Morphological and functional differences in the elbow extensor muscle between highly trained male and female athletes

Transcription

1 Eur J Appl Physiol (1998) 78: 109±114 Ó Springer-Verlag 1998 ORIGINAL ARTICLE Y. Ichinose á H. Kanehisa á M. Ito Y. Kawakami á T. Fukunaga Morphological and functional differences in the elbow extensor muscle between highly trained male and female athletes Accepted: 7 January 1998 Abstract The purpose of this study was to investigate the di erence in the morphological and functional aspects of the triceps brachii muscle between highly trained male and female athletes who were members of the 1996 Japanese Olympic teams in each of three different events: soccer, gymnastics and judo. The thickness (TB mt ) and bre pennation angle (TB pen ) of the triceps brachii muscle and force output during elbow extensions were determined using a B-mode ultrasound apparatus and an isokinetic dynamometer, respectively. The TB mt and its value relative to upper arm length (TB mt /l ua ) were signi cantly larger in the men than in the women in all the events except judo. In all the subjects, a signi cant correlation was found between TB mt /l ua ) and TB pen (r ˆ 0.721, P < 0.05). The existence of the sex di erence in TB pen within the same event was in agreement with that observed in TB mt /l ua except for the soccer players. The TB pen of the soccer players were similar in both sexes although a signi cant sex di erences was found in TB mt /l ua. The isokinetic forces measured using the two velocities 60 á s )1 (F 60 ) and 180 á s )1 (F 180 ) were signi cantly correlated to the cross-sectional area (CSA) of the triceps brachii muscle estimated from TB mt (r ˆ 0.702, P < 0.05 for F 60, and r ˆ 0.776, P < 0.05 for F 180 ). No signi cant sex di erences were found in either F 60 /CSA or F 180 /CSA in any of the events. From these results, it could be assumed, at least in the Olympic athletes tested in this study, that the bre angulation of the triceps brachii muscle was almost the same in the two sexes if allowance was made for the di erence in the muscle size, and the sex di erence in force generation capability of the triceps brachii muscle could in the main be attributed to the di erence in CSA rather than in the architectural characteristics. Y. Ichinose á H. Kanehisa á M. Ito á Y. Kawakami T. Fukunaga (&) Department of Life Sciences (Sports Sciences), University of Tokyo, Komaba, Meguro-ku, Tokyo, 153 Japan Key words Olympic athletes á Sex di erence á Muscle thickness á Fibre pennation angles á B-Mode ultrasound Introduction It has been well documented that an important factor responsible for the sex di erence in muscle strength is the di erence in muscle size. Recent improvement in the technologies for assessing body composition such as magnetic resonance imaging (MRI) and ultrasonography using brightness mode (B-mode) have made possible the measurement of both the size and bre pennation angles of human muscles (Henriksson-Larse n et al. 1992; Kawakami et al. 1993, 1995; Narici et al. 1992; Rutherford and Jones 1992). While the force produced in the tendon by the contraction of muscle bres will be less than the sum of the forces produced in the individual bres due to angular e ects, it has been shown that pennation enables more contractile material to be placed in parallel for a given anatomical cross-section (Rutherford and Jones 1992). Alexander and Vernon (1975) have presented a model to estimate the force at the tendon in the lower limb muscles from muscle volume, muscle layer thickness, pennation angles and force per unit of cross-sectional area (CSA), and suggested that the tendon force would increase up to a pennation angle of 45. However, little information is available from previous studies on the relationship between bre pennation angles and force generation capability in human muscles. Moreover, since less attention has been paid to possible di erences between the sexes with respect to muscle architecture, it is not well known whether bre angulation as well as muscle size could explain the sex di erence in the ability to generate force. While previous ndings have shown a great deal of variation in the magnitude of the sex di erences in muscle strength, one consistent nding has been that larger sex di erences exist in upper body strength relative to lower body strength (Bishop et al. 1987). On the other hand, these observed di erences may have been

2 110 in uenced by various factors such as genetic di erences in muscle mass and/or neuromuscular function, behavioural di erences in the degree of participation of strength-developing activities, experience in maximal voluntary e ort and/or motivation to perform (Bishop et al. 1987). To minimize the in uence of behaviourally linked sex di erences and to permit a more objective comparison of muscle size and strength capability in the upper body, it is necessary to measure these variables in groups of men and women who have similar backgrounds of physical activity. In the present study the thickness and bre pennation angle of the triceps brachii muscle were measured using a B-mode ultrasound technique in highly trained male and female athletes and the force output during elbow extensions was determined isokinetically. The purpose of this study was to investigate the di erences in the morphological and functional aspects of the triceps brachii muscle between the elite male and female athletes. Methods Subjects The subjects in this study was 28 male and 33 female athletes who were members of the 1996 Japanese Olympic team, as follows ± 16 male soccer players [mean age 20.8 (SEM 0.3) years, mean height (0.3) cm, mean body mass 72.3 (0.9) kg], 20 female soccer players [23.7 (0.8) years, (0.8) cm, 57.3 (1.3) kg], 7 male gymnasts [24.9 (0.8) year, (2.2) cm, 59.9 (2.3) kg], 7 female gymnasts [17.6 (0.5) years, (2.0) cm, 41.8 (1.4) kg], 5 male judo athletes [23.8 (1.2) years, (3.2) cm, 76.5 (6.0) kg] and 6 female judo athletes [21.2 (1.0) years, (2.7) cm, 63.6 (3.9) kg]. The sample represented the top athletes in these sports in Japan at the time data were collected. This study was approved by the o ce of the Department of Sports Sciences, University of Tokyo, and was consistent with their requirements for human experimentation. All the subjects were volunteers; they were informed of the potential risks and bene ts of the various tests and all gave written informed consent. Measurement of muscle thickness and bre pennation angles Muscle thickness (TB mt ) and bre pennation angles (TB pen ) of the triceps brachii muscle in the right arm were measured by a B-mode ultrasound apparatus (SSD-500, Aloka) as has been described by Kawakami et al. (1993). The subjects stood with their arms relaxed in the extended position. A single cross-sectional plane was measured to determine TB mt at a position 40% of the distance from the lateral epicondyle to the acromion process of the scapula. A transducer with a 7.5-MHz scanning head was placed perpendicular to the tissue interface and perpendicular to the underlying humerus. The scanning head was coated with water-soluble transmission gel, which provided acoustic contact without depressing the dermal surface. The subcutaneous adipose tissue-muscle interface and the muscle-bone interface were identi ed from the ultrasonic image, and the distance from the adipose tissue-muscle interface to the muscle-bone interface was taken as being representative of muscle thickness. Both the long and medial heads of the triceps muscle were included in the muscle layer. For measuring TB pen, the centre of the transducer of the ultrasonic apparatus was placed in the same position as for TB mt, this time parallel to the long head of the triceps. Before the measurement, the observer visually con rmed the muscle belly of the long head and drew a longitudinal line on the dermal surface in the centre of the belly with a surgical marking pen. The angles between the echoes of the aponeurosis of the triceps muscle and the echoes from interfaces among the fascicles of the long head were measured and de ned as muscle- bre pennation angles. The accuracy and test-retest reproducibility of measuring TB mt and TB pen have been identi ed in an earlier study (Kawakami et al. 1993). Measurement of force The present study employed a Cybex II isokinetic dynamometer to determine the force output during elbow extension. Prior to the tests, the Cybex system was calibrated using known masses on the lever arm. The maximal voluntary torque in elbow extension was measured isokinetically at two constant velocities of 60 á s )1 and 180 á s )1. To standardize the measurement and to localize the action to the appropriate muscle group, the subjects lay supine on a bed with support for the back, elbow, shoulders, and hips. The axis of the elbow joint was aligned with the axis of the dynamometer lever arm of the Cybex machine. After a period of standardized warm-up and familiarization with the measurement apparatus the subjects were encouraged to perform maximal voluntary muscle actions isokinetically throughout a range of elbow-joint angles from 110 exed position to full extension. The test order of the experimental conditions was random for each subject. A period of 10 s was permitted between successive attempts and a 1-min rest allowed between each contraction. The greatest torque from three to ve attempts was recorded for each test velocity. The force was calculated by dividing the torque by the length of the lever arm. Force values obtained at 60 á s )1 and 180 á s )1 were referred to as F 60 and F 180. Calculation of CSA To allow comparison between the subjects of force generation capability normalized to unit muscle size, the CSA of the triceps brachii muscle was calculated. The equation used was: CSA ˆ p TB mt =2 2 where p is the constant , TB mt is muscle thickness in centimetres. The estimated CSA, was used to calculate the ratios F 60 / CSA and F 180 /CSA. Statistics The linear correlation coe cient (r) was calculated using the method of least squares. A two-way analysis of variance (ANOVA) was used to identify the e ect of sex or event on the measurements. When the signi cant e ect of sex or event was found, the Sche eâ test was used to identify the location of signi cant di erences among several means. The probability level accepted for statistical signi cance was set at P < Results Table 1 shows the results of TB mt and TB pen measurements. The TB mt and its value relative to upper arm length (TB mt /l ua ) were signi cantly larger in the men than in the women in all events except for judo. In comparisons among groups of athletes the male judo athletes and gymnasts showed signi cantly larger TB mt and TB mt /l ua than the male soccer players, and female judo athletes than the female gymnasts and soccer players. In all the subjects, the range of variation in

3 111 Table 1 Descriptive data on muscle thickness (TBmt) and bre angle (TBpen). SO soccer players, GY gymnasts, JU judo athletes, CSA estimated value for corss-sectional area derived from the equation (TBmt/2) 2 TBmt/lua TBmt per unit of upper arm length SO GY JU Di erences among events (P < 0.05) Men (n = 16) Women (n = 20) Men (n = 7) Women (n = 7) Men (n = 5) Women (n =6) mean SEM mean SEM Mean SEM mean SEM mean SEM mean SEM TB mt (mm) * * Men JU, GY > SO CSA (cm 2 ) Women JU > GY, SO TBmt/lua ( 10 )1 ) * * Men JU, GY > SO Women JU > GY, SO TBpen ( ) * Men JU > GY > SO Women JU? GY, SO * Signi cantly lower (P < 0.05) than for men within the same event TB pen was almost the same in the two sexes (from 12 )to 29 in the men and from 11 to 30 in the women). The TB pen was signi cantly correlated to TB mt /l ua (r ˆ 0.721, P < 0.05, Fig. 1) in all the subjects. The level of the correlation coe cient between TB mt /l ua and TB pen was almost the same even when the regression analysis was performed on the separate groups of the men and women [r ˆ (P < 0.05) for the men and r ˆ (P < 0.05) for the women]. In both sexes, TB pen was the largest in the judo athletes and the smallest in the soccer players. While the result of a twoway ANOVA on TB pen showed no signi cant sex e ect (F ˆ 2.007, P > 0.05), there was a signi cant sex event interaction (F ˆ 5.128, P < 0.05). In the comparison within the same event, the gymnasts only showed a signi cant sex di erence in TB pen. Both F 60 and F 180 were signi cantly correlated to the estimated CSA in all subjects [r ˆ (P < 0.05) for F 60 and r ˆ (P < 0.05) for F 180 ]. The descriptive data for F 60 and F 180 expressed in absolute terms and values relative to the estimated CSA, F 60 /CSA and F 180 / CSA, respectively, are summarized in Table 2. Both F 60 and F 180 were signi cantly larger in the men than in the women in all events except for judo athletes. In the men, the gymnasts and judo athletes showed signi cantly higher F 60 and F 180 values than the soccer players. For the women, F 60 and F 180 values were signi cantly larger in the judo athletes than in the other two groups. The F values of sex e ect on F 60 /CSA (F ˆ 0.636) and F 180 / CSA (F ˆ 0.072) were insigni cant, but those of event e ect were signi cant [F ˆ (P < 0.05) for F 60 / CSA and F ˆ (P < 0.05) for F 180 /CSA]. In the comparisons among athletes for F 60 /CSA, the male gymnasts and soccer players showed signi cantly higher average values than the male judo athletes, and the female soccer players higher average values than the female judo athletes. In the case of F 180 /CSA, only the female soccer players showed a signi cantly higher value than the female judo athletes. Fig. 1 Relationship between TB m /l ua and TB pen

4 112 Table 2 Descriptive data on isokinetic force and force per unit of estimated cross-sectional area (CSA). F60 Force developed at 60 á s )1, F180 force developed at 180 á s )1, F60/CSA force per unit of estimated CSA at 60 á s )1, F180/CSA force per unit of estimated CSA at 180 á s )1. For other de nitions see Table 1 SO GY JU Di erences among events (P < 0.05) Men (n = 7) Women (n = 7) Men (n = 5) Women (n =6) Men (n = 16) Women (n = 20) mean SEM mean SEM mean SEM mean SEM mean SEM mean SEM F60 (N) * * Men GY > JU, SO Women JU > SO, GY F180 (N) * * Men JU, GY > SO Women JU > SO, GY F60/CSA (N á cm )2 ) Men GY, SO > JU Women SO > JU F180/CSA (N á cm )2 ) Women SO > GY * Signi cantly lower (P < 0.05) than for men within the same event Fig. 2 Relationship between TB pen and F 60 /CSA In all the subjects, signi cant negative correlations were found between the estimated CSA and both F 60 /CSA (r ˆ )0.426, P < 0.05) and F 180 /CSA (r ˆ )0.484, P < 0.05). On the other hand, although there was a tendency for both F 60 /CSA and F 180 /CSA to be negatively correlated with TB pen (r ˆ )0.336 for F 60 / CSA and r ˆ )0.251 for F 180 /CSA the relationship between F 60 /CSA and TB pen alone was statistically signi cant (P < 0.05, Fig. 2). Discussion Among the athlete groups tested in this study, the soccer players of both sexes, who generally had less upper-body strength developing activities during actual competitions and/or training regiments, showed and lowest values in both TB mt and TB pen. The average TB pen of 15 in the male soccer players was almost equal to that observed previously in untrained males (Kawakami et al. 1993). On the other hand, although the judo athletes of both sexes showed larger TB pen than the other two athlete groups, their average values, 27 for the men and 28 for the women, were smaller compared to those reported previously for higher trained body builders (33 ; Kawakami et al. 1993). These results would suggested that these Olympic athletes had not enlarged their muscles as has been observed in exceptionally strength-trained persons. Positive correlations were found between TB mt /l ua and TB pen in both sexes. This result agrees with previous ndings by Kawakami et al. (1993), in which muscle thickness and bre pennation angles of the triceps brachii muscle were measured for untrained persons and body builders using the same procedure as in this study. On the other hand, previous ndings on the relationship between the muscle size and bre angulation of knee extensor muscles are still controversial, although similar

5 113 methods to those in the present study have been used to determine these two parameters. For example, Henriksson-Larse n et al. (1992) have found no correlation between muscle bre CSA and pennation angles and have concluded that bre angulation was independent of bre size. In contrast, Rutherford and Jones (1992) have reported that there was a signi cant correlation between quadriceps femoris muscle CSA and mean bre pennation angles for the vastus lateralis and intermedius muscles. Two possibilities might explain the discrepancy between the results on the existence of a statistical signi cance in the relationship between muscle size and bre angulation. Firstly, the contribution of size to bre angulation may vary from muscle to muscle. This assumption would be supported by previous reports that resistance training has resulted in similar relative gains in both muscle thickness and bre angles of the triceps brachii muscle (Kawakami et al. 1995), but no change in bre pennation angles of knee extensor muscles in spite of a signi cant increase in muscle CSA (Rutherford and Jones 1992). Secondly, the selection of subjects may have led to a discrepancy between results on the relationship between muscle size and bre angulation. The present and the studies of Kawakami et al. (1993, 1995) have tested subjects with di erent backgrounds, of physical activity and a wide range of variation in muscle size. On the other hand, Henriksson-Larsen et al. (1992) have tested women with average physical characteristics. The di erence in the range of variation in muscle size might therefore have in uenced the existence of a statistical signi cance in the relationship between muscle size and bre angulation. The gymnasts and judo athletes showed the same sex di erences in TB pen to those observed in TB mt /l ua. Moreover, the range of variation in TB pen was similar in both sexes. Therefore, it seems that the magnitude of TB pen might be similar in male and female athletes on average, if allowance is made for the di erence in the muscle size. In the comparison within the same event, however, the soccer players exceptionally showed a discrepancy between the sex di erences in these parameters, TB mt /l ua being signi cantly higher in the men than in the women, but TB pen being similar in both sexes. To interpret the discrepancy, rstly, the fact that the measured TB pen is in uenced by the curvature of the bre should be considered. This might be a factor responsible for the inter-individual variation existing in the relationship between TB mt /l ua and TB pen as shown in Fig. 1. On the other hand, in a simple model, the muscle layer thickness is dependent on bre pennation angle and fascicle length. The previous nding by Henriksson- Larse n et al. (1992), has shown that the bre angulation of the knee extensor muscles was negatively correlated to the length of the fascicle. Taking this nding into account, there is a possibility that the observed di erences in TB mt and TB mt /l ua between the male and female soccer players might have been attributed to the di erence in the fascicle length of the triceps brachii muscle. In any case, since no data are available on the existence of a di erence in fascicle length between the sexes, further investigation will be required in this respect. In the present study, CSA estimated from TB mt was used to analyse, the relationship between CSA and force output, and to compare force per unit of CSA among subjects. Therefore, we cannot directly compare the CSA with those obtained previously from measurements using imaging techniques such as computed tomography, MRI and ultrasound. However, in our earlier study on the relationship between the muscle layer thickness and measured CSA of the triceps brachii muscle (Fukunaga et al. 1991), a high correlation (n ˆ 59, r ˆ 0.822) has been found between the two parameters. Moreover, the observed correlation coe cients between the estimated CSA and forces, r ˆ for F 60 and r ˆ for F 180, are comparable to those reported in previous studies on the relationship between muscle CSA and isokinetic forces (Alway et al. 1990; Sale et al. 1987; r ˆ 0.58±0.86). In these studies imaging techniques have been used to determine muscle CSA for subject groups which consisted of both sexes and clearly belonged to separate populations with respect to the extent of muscle CSA and ability to generate force. Therefore, it is reasonable to assume that the CSA proposed in this study would have had almost the same variability as the measured CSA by imaging techniques for analysing the relationship between CSA and force output and performing comparisons on force per unit of CSA among subjects. No signi cant sex di erences were found in both F 60 / CSA and F 180 /CSA regardless of event. Some researchers have investigated the di erence in upper-body strength per unit of muscle or fat-free CSA between the sexes having similar physical activity backgrounds. Among these studies, Castro et al. (1995) and Miller et al. (1993) have found no signi cant di erence in force per unit of CSA as in the present study, but Bishop et al. (1987) have reported that male athletes showed signi cantly higher upper-body strength than female, even when adjustments were made for the di erence in CSA. The discrepancy might be referred to the di erence in the procedure for force measurements between studies. In the present and the above-cited two studies, force output during a single joint movement such as elbow exion or extension was determined using dynamometers. On the other hand, Bishop et al. (1987) have used, as the upper body force score, the largest mass that could be lifted in isotonic exercises such as bench press and arm curl tasks. The use of these multi-joint movements for assessing the upper-body force has the possibility of involving the activities of muscle groups where appropriate measurement of CSA has not been made, and so it might have led to a disagreement with other studies on the existence of signi cant sex di erence in the force per unit of CSA. Moreover, it is noteworthy that, in comparisons within the same event, the gymnasts alone showed a

6 114 signi cant sex di erence in TB pen among the three groups of athletes, but they did not in both F 60 /CSA and F 180 /CSA, as was the case for the other two events as well. This result would indicate the possibility that the di erence in the ability to generate force of the triceps brachii muscle might in the main be attributed to the sex di erence in the muscle size rather than to the architectural characteristics. Some researchers have suggested a negative in uence of the pennation on the ability to generate force in the hypertrophied muscles because of the signi cant negative correlation that has been observed between CSA and force per unit of CSA (Maughan et al. 1983, 1984; Sale et al. 1987). In fact, Kawakami et al. (1993, 1995) have found that the bre pennation angles of the triceps brachii muscle were greater in the hypertrophied condition than in the normal one, and that it increased after a period of resistance training, with a reduction of force per unit of muscle CSA. Signi cant negative correlations were also found between the estimated CSA and both F 60 /CSA and F 180 / CSA in the present study. Moreover, there was a tendency for F 60 /CSA and F 180 /CSA to be reduced with increasing TB pen. This might explain in part why the soccer players of both sexes, whose average values for both TB pen and TB mt /l ua were the smallest, showed the highest average values for F 60 /CSA and F 180 /CSA among subject groups tested (Table 2). However, although the relationship between TB pen and F 60 /CSA was statistically signi cant, the correlation coe cient between the two parameters was low. Moreover, as shown in Fig. 2, the distributions of F 60 /CSA on a TB pen were almost the same in the two sexes. Therefore it seems that, at least in the Olympic athletes tested in this study, the bre angulation of the triceps brachii muscle had less in uence on the ability to generate force than muscle size, and if some negative in uences existed, the magnitude of the sex di erence in the bre angulation might not have been so large as to induce a signi cant di erence in force per unit CSA as a result of comparison within the same event. In summary, at least in the Olympic athletes tested in this study, the magnitude of the bre angulation of the triceps brachii muscle was not as large as that observed in exceptionally strength-trained persons, and it might be more or less the same in both sexes if allowances are made for di erences in the muscle thickness. Moreover, the in uence of bre angulation on the ability to generate force might be less compared to that of muscle size, and the sex di erence in the ability to generate force could in the main be due to the di erence in muscle size rather than to the di erence in architectural characteristics. Acknowledgements This study was carried out in collaboration with the Japanese Olympic Committee and Sports Science Laboratory of the Japanese Amateur Sports Association. References Alexander R McN, Vernon A (1975) The dimensions of knee and ankle muscles and the forces they exert. J Hum Mov Stud 1:115±123 Alway SE, Stray-Gundersen J, Grumbt WH, Gonyea WJ (1990) Muscle cross-sectional area and torque in resistance-trained subjects. Eur J Appl Physiol 60:86±90 Bishop P, Cureton K, Collins M (1987) Sex di erence in muscular strength in equally-trained men and women. Ergonomics 30:675±687 Castro MJ, McCann DJ, Sha rath JD, Adams WC (1995) Peak torque per unit cross-sectional area di ers between strengthtrained and untrained young adults. Med Sci Sports Exerc 27: 397±403 Fukunaga T, Abe T, Ikegawa S, Ozawa H, Goda K, Ishida Y, Asami T (1991) Relationship between limb composition and thickness of subcutaneous fat and skeletal muscle. Rep Res Cent Phys Educ 19:1±6 Henriksson-Larse n K, Wretling ML, Lorentzon R, OÈ berg L (1992) Do muscle bre size and bre angulation correlate in pennated human muscles? Eur J Appl Physiol 64:68±72 Kawakami Y, Abe T, Fukunaga T (1993) Muscle- ber pennation angles are greater in hypertrophied than in normal muscles. J Appl Physiol 74:2740±2744 Kawakami Y, Abe T, Kuno S, Fukunaga T (1995) Traininginduced changes in muscle architecture and speci c tension. Eur J Appl Physiol 72:37±43 Maughan RJ, Watson JS, Weir J (1983) Strength and crosssectional area of human skeletal muscle. J Physiol 338:37±49 Maughan RJ, Watson JS, Weir J (1984) Muscle strength and crosssectional area in man: a comparison of strength-trained and untrained subjects. Br J Sports Med 18:149±157 Miller AED, MacDougall JD, Tarnopolsky MA, Sale DG (1993) Gender di erences in strength and muscle ber characteristics. Eur J Appl Physiol 66:254±262 Narici MV, Landoni L, Minetti AE (1992) Assessment of human knee extensor muscle stress from in vivo physiological crosssectional area and strength measurements. Eur J Appl Physiol 65:438±444 Rutherford OM, Jones DA (1992) Measurement of bre pennation using ultrasound in the human quadriceps in vivo. Eur J Appl Physiol 65:433±437 Sale DG, MacDougall JD, Alway SE, Sutton JR (1987) Voluntary strength and muscle characteristics in untrained men and women and male bodybuilders. J Appl Physiol 62:1786±1793