Although it is commonly accepted that muscle size determines the maximal isometric strength that a human skeletal muscle can produce in healthy adults regardless of the sex, it is unclear whether this relationship changes during growth and maturation. Results reported in pediatric populations are highly controversial. The ratio of isometric strength to maximal muscle cross-sectional area (MCSA) has been reported as unchanged throughout adolescence and early adulthood in elbow flexors and triceps surae muscles (8,16), whereas a significant increase has been reported during growth for isometric (13,19,36) and dynamic exercises (18,36). The ability to produce strength with respect to MCSA was lower in children as compared with young adults for knee extensor (19) and elbow flexor muscles (13). In other words, for a given MCSA, force would be higher in men as compared with boys, thereby suggesting that anthropometric factors on their own could not fully account for force changes during growth and maturation, whatever the exercise modality. A series of additional variables related to muscle metabolism, such as muscle enzyme activity (9,18) and muscle fiber-type composition (5), could explain part of these changes. Another hypothesis regarding the difference in muscle to force volume ratio between children and adults is related to the neuromuscular system (13,31,32) and suggests that recruitment and firing rate of motor units would not be optimal in children, thereby accounting for a decrease in the force to muscle size ratio. Overall, it still remains unclear whether force is similarly determined by muscle size, whatever the age.
Part of these controversies could be attributed to methodological discrepancies among studies. More particularly, precise quantification of muscle volume or MCSA from anthropometric measurements might not be as accurate in children as it is in adults. Although magnetic resonance imaging (MRI) is considered the gold standard, very few studies have reported local muscle volume quantification using MRI in children, and morphometric analyses have been mainly performed using ultrasound (19) or anthropometry (13,36). On the basis of skinfold caliper measurements and a set of assumptions, muscle volume can be estimated from anthropometric measurements. This method has been shown to be valid for lean, healthy adults (22,39), but potential biases in pediatric populations have never been clearly estimated regarding limb muscle volume quantification.
The main purpose of this study was to determine whether the relationship between muscle size and maximum isometric strength changed during growth and maturation. In addition to that, we quantified the potential measurements bias introduced by anthropometric estimations of local muscle volume considering MRI as the gold-standard technique. We also determined the differences when muscle size was estimated from volume and MCSA measurements in a pediatric population.
Forty-two healthy subjects were included in the present study. Written informed consent was obtained from all the subjects and from the parents for each minor volunteer. Pubertal stage was determined according to pubic hair and gonadal development, as described by Tanner et al., using a self-reporting method (38). From a practical point of view, each subject determined their pubic hair and gonadal pubertal stages according to drawings representing the five different Tanner stages. On that basis, subjects were classified as prepubertal (Tanner stage ≤ 2) or pubertal (Tanner stage > 2). All the subjects were physically active, but none of them were involved in any regular training program. The subjects were considered as normally thin according to the norm defined by the International Obesity Task Force. The protocol was approved by the Ethics Timone Hospital Committee (Marseille, France).
Maximum Voluntary Isometric Force Measurement.
Maximal isometric strength of the dominant forearm (Fmax) was measured using an ergometer designed and built in our laboratory. Each subject sat on a chair and held a handle bar at the distal phalange joint level while the dominant forearm was positioned horizontally. After a familiarization period, each subject was encouraged to perform a maximal handgrip contraction for a 3-s period. This measurement was repeated three times with at least a 60-s resting period between each measurement. Force output was measured using a calibrated force transducer (0- to 2000-N force range, ZF, Scaime, France) connected to the handle bar and transmitted to a PC using an analog/digital converter, thereby providing visual feedback to the subject. The maximum force was determined as the mean of three reproducible measurements. The variability of each trials was 3.0 ± 2.2%, whatever the age.
The forearm lean (muscle + bone) volume (VL) was determined from circumferences and skinfold thickness measurements according to the Jones and Pearson method (17) and using the following formula:
where VL1 and VL2 refer to the volume of the wrist-to-midforearm and midforearm-to-elbow segments respectively. A refers to the area at the three levels: wrist, midforearm, and elbow.
The forearm length and circumferences were measured to the nearest 1 mm at three levels (wrist, midforearm, and elbow) using a flexible standard measuring tape. Skin thickness was measured on the anterior and posterior parts of the dominant midforearm using a Harpenden skinfold caliper (British Indicators). The corresponding average was subtracted from the circumference of the midforearm to determine the forearm's lean volume. The skin thickness at the elbow and wrist levels was considered negligible.
MRI investigations were performed at 1.5 T on a whole-body Siemens-Vision Plus Imaging system (Siemens, Germany). The dominant arm was placed in the middle of an "extremity coil" (central diameter = 25 cm). Subjects were scanned in a prone position with the arm fully extended and relaxed. For each subject, the midforearm was ink-marked, and the corresponding mark was aligned with the crosshairs of the MRI system so that identical positioning was ensured in the magnet bore for each subject. Transverse T1-weighted images (9-13 slices, depending on the forearm length from wrist to elbow) were recorded with the following parameters (TR = 490 ms, TE = 12 ms, 200-mm field of view, 512 × 512 acquisition matrix, slice thickness = 5 mm, 10-mm gap, and a total duration of 122 s).
The cross-sectional area of each slice was determined using an IDL (Interactive Data Language, RSI) home-written routine (23). Briefly, on the basis of a signal-intensity threshold, fat, muscle, and bone signals were clearly separated and quantified. Muscle volume (VM) was calculated by summing the areas of all the slices, taking into account the slice thickness and the interslice space.
MCSA was defined as the highest area measured among different slices.
The ratio between muscle volume calculated with MRI (VM) and from anthropometric (VL) measurements was used as a comparative index between both techniques. The ratio between VL and VM was expressed as a percentage.
Statistical analyses were performed using JMP software 5.1.2 (SAS institute Inc.). All results were expressed as means and standard deviations (mean ± SD). Each variable distribution has been tested, and appropriate one-way ANOVA was used to compare the three groups. If the first test provided a significant difference, then each group was compared with the other two using post hoc Tukey-Kramer tests. Relationships between variables were analyzed using linear regressions, and the corresponding strength was assessed using Pearson's correlation coefficient. The limit for statistical significance was set at P < 0.05.
The physical characteristics of each group are shown in Table 1. Most of the variables were significantly different among the three groups.
Typical MR images of the forearm are displayed in Figure 1. Forearm muscle volume as determined from MRI (VM) or anthropometric (VL) measurements was significantly higher in adults than in children and adolescents (Table 2). Also, prepubertal boys had a significantly lower muscle volume as compared with adolescents. Regarding MCSA values, men had significantly larger values compared with adolescents and children, whereas adults and adolescents had similar values (Table 2).
Similar to muscle volume, Fmax significantly increased with respect to age from 174.9 ± 31.8 N in children to 456.2 ± 49.2 N in men. Men were significantly stronger than children and adolescents, while adolescents were stronger than children (Table 3). The ratio of Fmax to VM was identical in the three groups (Fig. 2A). On the contrary, when scaled to MCSA (Fig. 2C) or VL (Fig. 3A), maximum force was larger in adults as compared with both children and adolescents. Additionally, Fmax was positively related to muscle volume measured either with MRI (y = 0.83x, r2 = 0.90, P < 0.001) (Fig. 2B) or using anthropometry (y = 0.81x − 77.0, r2 = 0.85, P < 0.001) (Fig. 3B). A strong relationship between Fmax and MCSA (y = 11.9x − 41.1, r2 = 0.87, P < 0.001) was found (Fig. 2D).
MRI measurements of muscle volume were strongly correlated to the values estimated by anthropometry (y = 0.99x + 107.3, r2 = 0.90, P < 0.001) (Fig. 4), while anthropometric measurements (VL) were systematically overestimated with respect to MRI values (VM) (Table 2). This overestimation was significantly larger in children and adolescents as compared with adults (43.1, 38.5, and 20.5%, respectively) (Table 2).
The primary aim of this investigation was to determine whether the relationship between muscle volume and maximum isometric strength did change during growth and maturation.
The maximal isometric strength of forearm flexor muscles (Fmax) and muscle size significantly increased from childhood to adulthood. The present Fmax values measured during a maximal handgrip force test (174.9 ± 31.8, 266.9 ± 96.4, and 456.2 ± 49.2 in children, adolescents, and adults, respectively) are consistent with previous results reported in similar populations (175 ± 43 N in 10- to 11-yr-old boys to 498 ± 76 N in adults) (28,33). To our knowledge, no detailed pediatric studies on forearm muscle volume have been reported so far, but our measurements in adults are in agreement with prior data (15,26). In addition, our forearm MCSA values are comparable with earlier measurements in 10-yr-old boys to 40-yr-old men (28).
Taking into account the three different groups, we found a strong positive linear relationship between Fmax and muscle size estimated from VL, VM, or MCSA measurements. Similar correlations have been already reported in adults using different methods (3,24) and more scarcely in a pediatric population (22) for various muscles. These results strongly suggest that the maximal voluntary isometric force that can be exerted by a human muscle is mainly proportional to its size, whatever the age. They also refute previous reports suggesting, on the basis of measurements of force per unit of anatomic CSA, that an increase in this ratio from childhood to adulthood (13,36) would be explained by changes in fiber-type composition or maturation of the central nervous system conditioning changes in motor unit recruitment during growth.
This disagreement clearly points out that the method used and/or the index chosen to estimate muscle size can largely influence the corresponding strength-muscle size relationship and the accompanying explanations regarding changes during growth.
It is noteworthy that in the same group of subjects, the ratios of Fmax to muscle volume and to MCSA evolved differently from childhood to adulthood. In agreement with previous findings, Fmax to VL or MCSA ratio was significantly lower in children and adolescents as compared with adults (13,19,36). However, when muscle volume was assessed using MRI (VM), we found no difference among the three groups; MRI is currently accepted as the gold-standard technique for the morphometric studies. This noninvasive technique offers direct and accurate access to the different compartments of the body or limb composition, especially for skeletal MSCA (6) and volume (25). Although both MCSA and VM have been measured using MRI, we found significantly different results according to the index chosen. In the present study, we only measured the anatomic and not the physiological MCSA. This distinction is actually important in pennated muscles such as forearm muscles, for which the force produced depends on the fibers' length and on their pennation angle (3,40). These latter parameters are not taken into account with anatomic MCSA. Furthermore, in pennated muscles, the difference between anatomic and physiological MCSA will become larger as muscle length (i.e., muscle volume) increases (2), as is the case during growth. On that basis, the Fmax to anatomic MCSA ratio might have been systematically overestimated in adults due to their larger muscle volume, thereby introducing bias in the force to MCSA relationship. The increase of force per unit of MCSA reported in the present study during growth would come from the inaccuracy of muscle size measurements and might not have any physiological or neural origins as previously suggested (13,36). Physiological MCSA obviously has to be preferred to anatomic MCSA in order to characterize the relationship with the strength ability. However, a simple and accurate determination of pennation angle in vivo is still missing. Muscle volume measurements should be preferred to anatomic MCSA to properly investigate relationships between strength and muscle size during growth and maturation more particularly when the muscle studied has a pennated architecture.
Additionally, considering that VM was the index most strongly correlated with Fmax, and that the line representing the relationship between Fmax and VM crossed the y-axis at zero, we suggest that VM would be the most appropriate muscle size index of force. On the contrary, Bamman et al. reported that in exercising plantar flexor muscles, both anatomic and physiological CSA measured using MRI in women were the indices best correlated (i.e., with an intercept at zero) with maximal isometric force (3). In contrast to what we performed in the present study, they did not include the whole muscle area in their MRI measurements, but only the triceps surae, the activated muscle during plantar flexions. One might be concerned about the inclusion of nonactivated muscles in the measurements of muscle size. However, it seems reasonable to think that the relative bias introduced for each subject was similar, such that it cannot impact the final result. A localized hypertrophy could have modified our results, but this was not the case, given that the subjects were not involved in any localized training of forearm muscles.
In addition to the index chosen to characterize muscle size, the technique used can also greatly influence the strength-muscle size relationship. In our group of subjects, when muscle size was assessed using anthropometry, we found a significant increase of strength per unit of muscle volume in adults compared with children. This ratio was similar among the three groups when muscle volume was measured using MRI, but this difference is likely to be explained by the well-recognized approximations linked to the anthropometric truncated cone model (17). It is indeed well documented that anthropometry introduced systematic bias in the muscle size quantification. Overestimations of local muscle CSA or volume have been reported in various muscles of adults when anthropometry was compared with reference methods such as computed tomography (4,7,10,29,35) and MRI (11,22,39). Despite a systematic overestimation, absolute values were strongly correlated. More specifically, for the thigh muscles, a strong positive linear relationship has been reported between the two sets of muscle volume values assessed by anthropometry and MRI, with a slope of 1 and an intercept at 1052 cm3 (39). To our knowledge, no study has ever quantified the bias of anthropometric measurements in pre- and pubertal populations, and, more importantly, it is still unknown whether the size of this potential bias is independent of age.
In agreement with the aforementioned studies, we found a strong correlation between forearm muscle volume values assessed by anthropometry and MRI. The present correlation coefficient (r2 = 0.90, P < 0.001) was comparable with previous findings in the lower limb (11,22,39). As expected, muscle volume estimated by anthropometry was systematically overestimated compared with MRI measurements. In absolute terms, the systematic bias was similar in children and adults. Given the lower muscle volume in pre- and pubertal boys, muscle volume was significantly more overestimated in children (43.1 ± 15.2%) and adolescents (38.5 ± 18.8%) compared with adults (20.0 ± 10.5%). In other words, for a given muscle volume calculated from anthropometric measurements, force level was more underestimated in children and in adolescents than in adults. This bias would explain the erroneous increase of force per unit of muscle volume reported from childhood to adulthood when muscle volume was calculated from anthropometric measurements. This would also account for the negative intercept observed on the Fmax-VL relationship.
Several explanations could be put forth in order to explain the bias related to anthropometric muscle volume measurements. One has to keep in mind that anthropometry provides an estimation of the lean limb volume including not only muscle but also bone so that muscle CSA and volume are systematically overestimated. Using MRI, it is possible to distinguish muscle and bone. We found that the bone to muscle volume ratio was indeed significantly larger in children than in adults (0.14 ± 0.007 vs 0.10 ± 0.007, P < 0.05), thereby clearly indicating that bone inclusion in anthropometric measurements introduced a bias in the comparisons between children and adults, and that anthropometric estimation of muscle volume is more accurate in adults than in children. The bone volumes measured using MRI were 29.07 ± 5.42, 37.20 ± 9.66, and 52.87 ± 13.72 cm3 in children, adolescents, and adults, respectively, while the absolute bias introduced by anthropometry amounts to 104 ± 36 cm3. These results clearly indicate that, although a contributing factor, bone inclusion is not the only parameter accounting for the systematic bias related to anthropometric measurements. The comparison of lean volume (including bone) measured using MRI and anthropometry still indicates a larger overestimation in children (25.96 ± 12.32%) and in adolescents (24.27 ± 15.20%) than in adults (9.70 ± 9.11%). It is noteworthy that bone inclusion in the comparative analysis significantly reduced but did not erase the corresponding overestimation, illustrating that other factors should influence the accuracy of anthropometric measurements.
In addition to bone, intramuscular fat can be another potential source of error. It has been shown that aging increases the relative content of nonmuscle tissue in the muscle CSA or volume (21,34). In the present study, we did not measure any difference in the level of intramuscular adiposity among the three groups.
Assumptions related to anthropometry calculations and technical difficulties are also potential confounding factors in muscle volume measurements. On the one hand, the truncated cone model oversimplifies the limb anatomy, assuming that the limb is circular and that subcutaneous fat is homogeneously distributed around the limb. MR images clearly show that is not necessarily the case for the forearm, as it has been already reported for the arm, the thigh, and the calf (11,14,22,29). The accuracy of anthropometric measurements depends on the agreement between the model and the actual form, which are known to change with nutritional status or diseases (10,29,39) and training (20,27). Because morphology strongly changes during growth, these assumptions might be more adapted for the mature rather than the immature form of the limb.
On the other hand, measurement of subcutaneous adipose tissue with a caliper is technically demanding (4,14). For instance, it has been reported that the disparity between anthropometry and MRI is likely to increase with increasing adiposity (4,10,39). However, in the present study, all the subjects have a normal BMI and were normally active, such that adiposity might not be considered as a confounding factor.
Our results clearly showed that anthropometric measurements cannot be reliably used in order to investigate the strength-muscle size relationship during growth and maturation, and that MRI should be preferred. Other techniques such as ultrasound have been used in order to measure muscle volume (19). Given that these methods are based on assumptions similar to anthropometry, one might reasonably assume that the corresponding results are equally biased.
The most commonly evoked hypothesis accounting for the lesser ability of a children to produce strength compared with adults is related to an insufficient recruitment of motor units and also under isometric (13,31) or dynamic conditions (19,36); this would be related to an immature neuromuscular system. Paasuke and coworkers reported a higher twitch tetanic force related to maximal voluntary force in prepubertal boys than in postpubertal boys and men and suggested an increased motor unit activation under maximal voluntary contraction in children (31). As is commonly the case in this type of study, one can wonder how reliable the MVC measurement was in each group of subjects included in the present study. Each subject was verbally encouraged in order to produce the maximum force, which was averaged over three reproducible measurements. The within-subject variation for MVC was 3.0 ± 2.2% for the entire population and was similar among the three groups, clearly indicating that motivation cannot be invoked as a confounding factor. In addition, the zero-intercept of the Fmax-VM relationship is an additional quality factor of our measurements.
Although our results strongly suggest that muscle strength is mainly determined by muscle volume regardless of age, we cannot refute the involvement of other factors. Improvement of maximal voluntary isometric and dynamic muscle strength has been reported in pre- and pubertal boys after resistive training (30,32) and in adults (1), whereas muscle size was unchanged. An increased neuromuscular activation has been suggested to explain this improvement, but its magnitude was smaller than the corresponding strength increase (32), thereby suggesting that other factors should explain this phenomenon. A reduced activation of the muscle sensory receptors (12) and a change in motor skill coordination (37) have also been suggested as accounting for the training-induced strength gain not associated with hypertrophy. The similar training adaptation pattern reported before and after puberty (34) suggests that the neuromuscular system, including the peripheral (motor unit activation) and the central (motor ability) components, did not change during maturation. It is important to keep in mind that these changes-the strength improvement with no corresponding hypertrophy-resulted from training, and that no clear evidence is available regarding a difference occurring with growth and maturation.
Overall, our results clearly and strongly suggest that in subjects not involved in regular training, the ability of a given muscle volume to produce force does not change during growth. Previous results reporting changes likely suffer from inaccurate measurements of muscle volume, especially in children.
We thank all the participants for their effort and patience during data collection.
1. Aagaard P. Training-induced changes in neural function. Exerc Sport Sci Rev
2. Aagaard P, Andersen JL, Dyhre-Poulsen P, et al. A mechanism for increased contractile strength of human pennate muscle in response to strength training: changes in muscle architecture. J Physiol
3. Bamman MM, Newcomer BR, Larson-Meyer DE, Weinsier RL, Hunter GR. Evaluation of the strength-size relationship in vivo using various muscle size indices. Med Sci Sports Exerc
4. Baumgartner RN, Rhyne RL, Troup C, Wayne S, Garry PJ. Appendicular skeletal muscle areas assessed by magnetic resonance imaging in older persons. J Gerontol
5. Bell RD, MacDougall JD, Billeter R, Howald H. Muscle fiber types and morphometric analysis of skeletal msucle in six-year-old children. Med Sci Sports Exerc
6. Beneke R, Neuerburg J, Bohndorf K. Muscle cross-section measurement by magnetic resonance imaging. Eur J Appl Physiol Occup Physiol
7. Buckley DC, Kudsk KA, Rose BS, Fatzinger P, Koetting CA, Schlatter M. Anthropometric and computerized tomographic measurements of lower extremity lean body mass. J Am Diet Assoc
8. Davies CT, White MJ, Young K. Muscle function in children. Eur J Appl Physiol Occup Physiol
9. Eriksson BO, Gollnick PD, Saltin B. Muscle metabolism and enzyme activities after training in boys 11-13 years old. Acta Physiol Scand
10. Forbes GB, Brown MR, Griffiths HJ. Arm muscle plus bone area: anthropometry and CAT scan compared. Am J Clin Nutr
11. Fuller NJ, Hardingham CR, Graves M, et al. Predicting composition of leg sections with anthropometry and bioelectrical impedance analysis, using magnetic resonance imaging as reference. Clin Sci (Lond)
12. Gabriel DA, Basford JR, An K. Training-related changes in the maximal rate of torque development and EMG activity. J Electromyogr Kinesiol
13. Halin R, Germain P, Bercier S, Kapitaniak B, Buttelli O. Neuromuscular response of young boys versus men during sustained maximal contraction. Med Sci Sports Exerc
14. Heymsfield SB, McManus C, Smith J, Stevens V, Nixon DW. Anthropometric measurement of muscle mass: revised equations for calculating bone-free arm muscle area. Am J Clin Nutr
15. Holzbaur KR, Murray WM, Gold GE, Delp SL. Upper limb muscle volumes in adult subjects. J Biomech
16. Ikai M, Fukunaga T. Calculation of muscle strength per unit cross-sectional area of human muscle by means of ultrasonic measurement. Int Z Angew Physiol
17. Jones PR, Pearson J. Anthropometric determination of leg fat and muscle plus bone volumes in young male and female adults. J Physiol
18. Kaczor JJ, Ziolkowski W, Popinigis J, Tarnopolsky MA. Anaerobic and aerobic enzyme activities in human skeletal muscle from children and adults. Pediatr Res
19. Kanehisa H, Yata H, Ikegawa S, Fukunaga T. A cross-sectional study of the size and strength of the lower leg muscles during growth. Eur J Appl Physiol Occup Physiol
20. Katch FI, Hortobagyi T. Validity of surface anthropometry to estimate upper-arm muscularity, including changes with body mass loss. Am J Clin Nutr
21. Kent-Braun JA, Ng AV, Young K. Skeletal muscle contractile and noncontractile components in young and older women and men. J Appl Physiol
22. Knapik JJ, Staab JS, Harman EA. Validity of an anthropometric estimate of thigh muscle cross-sectional area. Med Sci Sports Exerc
23. Mattei JP, Fur YL, Cuge N, Guis S, Cozzone PJ, Bendahan D. Segmentation of fascias, fat and muscle from magnetic resonance images in humans: the DISPIMAG software. Magma
24. Maughan RJ, Watson JS, Weir J. Strength and cross-sectional area of human skeletal muscle. J Physiol
25. Mitsiopoulos N, Baumgartner RN, Heymsfield SB, Lyons W, Gallagher D, Ross R. Cadaver validation of skeletal muscle measurement by magnetic resonance imaging and computerized tomography. J Appl Physiol
26. Miyatani M, Kanehisa H, Fukunaga T. Validity of bioelectrical impedance and ultrasonographic methods for estimating the muscle volume of the upper arm. Eur J Appl Physiol
27. Narici MV, Hoppeler H, Kayser B, et al. Human quadriceps cross-sectional area, torque and neural activation during 6 months strength training. Acta Physiol Scand
28. Neu CM, Rauch F, Rittweger J, Manz F, Schoenau E. Influence of puberty on muscle development at the forearm. Am J Physiol Endocrinol Metab
29. Overend TJ, Cunningham DA, Paterson DH, Lefcoe MS. Anthropometric and computed tomographic assessment of the thigh in young and old men. Can J Appl Physiol
30. Ozmun JC, Mikesky AE, Surburg PR. Neuromuscular adaptations following prepubescent strength training. Med Sci Sports Exerc
31. Paasuke M, Ereline J, Gapeyeva H. Twitch contraction properties of plantar flexor muscles in pre- and post-pubertal boys and men. Eur J Appl Physiol
32. Ramsay JA, Blimkie CJ, Smith K, Garner S, MacDougall JD, Sale DG. Strength training effects in prepubescent boys. Med Sci Sports Exerc
33. Rauch F, Neu CM, Wassmer G, et al. Muscle analysis by measurement of maximal isometric grip force: new reference data and clinical applications in pediatrics. Pediatr Res
34. Rice CL, Cunningham DA, Paterson DH, Lefcoe MS. Arm and leg composition determined by computed tomography in young and elderly men. Clin Physiol
35. Rice CL, Cunningham DA, Paterson DH, Lefcoe MS. A comparison of anthropometry with computed tomography in limbs of young and aged men. J Gerontol
36. Saavedra C, Lagasse P, Bouchard C, Simoneau JA. Maximal anaerobic performance of the knee extensor muscles during growth. Med Sci Sports Exerc
37. Shima N, Ishida K, Katayama K, Morotome Y, Sato Y, Miyamura M. Cross education of muscular strength during unilateral resistance training and detraining. Eur J Appl Physiol
38. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child
39. Tothill P, Stewart AD. Estimation of thigh muscle and adipose tissue volume using magnetic resonance imaging and anthropometry. J Sports Sci
40. Wickiewicz TL, Roy RR, Powell PL, Edgerton VR. Muscle architecture of the human lower limb. Clin Orthop Relat Res