Eur J Appl Physiol (2002) 87: 409–415 DOI 10.1007/s00421-002-0643-z
O R I GI N A L A R T IC L E
Renaud Halin Æ Philippe Germain Æ Olivier Buttelli Branislaw Kapitaniak
Differences in strength and surface electromyogram characteristics between pre-pubertal gymnasts and untrained boys during brief and maintained maximal isometric voluntary contractions Accepted: 18 April 2002 / Published online: 15 June 2002 Springer-Verlag 2002
Abstract The present study was aimed at investigating differences of maximal strength (Fmax) of the elbow flexors and characteristics of the surface electromyogram (EMG) between six gymnasts (G) and six untrained (UT) 10-year-old boys during brief and maintained maximal voluntary isometric contraction (MVC). The Fmax was estimated during 5 s MVC (maximal test, MT) and normalized to the cross sectional area (CSA) of the arm. The EMG signal of the biceps brachii was recorded during MT and during a 25 s maintained MVC (fatigue test). Values were calculated for root-mean-square (rmsMT) and mean power frequency (MPFMT) of the EMG signal for the duration of the MT. For the fatigue test, MPF were normalized to the initial value (MPFn) and kinetics were expressed by the slope coefficient of linear regression. Although Fmax and Fmax/CSA tended to be higher for G than UT, the differences did not reach significance. The MPFMT was significantly higher for G [mean (SD)][136 (8) Hz] than for UT [125 (9) Hz]. The MPFn slope coefficients were significantly greater for G than for UT [–1.0 (0.2) and –0.5 (0.3), respectively]. When all the children were considered, Fmax was significantly correlated to MPFMT (r=0.61). These results showed that gymnasts tend to have higher Fmax and Fmax/CSA accompanied by a significantly higher MPFMT and a steeper MPF downshift. Moreover, children with greater strength tended to have higher MPFMT. It is suggested that spatial and/or
R. Halin (&) Æ P. Germain Æ O. Buttelli Laboratoire de la Performance Motrice, Faculte´ du Sport et de l’Education Physique d’Orle´ans, Rue de Vendoˆme, BP 6737, 45062 Orle´ans Cedex 2, France E-mail:
[email protected] Tel.: +33-02-38690003 Fax: +33-02-38417260 B. Kapitaniak Laboratoire de Physiologie du Travail et du Sport, Universite´ Paris VI Pierre et Marie Curie, Faculte´ de Me´decine de la Pitie´ Salpetrie`re, 91 boulevard de l’Hoˆpital, 75630 Paris Cedex 13, France
temporal recruitment of more fatigable fast motor units could have been enhanced in G and more generally, that it could be a mechanism that would explains, in part, the level of force production in children. Keywords Children Æ Gymnastic Æ Electromyography Æ Maximal voluntary contraction
Introduction Children are participating in more and more intense training and for some sports, it involves very young children. This is particularly true for gymnastics for whom training can begin as young as 6–7 years old and for boys aged 9–10 years can be intense (10 h a week or more). It has been observed that gymnasts over 11 years old were stronger than normal untrained boys (Maffulli et al. 1994). More generally, it is well established that resistance training can be effective in increasing strength in children (see review of Blimkie 1993). Therefore, it would seem that such training could have an influence on the neuromuscular system. Classically, in adults, long-term resistance training induces strength increases accompanied by marked muscle hypertrophy (MacDougall et al. 1979). Hence, the increase in strength is essentially, but not fully, explained by this augmentation of the muscle mass. However, it has been proposed that the increase in strength exhibited by individuals during the initial few weeks of training can be attributed to neural factors (Moritani and DeVries 1979) as no hypertrophy is generally observed. Those neural mechanisms could be an increase in the maximal firing rate of motor units (MU; temporal recruitment, Kamen et al. 1998; Leong et al. 1999) and an increase in MU recruitment (spatial recruitment; Akima et al. 1999). For pre-pubertal children, the majority of the studies have shown that muscle strength could significantly increase without a concomitant augmentation in muscle size as assessed by anthropometrical measurements, and
410
this, even after a period and intensity of training for which muscle hypertrophy is evident in adults (Weltman et al. 1986; Ramsay et al. 1990). Even when slight muscle hypertrophy was sometimes reported for children (Mersch and Stoboy 1989) it was always weak compared to the augmentation in strength. As for adults, in the absence of marked hypertrophy, strength increase after training in children has been essentially attributed to neural factors and muscle recruitment (see review of Blimkie 1993). Surface electromyography (EMG) is a useful means of uninvasively studying the neuromuscular system (De Luca 1997). For example, an enhanced neuromuscular activation after resistance training can be illustrated by an increase of the quantitative parameters of the EMG [root mean square (rms) or integrated EMG] in adults (Bandy and Hanten 1993; Ha¨kkinen and Komi 1983) as well as in children (Komi et al. 1978; Ozmun et al. 1994). In the frequency domain, studies both in the unfatigued and in the fatigued state indicate that the fibre-type proportions might influence qualitative parameters of the EMG signal such as the mean (MPF) or the median (MDF) frequency of the power spectrum during static (Viitasalo and Komi 1978; Gerdle et al. 1997) and dynamic contractions (Komi et Tesch 1979; Taylor et al. 1997). Moreover, it has been observed that training can have an impact on localised muscle fatigue in children who are ill (O¨berg et al. 1994) or in healthy adults (Gabriel et al. 2001) as shown by the greater downshift in MPF observed during a fatigue test after training. A recent study has also reported greater downshifts in MDF in weightlifters compared to untrained adults during maintained isometric contractions (Felici et al. 2001). Those observations seemed to be related to changes in the structure of the muscle or the pattern of activation. Therefore, neuromuscular adjustments might be illustrated by analyses of the EMG signal in the time and/or frequency domain. Our experiment was designed to study the effects of long-term and intense gymnastic training on the manifestation of strength and EMG characteristics during maximal effort in young boys. The assumption is being made that intense and long-term competitive gymnastic training in children would lead to an augmentation of muscle strength without a marked hypertrophy of the muscles. This augmentation in strength would be essentially due to an enhancement of the command drive (spatial and/or temporal recruitment), which then would be estimated by an analysis of the EMG signal during brief and maintained maximal voluntary contraction (MVC).
6 trained gymnasts (G) and 6 untrained boys (UT). The G had been practising their sport for several years (3–5 years) for 12–15 h a week. The UT group was composed of boys who had had no regular sport training during the past 3 years. The mean ages, heights and body masses of the two groups were not significantly different (Table 1).
Methods
Table 1. Characteristics of the experiment population
Muscle The muscle studied was the biceps brachii. This muscle was chosen because it is intensively used during gymnastic training but less used in the usual daily life of a child compared to the muscles of the lower limb. Consequently, greater training-induced differences could be expected between the groups. Anthropometrical measurements The skinfold thickness of the arm was measured using skinfold callipers. Skinfolds were measured at the anterior, posterior, medial and lateral aspects of the right arm (circumferential skinfolds). The circumference was measured at the thickest section of the arm with subjects standing in the anatomical position. The cross-sectional area (CSA) of the arm was then calculated according to the equation of Moritani and DeVries (1979): h X . i2 CSA ¼ p ðC=2pÞ f 4 where C is the circumference of the arm and f represents the circumferential skinfolds. Procedure and tests As the aim of the present experimentation was to study the behaviour of the neuromuscular system when it was working at its maximal capacity, either during brief or fatiguing efforts, tests were composed only of MVC. A specially designed ergometer was built to measure isometric contractions of the elbow flexors in a standardised position. The subjects were seated on a chair. The right arm was positioned horizontally in the sagittal plane of the body passing by the right shoulder. The forearm was vertical and the hand was in a supine position. A wrist-cuff was attached proximal to the styloid process. The force exerted was measured using a strain gauge (Alpha-P Kosmos Ditel digital meters, A-D converter with a sampling frequency of 16 Hz, and a range of force of 0–2,000 N). The angles of forearm-arm and arm-trunk were both fixed at 90. The angles were controlled using an electronic goniometer (Alpha-C Kosmos Ditel digital meters, A-D convertor with a sampling frequency of 16 Hz). The back and shoulders of the subject were tightly secured with straps to limit undesirable movements. A standardised warm-up was performed 3 min before the tests. It was composed of ten contractions of increasing intensity. Each contraction lasted 5 s and was followed by 20 s of rest. Maximal test In the maximal test (MT) each subject performed three brief MVC of 5 s duration separated by 2 min of complete rest. The highest
Subjects A group of 12 right-handed young boys participated in the study. These boys and their parents were informed about the experiment procedure and all gave their signed consent. This study was approved by the Ethics Committee of La Pitie´ Salpetrie`re, Faculty of Medicine, Paris. The experiment population was composed of
Age (years) Height (m) Body mass (kg)
Gymnasts (n=6)
Untrained boys (n=6)
Mean
SD
Mean
SD
10.5 1.36 30.2
0.6 0.04 3.4
10.6 1.43 35.5
0.5 0.06 6
411 value was considered to be the maximal force (Fmax) and was retained for EMG analysis. The Fmax was standardized according to the CSA of the arm (Fmax/CSA).
MPFn kinetics were compared between the two groups. Relationships between variables were established using the Pearson product moment. In all statistical tests the difference was considered to be significant at the P‡0.05 level.
Fatigue test At 3 min after the last brief contraction, the subjects performed one maintained MVC lasting 25 s. This duration was chosen because anything longer could have led to a fall in motivation of the child. The subjects were vigorously encouraged to exert their maximal strength during the whole of the contraction time. EMG signal recording and processing The EMG signal from the antero-internal surface of the arm was detected during the MT and the fatigue test using two Ag-AgCl surface electrodes of 8 mm diameter. The subjects were prepared for the placement of the EMG electrode by shaving the skin and cleaning it with alcohol wipes. Electrodes were placed away from the motor point (Zipp 1982) on the distal third of the short head of the belly of the contracted biceps brachii, parallel to the main direction of the muscle fibres. A reference electrode was placed over the acromioclavicular joint. A 20 mm inter-electrode distance was used for all subjects and all inter-electrode resistances measured were below 10 kW. The raw EMG signal was preamplified (·600) near the site of the electrodes, and recorded on a Myodata portable system (Mazet Electronique, France) at an acquisition rate of 1,024 Hz (5–400 Hz bandwidth) and A-D converted (12 bit resolution). The Myodata system had a minimal common mode rejection ratio of 100 db, an input impedance of 10 GW and a maximal output impedance of 10 W. Data were stored on a flash card memory and transferred to a personal computer for treatment. Myoelectric signal treatment was performed using the Me12F software (Mazet Electronique). To avoid transient phenomena due to rising force, the first second of contraction was discarded in the MT as well as in the fatigue test. For the fatigue test, the last 2.5 s were also discarded to avoid the risk that some children could have anticipated the end of the effort. Thus, only 21.5 s of effort were kept for EMG analysis. For MT, the rms amplitude and the power density spectrum (PDS) of the signal were calculated for the best trial. The rms was calculated for a 2 s epoch (rmsMT). For the same epoch, three successive PDS were calculated after Hamming windowing using the fast Fourier transform technique for 1 s windows overlapping each other by 0.5 s. The MPF of the PDS was determined for each epoch. The three MPF values were then averaged, thus giving the maximal MPF value (MPFMT). For the fatigue test, the MPF kinetic was determined by calculating successive PDS using signals epoch of 1 s overlapping each other by 0.5 s, thereby generating 42 values during the 21.5 s of contraction. All MPF values were normalized (MPFn) in relation to the initial MPF value. In order to use a standard methodology and the same analytical approach for all subjects, we have chosen to quantify the MPFn kinetic by the slope coefficient of a least square linear regression. A linear regression model was adopted because from a statistical point of view, it is the first model that has a significant increase in the incremental F value, and is therefore assumed to be the best model to use. The MPFn kinetics were also determined using the percentage of variation between the mean of the first three MPFn values (s 1–2) and the mean of the last three MPFn values (s 20.5–21.5).
Results Anthropometrical measurements Although G tended to have larger CSA of the arm than UT, the difference was not significant [23.0 (2.9) cm2 and 20.8 (2.4) cm2, respectively]. Maximal test When absolute Fmax were considered, G was stronger [119 (29) N] than UT [96 (16) N] but the difference was not significant. When Fmax was adjusted according to muscle CSA, G was stronger than UT [5.2 (1.1) NÆcm–2 compared to 4.6 (0.7) NÆcm–2, respectively] but the difference was not significant. The MPFMT of G was significantly (P
