repetitions at

0◦/s and 10 repetitions at 360◦/s. The 60◦/s velocityas chosen due to the fact that this isokinetic speed isonsidered representative of muscle power [4] and the60◦/s speed was chosen as an attempt to evaluate thethletes in circumstances that more closely resemble theunctional activities observed in sports, considering theimitations of the dynamometer [14]. The procedure waslways initiated by the slowest velocity, with 2 minutes ofest between test velocities. Prior to testing, the subjectserformed, in each velocity, three sub-maximal and twoaximal contractions for familiarization purposes. During

ach series of contractions the same examiner was ver-ally encouraging the subjects, in a standardized vigorousay, attempting to stimulate them to perform maximalffort.

e

2

IiapShtavflpt1

lastupfos

2

T5mvWnobfttaTa

3

Dendsnt

fldw(lcfltne(

besia

4

Omitmtcl

34

.3. Data gathering

n the data collection, all contractions were considered andndividually analyzed by a single examiner. Each curve wasnalyzed manually, moving the cursor of the curve analysisrogram of the isokinetic dynamometer (Biodex Advantageoftware 4.0). The peak torque was considered to be theighest point at the chosen velocity in each curve. The peakorque values were normalized by the subject’s body massnd multiplied by 100. Mean values of each series of eacholunteer were considered in the statistical analysis. Theexor/extensor (F:E) ratio was calculated dividing the meaneak torque of the joint flexors by the mean peak torque ofhe joint extensors, with the obtained result multiplied by00.

Acceleration time (AcT) was obtained through the col-ection of the initial time of each contraction (speed = 0◦/s)nd the time at the instant in which the desired angularpeed was reached (60◦/s or 360◦/s). In order to obtainhe time to peak torque (TPT), the same initial time wassed (speed = 0◦/s) and the time in the instant that theeak torque was reached. These times were subtractedor each contraction, and the mean values of each seriesf each volunteer were considered in the statistical analy-is.

.4. Statistical analysis

he statistical analysis was carried out using the Statistica.0 for Windows software (StatSoft, Inc, Tulsa, USA). Theeans and standard deviations were calculated for all the

ariables using standard statistical procedures. The Shapiro-ilks and Kolmogorov-Smirnoff tests were used to test the

ormality of distribution. Accordingly, the paired t-testsr Wilcoxon tests were used to compare the differencesetween the scores of the dominant and non-dominant limbsor each dependent variable (peak torque, accelerationime and time to peak torque), for each movement pat-

ern (extension and flexion), for each joint complex (kneend elbow) and for each test velocity (60◦/s and 360◦/s).he level of significance (�) was set at 5% for all statisticalnalyses.

stws

Table 1 Knee extensor and flexor peak torque/body mass, accel

60◦/s

Dominant Non-dominan

ExtensionPT/BM 267.24 ± 47.02 262.26 ± 51.AcT 0.031 ± 0.006 0.032 ± 0.0TPT 0.428 ± 0.051 0.430 ± 0.0

FlexionPT/BM 139.81 ± 24.86 135.26 ± 23.AcT 0.047 ± 0.006 0.048 ± 0.0TPT 0.510 ± 0.095 0.513 ± 0.1

PT/BM: mean peak torque normalized by body mass × 100 (%); AcT: accstandard deviation; none of the comparisons reached statistical signifi

R. Scattone-Silva et al.

. Results

ata from the isokinetic assessment of the flexion andxtension of the knee is presented in Table 1. There waso significant difference between the dominant and non-ominant limbs in any of the evaluated parameters at bothpeeds. Moreover, the difference between mean peak torqueormalized by body mass (PT/BM) between limbs was lesshan 10% for both flexion and extension data.

Table 2 presents the data from the elbow extension andexion isokinetic assessment. In the elbow extension, theominant limb presented a smaller AcT (faster recruitment)hen compared to the non-dominant side at the 60◦/s speed

P = 0.01). Conversely, at the 360◦/s speed the non-dominantimb presented a smaller AcT (faster recruitment) whenompared to the dominant side (P = 0.05). Regarding elbowexion, the dominant upper limb presented greater peakorque normalized by body mass when compared to theon-dominant side at the 60◦/s speed (P = 0.02). No differ-nces were found at the 360◦/s speed in any of the variablesP > 0.05).

Data from flexor/extensor ratio for elbow and knee atoth speeds is presented in Fig. 1. No significant differ-nce was found in the knee F:E ratio between limbs at bothpeeds. At the elbow, significant difference was found onlyn the F:E ratio at 360◦/s, with the dominant limb presenting

higher F:E ratio than the non-dominant limb (P = 0.04).

. Discussion

ur objective was to assess the rate of torque develop-ent of the knee and elbow muscles of elite karate athletes

n order to recognize adaptations to regular training ofhis martial art, and to identify muscle torque or recruit-ent asymmetries that could predispose these individuals

o injuries in these joints. Our results showed no signifi-ant difference between the dominant and non-dominantower limbs in any of the variables in knee flexion or exten-

ion at both speeds. Moreover, the PT/BM values in bothhe dominant and non-dominant limbs were very similar, asere the muscle recruitment parameters, with differences

maller than 10% between limbs. This indicates that the daily

eration time and time to peak torque.

360◦/s

t Dominant Non-dominant

01 124.69 ± 26.07 123.80 ± 35.5103 0.095 ± 0.012 0.098 ± 0.02399 0.156 ± 0.007 0.155 ± 0.015

10 74.69 ± 16.71 74.41 ± 19.4011 0.138 ± 0.019 0.140 ± 0.02720 0.190 ± 0.025 0.193 ± 0.032

eleration time (s); TPT: time to peak torque (s); mean values andcance (P > 0.05).

Acceleration time and time to peak torque in karate athletes e35

Table 2 Elbow extensor and flexor peak torque/body mass, acceleration time and time to peak torque.

60◦/s 360◦/s

Dominant Non-dominant Dominant Non-dominant

ExtensionPT/BM 60.79 ± 17.94 58.87 ± 15.43 29.38 ± 6.79 33.72 ± 11.46AcT 0.050 ± 0.013a 0.058 ± 0.012 0.182 ± 0.027 0.169 ± 0.011a

TPT 1.100 ± 0.319 0.752 ± 0.423 0.238 ± 0.025 0.230 ± 0.017

FlexionPT/BM 84.73 ± 19.29a 79.33 ± 16.78 38.14 ± 6.23 37.47 ± 11.39AcT 0.035 ± 0.010 0.040 ± 0.005 0.130 ± 0.018 0.155 ± 0.027TPT 0.570 ± 0.185 0.472 ± 0.276 0.171 ± 0.019 0.209 ± 0.043

PT/BM: mean peak torque normalized by body mass × 100 (%); AcT: acceleration time (s); TPT: time to peak torque (s); mean values andstandard deviation.

limb

tdil

bmistbtaosds[s

a Denotes significantly different comparing to the contra-lateral

practice of karate does not produce bilateral asymmetriesin the lower limbs that could be related to an increased riskof injury. These results are in agreement with the ones ofProbst et al. [11] in which karate athletes presented bilat-eral differences in knee flexion and extension torque smallerthan 10% at 60◦/s and 180◦/s. Even considering a preva-lent use of the dominant lower limb in performing kicksduring combat, the workload intensity developed by thenon-dominant leg as support limb may have been sufficientto produce strength gains which reduced the asymmetriesbetween lower limbs of the athletes [7].

Studies assessing isokinetic muscle recruitment patterns(AcT and TPT) are extremely scarce in the literature and,therefore, comparison of our results is made difficult. Wefound no significant difference in the AcT and TPT in kneeflexion or extension in the comparison between the domi-nant and non-dominant limbs at both speeds. These resultsare in agreement with the ones of Probst et al. [11], whohave also found no asymmetry between limbs in the TPT

neither in knee flexion or extension. However, they foundthat karate athletes had a smaller TPT (faster recruit-ment) of the knee flexor muscles than a control groupof active subjects [11]. These results indicate that karate

tWoc

Figure 1 Knee and elbow peak torque flexor/extensor ratios (%).(P < 0.05).

(P < 0.05).

raining is capable of inducing gains in rate of torqueevelopment at the knee muscles without, however, produc-ng asymmetries regarding the dominant and non-dominantimbs.

The knee F:E ratio (hamstrings to quadriceps ratio) haseen considered a useful measure in detecting lower limbuscle imbalances that could predispose young athletes to

njuries such as hamstrings muscle strains [12], knee overuseyndromes [3], and ACL injuries [2,13]. It has been statedhat knee F:E ratios below 60% at slower speeds (60◦/s) andelow 80% at faster speeds (300◦/s) are significantly relatedo lower limb injuries [2,3,8] and, therefore, need to beddressed in injury prevention programs. The athletes fromur study presented F:E ratios smaller than these values,uggesting that they might be at increased risk of injury. Inirect contrast to our results, the karate athletes in anothertudy presented knee F:E ratios greater than 60% at 60◦/s11], but it should be noted that this previous study used aample of non-elite athletes of both genders, with smaller

ime of karate practice (belts ranged from green to black).e couldn’t find any other study assessing the knee F:E ratiof karate athletes at higher isokinetic speeds, so no directomparison can be made from our results.

* Significantly different comparing to the contra-lateral limb

e

tPdttidsipFlr

aheptpceeefbmgThwife1

stnitefadeFti

ctlbiousAlt

awtctflptruTdf

ahviitanwniftiijmnmtabhs

l[faaslasptectui

36

The results of the elbow assessment of this study indicatehat the athletes presented higher values of elbow flexionT/BM in the dominant limb when compared to the non-ominant limb at the speed of 60◦/s. One could assume thathese results might indicate some sort of muscle adaptationo karate training, however it is possible that this asymmetryn the elbow flexion strength could be the result of a pre-ominant use of the dominant limb in daily life activities,ince similar asymmetries have been found in previous stud-es of healthy non-athlete subjects [34,35]. A higher value ineak torque for elbow flexion also results in a higher elbow:E ratio as observed in this study on the dominant upperimb, but this doesn’t necessarily imply a muscle imbalanceesultant of sport’s practice.

The data regarding the elbow F:E ratio are also scarcend somewhat conflicting in the literature. Lin et al. [4]ave recently assessed the elbow F:E ratio in concentric andccentric isokinetic modes at 60◦/s and 240◦/s in baseballlayers. Interestingly, they found that elbow F:E concen-ric ratio greater than 76% at the higher speed was a strongredictor of elbow injury in these athletes. The authorsouldn’t find an explanation for these results, since it wasxpected that a higher F:E ratio could imply in a moreffective action of the elbow flexors decelerating the elbowxtension during pitching, which would protect the jointrom injury [4]. A greater F:E ratio at the elbow coulde the result of reduced strength of the elbow extensoruscles, which could imply in less force dissipation and

reater overload at the joint structures, resulting in injury.he elite karate athletes of this study presented muchigher values of elbow F:E ratio both at 60◦/s and 360◦/s,hich could imply a higher susceptibility to elbow injuries

n this population. Ellenbecker and Roetert [9] have alsoound results greater than that in young elite tennis play-rs, who presented elbow F:E ratios ranging from 90 to05%.

It should be noted that in the study of Lin et al. [4], theubjects were positioned for the isokinetic assessment withhe forearm in neutral position, while we assessed the isoki-etic elbow flexion/extension with the forearm positionedn supination, as recommended by the dynamometer instruc-ions manual [31]. It has been shown that higher isokineticlbow flexion moments are obtained in assessments with theorearm positioned in supination when compared to the fore-rm in neutral position [35]. This methodological differenceuring the isokinetic assessment could explain the higherlbow flexor peak torque values (and the consequent higher:E ratio) observed in our study, not necessarily implyinghat the athletes of our study are at a higher risk of elbownjuries.

Regarding the recruitment patterns of the elbow mus-les, the athletes from this study presented a smaller AcT ofhe elbow extensors at the speed of 60◦/s in the dominantimb when compared to the non-dominant. This might alsoe the result of a predominant use of the dominant limbn daily life activities, especially since this difference wasnly found in the slower velocity, which is closer to the onessed in routine non-athletic activities. Conversely, at the

peed of 360◦/s the non-dominant limb presented a smallercT for elbow extension when compared to the dominant

imb. This result probably indicates a neuromuscular adap-ation to regular karate training, considering that these

D

Tc

R. Scattone-Silva et al.

thletes usually perform dozens of high velocity punchesith both upper limbs every day. The faster recruitment of

he non-dominant upper limb extensor muscles may be aonsequence of the fighting position adopted by some ofhe athletes during training and competition. It is commonor the athletes to position their dominant upper and lowerimbs forward in the sagittal plane while assuming fightingosition. In this posture, the athletes can kick faster withhe dominant lower limb, and the dominant upper limbemains used to perform blocks, while the non-dominantpper limb performs fast attack and counter-attack actions.hus, the non-dominant upper limb is highly emphasizeduring training sections, which may explain why they had aaster recruitment time at the high velocity test.

This study had several limitations that should beddressed. The relatively small sample of subjects mayave interfered in the identification of asymmetries of someariables. This small sample was resultant of the difficultyn finding elite competitor karate athletes that would fit thenclusion criteria. The absence of the evaluation of a con-rol group is another important limitation of our study. Thessessment of healthy subjects would allow us to separateatural from training-induced asymmetries. Nevertheless,e have established comparisons from our results with theormative data available in the literature, in an attempt todentify adaptations that could be considered risk factorsor musculoskeletal injuries. Furthermore, it is importanto be aware that the relationship of F:E ratio data andnjury proneness in applied sport situations has methodolog-cal limitations since this measurement is based on singleoint maximal voluntary contractions and a limited range ofotion of the specific joint [7]. In spite of that, it is worth

oting that in the study of Lin et al. [4], none of the torqueeasures (concentric or eccentric) was significantly related

o elbow injury in the athletes, and the F:E concentric ratiot the 240◦/s speed was the only parameter that was capa-le of predicting elbow injury in young baseball players. Thisighlights the importance of the F:E ratio as a preseasoncreening measure for injury prevention.

In summary, the differences between limbs in the ath-etes of our study were considered within normative data1,2] — less than 10 to 15% — and seem to emphasize therequent use of bilateral patterns in this martial art. In anttempt to achieve the highest performance possible, elitethletes demand a great deal of time training to improvetrength and technique in both dominant and non-dominantimbs, in order to become more effective in combat. Thebsence of bilateral lower limb differences observed in ourtudy supports this theory. However, the athletes of theresent study showed agonist/antagonist asymmetries athe knee and elbow joints that have been related in the lit-rature to a higher predisposition to injuries. Based on theseomparisons, strengthening programs may be indicated forhis population. More studies, with more methodologicalniformity, are necessary for further enlightenment regard-ng the rate of force development in athletes.

isclosure of interest

he authors declare that they have no conflicts of interestoncerning this article.

s

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

Acceleration time and time to peak torque in karate athlete

Acknowledgements

The authors would like to thank all of the volunteers andthe Conselho Nacional de Desenvolvimento Científico e Tec-nológico (CNPq) for the financial support to this research.

References

[1] Safran MR, Seaber AV, Garrett WE. Warm-up and muscularinjury prevention: an update. Sports Med 1989;8:239—49.

[2] Knapik JJ, Bauman CL, Jones BH, Harris JM, Vaughan L. Pre-season strength and flexibility imbalances associated withathletic injuries in female collegiate athletes. Am J Sports Med1991;19:76—81.

[3] Devan MR, Pescatello LS, Faghri P, Anderson J. A prospectivestudy of overuse knee injuries among female athletes withmuscle imbalances and structural abnormalities. J Athl Train2004;39:263—7.

[4] Lin YC, Thompson A, Kung JT, Chieh LW, Chou SW, Lin JC. Func-tional isokinetic strength ratios in baseball players with injuredelbows. J Sport Rehabil 2010;19:21—9.

[5] Calmels PM, Minaire P. A review of the role of the ago-nist/antagonist muscle pairs ratio in rehabilitation. DisabilRehabil 1995;17:265—76.

[6] Siqueira CM, Pelegrini FR, Fontana MF, Greve JM. Isokineticdynamometry of knee flexors and extensors: comparative studyamong non-athletes, jumpers athletes and runner athletes. RevHosp Clin Fac Med São Paulo 2002;57:19—24.

[7] Magalhães J, Oliveira J, Ascensão A, Soares J. Concentricquadriceps and hamstrings isokinetic strength in volleyball andsoccer players. J Sports Med Phys Fitn 2004;44:119—25.

[8] Aagaard P, Simonsen E, Magnusson S, Larsson B, Dyhre-PoulsenP. A new concept for isokinetic hamstring: quadriceps musclestrength ratio. Am J Sports Med 1998;26:231—7.

[9] Ellenbecker TS, Roetert EP. Isokinetic profile of elbow flexionand extension strength in elite junior tennis players. J OrthopSports Phys Ther 2003;33:79—84.

[10] Fonseca ST, Ocarino JM, Silva PL, Brício RS, Costa CA, Wan-ner L. Characterization of professional soccer players’ muscleperformance. Rev Bras Med Esporte 2007;13:143—7.

[11] Probst MM, Fletcher R, Seelig DS. A comparison of lower-body flexibility, strength, and knee stability between karateathletes and active controls. J Strength Cond Res 2007;21:451—5.

[12] Crosier JL, Ganteaume S, Binet J, Genty M, Ferret JM. Strengthimbalances and prevention of hamstring injury in profes-sional soccer players: a prospective study. Am J Sports Med2008;36:1469—75.

[13] Myer GD, Ford KR, Foss KD, Liu C, Nick TG, Hewett TE. Therelationship of hamstrings and quadriceps strength to anteriorcruciate ligament injury in female athletes. Clin J Sports Med2009;19:3—8.

[14] Mendonca LM, Bittencourt NF, Anjos MT, Silva AA, Fonseca ST.Isokinetic muscular assessment of the shoulder joint in ath-letes from the male under-19 and under-21 Brazilian volleyballteams. Rev Bras Med Esporte 2010;16:107—11.

[15] Sbriccoli P, Camomilla V, DiMario A, Quinzi F, Figura F, FeliciF. Neuromuscular control adaptations in elite athletes: thecase of top level karateka. Eur J Appl Physiol 2010;108:1269—80.

[

e37

16] Kannus P. Isokinetic evaluation of muscular performance: impli-cations for muscle testing and rehabilitation. Int J Sports Med1994;15(Suppl. 1):S11—8.

17] Miller LE, Pierson LM, Nickols-Richardson SM, Wootten DF, Sel-mon SE, Ramp WK, et al. Knee extensor and flexor torquedevelopment with concentric and eccentric isokinetic training.Res Q Exerc Sport 2006;77:58—63.

18] Schlumberger A, Laube W, Bruhn S, Herbeck B, Dahlinger M,Fenkart G, et al. Muscle imbalance: fact or fiction? Isok ExercSci 2006;14:3—11.

19] Chen WL, Su FC, Chou YL. Significance of acceleration periodin a dynamic strength testing study. J Orthop Sports Phys Ther1994;19:324—30.

20] Van Cingel RE, Kleinrensink G, Uitterlinden EJ, Rooijens PP,Mulder PG, Aufdemkampe G, et al. Repeated ankle sprains anddelayed neuromuscular response: acceleration time parame-ters. J Orthop Sports Phys Ther 2006;36:72—9.

21] Tan KS. Constructing a martial tradition: rethinking a popularhistory of Karate-Do. J Sport Soc Issues 2004;28:169—92.

22] Sorensen H, Zacho M, Simonsen E, Dyhre-Poulsen P, KlausenK. Dynamics of martial arts high frontal kick. J Sport Sci1996;14:483—95.

23] Mori S, Ohtani Y, Imanaka K. Reaction times and anticipatoryskills of karate athletes. Hum Mov Sci 2002;21:213—30.

24] Camomilla V, Sbriccoli P, DiMario A, Arpante A, Felici F. Com-parison of two variants of a Kata technique (Unsu): theneuromechanical point of view. J Sport Sci Med 2009;8:29—35.

25] Myer GD, Ford KR, Hewett TE. Rationale and clinical techniquesfor anterior cruciate ligament injury prevention among femaleathletes. J Athl Train 2004;39:352—64.

26] MacWilliams BA, Wilson DR, Desjardins JD, Romero J, Chao EY.Hamstrings co-contraction reduces internal rotation, anteriortranslation, and anterior cruciate ligament load in weight-bearing flexion. J Orthop Res 1999;17:817—22.

27] Lloyd DG, Buchanan TS. Strategies of muscular support of varusand valgus isometric loads at the human knee. J Biomech2001;34:1257—67.

28] Valkering KP, van der Hoeven H, Pijnenburg BC. Posterolateralelbow impingement in professional boxers. Am J Sports Med2008;36:328—32.

29] Santos JRM, Antonelo MC, Coelho RBS, Takahashi R, SerrãoFV. Evaluation of eccentric and concentric peak torque inshoulder rotator muscles of young tennis players. Fisioter Bras2006;7:204—8.

30] Madsen OR. Torque, total work, power, torque accelerationenergy and acceleration time assessed on a dynamometer:reliability of knee and elbow extensor and flexor strength mea-surements. Eur J Appl Physiol Occup Physiol 1996;74:206—10.

31] Wilk KE. Isokinetic testing: goals, standards and knee test inter-pretation. Biodex Medical System Operation Manual; 1991.

32] Bohannon RW, Gajdosik RL, Leveau BF. Isokinetic knee flexionand extension torque in the upright sitting and semi-circledsitting positions. Phys Ther 1986;66:1083—6.

33] Dvir Z. Isokinetics: muscle testing interpretation and clinicalapplications. 1st ed. Edinburg: Churchill Livingstone; 1995.

34] Gallagher MA, Cuomo F, Polonsky L, Berliner K, Zuckerman JD.Effects of age, testing speed, and arm dominance on isokinetic

strength of the elbow. J Shoulder Elbow Surg 1997;6:340—6.

35] Mandalidis DG, O’Brien M. Isokinetic strength of the elbow flex-ors with the forearm in supination and in neutral position. IsokExerc Sci 2001;9:111—7.