Ultrasound (US) technique was applied to measure the thickness, crosssectional area (CSA), and internal structure of the quadriceps muscle in 70to 80-year-old endurance and power athletes and untrained men. Isometric knee extension strength was higher in the power athletes than in the other groups. The mean values for thickness and CSA did not differ between the athletes and the untrained men. The intensity of the intramuscular echo was, however, lower and the echo reflected from the connective tissue septa and bone was higher in the athletes than in the untrained men. Both the CSA and internal structure of the muscle group correlated significantly with muscle strength and number of training kilometers per year. The results suggest that long-term training maintains the muscle architecture and counteracts the age-related replacement of contractile tissue by other tissues such as fat. Key words: aging exercise quadriceps ultrasound imaging MUSCLE 81 NERVE 1 4 5 2 7 6 3 3 1991
ULTRASOUND IMAGING OF THE QUADRICEPS MUSCLE IN ELDERLY ATHLETES AND UNTRAINED MEN SARIANNA SIPILA, MSc, and HARRI SUOMINEN, PhD
W i t h increasing age, human muscles undergo both structural and functional changes, the most striking of which are reduction in muscle mass and strength."42"922 T h e decline of muscle strength is, to some extent, due to muscle atrophy, which is caused by a loss of muscle and by a reduction, especially in type 2, in fiber size.3~19322 A marked reduction in muscle strength accompanied by only a marginal reduction in muscle mass with aging may involve a replacement of muscle tissue by connective tissue. 1,18323 Borkan et a16 also found an age-related infiltration of fat into the lean tissue of the leg muscles. A reduction in physical activity is one reason for the impaired functional capacity of elderly people. Several studies have shown the positive effects of 2 to 6 months of physical training on eldAccording to Kovanen erly muscles.2~7~9~10~20~26~27 et all7 and Suominen et a1," lifelong physical
training counteracts some age-related changes in muscles, although endurance-type training also exerts some effects similar to age-associated changes, such as slowing of muscle fibers.17 Ultrasound (US) imaging has been used for noninvasive, quantitative characterization of muscle tissue. Muscle mass has been assessed from the ~s~8 1,16,31,32 .1 or from the thickness2Y330 of the muscle. US imaging has mostly been used in studies with young or middle-aged subjects, whereas relatively few authors have reported on US imaging of muscles in elderly T h e purpose of this study was to determine the thickness, cross-sectional area, and internal structure of the quadriceps muscle using two different US imaging techniques, and to relate these properties to muscle strength and amount of training in elderly athletes and untrained men. MATERIALS AND METHODS
From the Department of Health Sciences, Unlversity of Jyvaskyla, Jyvaskyla, Finland Acknowledgments: Supported by grants from the State Council for Research in Sport and Physical Education of the Ministry of Education, Finland. The authors thank Mrs. Merja Perhonen and Mr. Pertti Pykala for their skillful technical assistance. Address reprint requests to s. Sipila, Department of Health Sciences, University of Jyvaskyla Seminaarinkatu 15, SF-40100 Jyvaskyla, Finland. Accepted for publication May 8, 1990 CCC 0148- 639)(/91/060527-07 504.00 0 1991 John Wiley & Sons, Inc.
Ultrasound Imaging of Muscle
This study was part of a larger study on health and functional capacity among Finnish male endurance athletes (long-distance runners, orienteers, and cross-country skiers; n = 67) and power athletes (sprinters, jumpers, and throwers; n = 31) aged 70 to 90 years. Most had a lifelong training history and were still active in competitive sports. A random sample of 70- to 81-year-old men (n = 43) from the population register of the rural municipality of Jyvaskyla served as an
Subjects.
MUSCLE & NERVE
June 1991
527
untrained control group. From these groups, 14 endurance athletes, 7 power athletes, and 11 untrained men were randomly selected for the US measurements. T h e physical characteristics of the subjects are shown in Table 2. The athletes had been training regularly for 30 to 70 years, except for one endurance athlete who had been training for 15 years. When a weighted sum index was computed on the basis of the relative energy expenditure and efficiency from selfreported running (1.0), cross-country skiing (0.7), cycling (0.3), walking (0.6), and swimming (4.0), the annual training distance of the endurance athletes averaged 1839 km (range 785 to 3700 km). The power athletes did not usually log training kilometers. Two of them participated in endurance events and recorded their training kilometers. The indices for those two men were 520 and 1426. The power athletes had, however, participated in 3 to 7 training sessions (mean 4.6) per week for track and field, gymnastics, etc. Of the untrained subjects, only one man had participated in more intensive physical activity during the preceding year. His training kilometers yielded an index of 550. T h e other 5 men who reported their training kilometers during the preceding year all had indices below 100. A written informed consent was obtained from all subjects before the laboratory examination. In the clinical examination, no neurological problems were perceived in any of the subjects. Isometric Strength. Maximal isometric quadriceps muscle strength was measured in the right leg in a sitting position on a custom-made dynamometer chair which was a modification of that described by Heikkinen et al.'" T h e earlier apparatus was improved so that knee extension strength could, if required, be measured at different knee joint angles. T h e knee was set at an angle of 60" (full extension O0) and the ankle was attached above the malleolus to a belt-strain-gauge system. The subjects were not encouraged during the trials. The best performance of three trials was taken as the result.
Ultrasound Imaging. T h e US imaging was obtained from the right m. quadriceps by selecting, as the scanning site, the midpoint between the great trochanter and the lateral joint line of the knee. The mid-thigh was located by means of Hexible tape. During scanning, the subjects lay supine
528
Ultrasound Imaging of Muscle
with the leg extended and relaxed on the examination table (Fig. 1). T h e thickness of the muscle was assessed with a real-time scanner (Aloka SSD-280 LS) fitted with a 7.5 MHz linear array transducer. Both longitudinal and transverse scans were performed. The cross-sectional area (CSA) was assessed with a compound US scanner (Aloka SSU-190) fitted with a 5 MHz transducer. The outline of the quadriceps was traced visually using electronic calipers. Both the muscle thickness and CSA were measured by the scanner's own computer. T h e axial and lateral resolution of the former transducer was set at 0.4 mm and 0.9 mm ( - 3 dB), and that of the latter at 0.4 mm and 1.6 mm, respectively. Gain settings and near- and far-field scales were set equally for all subjects. T o avoid tissue compression, a generous amount of gel was used under the probe. The positioning and orientation of the probe was altered until the best bone echo was achieved. The probe was then assumed to be at right-angles to the femur. T h e compound US scans were photographed with a Polaroid camera and the real-time scans were printed on film paper by an Aloka Ultrasono SSZ-95 printer. From the US scans, four different characteristics were evaluated on a four-point scale by two blind observers (Table 1). T h e scales were developed from those reported by Forst et aL9 T o assess the reproducibility of the method, both observers evaluated the pictures twice. Every picture was thus evaluated four times. T h e sum of the four observations was taken as the result for each variable. T h e coefficients of variation for the measure-
FIGURE 1. Position of probe and leg during compound scanning.
MUSCLE & NERVE
June 1991
Table 1. Four-point scales for four different features of US scans and abbreviations used in tables. Discerning fasciae (real-time = RFa, compound = CFa): 1. Fasciae not discernible 2. Fasciae poorly discerned 3. Fasciae moderately discerned 4 Fasciae sharply discerned Intramuscular septa formation (real-time = RS, compound =
CS): 1. 2. 3. 4.
Table 3. Intra- and inter-observer correlation coefficients of analysis of internal muscle structure ( P < 0,001) Variable RFa RS RM RFe CFa
cs
No septa formation Poor septa formation Moderate septa formation Sharp septa formation
CM CFe
Intramuscular echo intensity (real-time = RM, compound = CM): 1 . Echo poor intramuscular structure 2. Small echogenicity 3. Moderately increased echogenicity 4. Strongly increased echogenicity Discerning femur (real-time = RFe, compound 1 . Femur not discernible 2. Femur poorly discerned 3. Femur moderately discerned 4. Femur sharply discerned
=
CFe):
ments repeated over consecutive days were 4.1% for thickness in the transverse scans, 4.8% in the longitudinal scans, and 4.3% for CSA. T h e correlation coefficients were 0.96, 0.95, and 0.99, respectively. The intra- and inter-observer correlation coefficients for the variables representing the internal muscle structure are shown in Table 3. Standard procedures were used to calculate means and standard deviations (SD). The statistical significance of the differences between the two groups of subjects were calculated by Student’s t test (two-tailed) for independent samples and between three groups of subjects by the analysis of variance. T h e differences between the athletes and the untrained men in the Statistical Methods.
Intra-observer
Inter-observer
0.932 0.879 0.836 0.977 0.899 0.853 0.916 0.844
0.830 0.863 0.811 0.939 0.895 0.838 0.713 0.770
internal muscle structure were also calculated by the Mann- Whitney U test. Because there were no differences in the statistical significance between Student’s t test and the Mann-Whitney test, only the results of the t test are reported. Pearson’s coefficient of correlation was used to express the correlations between duplicated measurements and between different variables. Spearman’s coefficient of correlation was used to calculate intraand inter-observer correlations in the evaluation of the internal muscle structure. RESULTS
There were no significant differences between the athletes and the untrained men in age, height, weight, length of the femur, or the content of subcutaneous fat in the middle thigh. Isometric knee extension strength was significantly higher in the power athletes than in the untrained men. No significant differences existed between the power and the endurance athletes, or between the endurance athletes and the untrained men in isometric muscle strength (Table 2). The mean values for thickness and CSA of the quadriceps in the endurance athletes and in the power athletes did not differ from those obtained in the untrained men (Table 4).
Table 2. Physical characteristics of elderly athletes and untrained men (mean Variable Age (years) Height (cm) Weight (kg) Length of femur (cm) Thickness of thigh subcutaneous tissue (mm) Isometric strength of quadriceps muscle (N)
Power athletes (n = 7) Endurance athletes (n
*
77.1 3.5 169.0 & 6.0 69.8 9.3 43.6 ? 3.1 4.3 1.0
* *
74.2 ? 3.0 171.0f 7.3 68.7 2 8.6 43.8 f 2.2 4.7 f 2.3
409* t 53
362t 2 76
=
* SD)
14) All athletes ( n = 21) Untrained men ( n
*
75.2 3.4 170.3 f 6.8 69.1 ? 8.2 43.7 2.5 4.6 1.9
* 377$ * 71
=
11)
73.4 f 2.4 167.6 ? 4.3 74.7 ? 11.9 43.1 ? 2.6 5.8 f 2.1 300
2
83
*P = 0.007, fP = 0.065, #P = 0.010 (difference between the athletes and the untrained men).
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
529
~~
Table 4. Quadriceps muscle thickness and CSA of elderly athletes and untrained men (mean k SD) Variable
Power athletes ( n
=
7)
Endurance athletes (n = 14)
Thickness (cm)
2.73
k
0.37
2.77
CSA (cm’)
52.5
k
10.8
52.7
? 2
The differences in the internal muscular structure between the athletes and the untrained men were more pronounced in the compound scans compared to the real-time scans (Table 5 ) . T h e fasciae and femur were better discerned and septa formation was more pronounced in the athletes
All athletes (n = 21)
0.39
2.76
8.8
52.7
k k
Untrained men ( n = 11)
0.37
2.80
9.2
48.4 2 11.1
k
0.56
than in the untrained men. Intramuscular echo intensity was, on the other hand, smaller in the athletes than in the untrained men (Fig. 2). Isometric quadriceps muscle strength correlated significantly with the CSA of the same muscle group, but not with muscle thickness (Table 6).
FIGURE 2. Transverse and longitudinal real-time (a, b) and compound (c, d) US scans of the mid-thigh in an untrained man (a, c) compared with those of a highly trained endurance athlete (b, d). The untrained man shows a marked increase in the echo reflected from intramuscular tissue (fat) and virtually complete loss of the echo reflected from the connective tissue septa and femur.
530
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
~
Table 5. Scores for muscle structure of elderly athletes and untrained men (mean Variable
Power athletes (n = 7)
Endurance athletes (n = 14)
All athletes (n = 21)
11.4 t 4.4 11.7 5 4.2* 9.0 t 2.4* 13.4 t 3.7t 12.1 t 2.3* 11.7 ? 2.1$ 8.0 t 2.9$ 12.3 t 2 . l t
11.1 t 2.8* 11.1 2 2.4' 7.9 2 2.6$ 12.4 t 3.2t 11.2 2.6$ 11.1 t 2.5$ 8.7 t 2.5$ 12.1 t 2.7t
11.2 t 3.3* 11.3 2 3.0t 8.9 t 2.5$ 12.7 t 3.3$ 11.5 t 2.54 11.3 t 2.3$ 8.5 t 2.6$ 12.1 t 2.5$
RFa RS RM RFe CFa
cs CM CFe
*
?
~
~~
SD). Untrained men (n
=
11)
8.4 t 3.7 7.9 t 3.3 11.7 t 2.6 7.7 t 4.1 6.9 t 2.7 6.4 t 2.1 13.8 t 2.2 7.9 t 3.1
'P < 0 05, t P < 0.01, #P < 0.007 (difference between the athletes and the untrained men).
Muscle strength also correlated with the variables of the internal muscle structure evaluated from the compound scans. T h e index of training kilometers correlated with the CSA, strength, and, most particularly, with the internal structure of the muscle group. The greater the number of training kilometers logged, the more discernible were the fasciae, connective tissue septa, and femur, and the smaller was the intensity of the intramuscular echo. Taking the athletes separately, the correlation coefficients were lower, but still in the range 0.19 to 0.62. DISCUSSION
The results of the present study support earlier results on the reliability of US imaging as a research method for the quantitative characterization of skeletal muscle. The coefficient of variation for repeated measurements has been reported as 4%31 and the correlation coefficient 0.98 to 0.99,s,29,30both of which are comparable with our measurements. On the basis of the intra- and inter-observer correlation coefficients, the reproduc-
ibility of the internal muscle structure evaluation developed for this study is also good. Certain special problems occur when quantifying aging muscles with US scanning. In animal studies, the results show an age-related increase in the intramuscular connective tissue content of endomysial" and perimysial origin. Borkan et al' found fat infiltration in the lean tissue when studying the legs of elderly sedentary men. These two age-related phenomena may mask atrophy of the muscle tissue. In this study, US scans of the untrained men and the athletes were different. In the scans of the untrained men, increased intramuscular echogenicity with decreased echoes reflected from larger connective tissue septa and the femur are obviously due to the increase in the proportion of intramuscular connective tissue and fat.5,10,15The reason for these changes in US scans may be multiple interfaces, which are present in lipomatous tissue and collagen. The sound beam is attenuated more rapidly because of these interfaces. A decreased sound beam results in decreased echoes
'
Table 6. Correlation coefficients between training kilometers. quadriceps muscle strength, thickness, CSA, and internal structure variables in elderly men (n = 32, 21 for TRKM) Variable TRKM Strength Thickness CSA RFa RS RM RFe CFa
Strength
Thickness
CSA
RFa
RS
RM
RFe
CFa
cs
CM
CFe
0.455* 1.000
-0.092 0.201 1.000
0.368* 0.488t 0.6954 1.000
0.609t 0.144 -0.240 -0.191 1.000
0.632$ 0.246 -0.275 -0.148 0.898$ 1.000
-0.7404 -0.209 0.018 -0.055 -0.672$ -0.659$ 1.000
0.670$ 0.230 -0.437 -0.233 0.803$ 0.877$ -0.7274 1.000
0.6554 0.332* -0.290 -0.062 0.526$ 0.605$ -0.601$ 0.7854 1.000
0.717$ 0.455t -0.186 -0.01 7 0.528$ 0.595$ -0.6694 0.749$ 0.937$ 1.000
-0.672$ -0.417t 0.048 -0.185 -0.343* -0.4184 0.648$ -0.616$ -0.851$ -0.891 $ 1.000
0.664$ 0.298' -0.259 -0.067 0.569$ 0.615$ -0.5584 0.7644 0.8574 0.849% -0.6974 1.000
cs
CM CFe
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
531
reflected from deeper structures such as larger septa and bone. Heckmatt et a112,13found similar changes in US muscle scans of children with muscular dystrophy. In the needle muscle biopsy, they also found adipose and connective tissue infiltration. Muscle strength correlated with the CSA of the muscle. N o significant correlation was found between strength and muscle thickness. This was not surprising because the measurement of quadriceps muscle thickness in the central mid-thigh excludes both the vastus lateralis and medialis. Strength also correlated with the internal muscle structure evaluated from the compound scans. we is^'^ criticized muscle mass assessment on the basis of CSA because compound scanning causes some distortion of the muscle tissue. Although the compound technique is more sensitive to sources of error, and the lateral resolution of the transducer used in compound scanning is inferior to that used in real-time scanning, the compound technique gives more information about large muscle groups than the real-time technique. Muscle mass and muscle strength decline with age due to muscle fiber loss2o721 and reduction in type I1 fiber It is likely that many elderly people make only few rapid, intensive movements in their daily lives. This slowing of movements does not recruit fast type I1 fibers which may induce atrophy of such fibers. Endurance trainin and aging show similar effects on fiber types.lF The basic reason for slowing due to endurance training is, however, conversion of type I1 fibers to type I and type IIB to type IIA. Intensive strength training may counteract these age-related changes, although training-induced changes in muscle CSA are not so evident in elderly as in younger people. A training-induced increase in muscle strength in elderly people is not necessarily accompanied by an increase in CSA. Moritani et a12* and Penman25 ascribed a greater importance to neural factors compared to hypertrophy in ac-
counting for increased muscle strength with training in elderly men. The groups in this study showed no differences in quadriceps thickness and CSA. The athletes had been training for several years and were still active in competitive sports. The power athletes were, however, mainly jumpers and sprinters. They also participated in endurance-type training, but none of them had included intensive strength training in their training programs. It is clearly a disadvantage for jumpers to have too heavy a muscle mass. Endurance training, which does not hypertrophy muscle fibers, together with the characteristics of the power athlete group (no weight lifters or throwers) may thus explain the insignificant differences between the groups in quadriceps thickness and CSA. It was also difficult to measure the quadriceps of the untrained men because of the poor echoes reflected from the fasciae and bone. The differences between the groups in quadriceps muscle strength might also have been more pronounced had the training of the elderly power athletes borne a closer resemblance to that of their younger counterparts. The cross-sectional nature of the present study has also to be taken into account when interpreting the results. T h e results of the present study, however, indicate that long-term physical training has beneficial effects on muscle functioning in elderly people. Both endurance and power athletes seem to have better muscle “quality” than untrained men. The results also indicate that the more training involved, the higher the muscle strength, the bigger the cross-sectional area, and the better the internal muscular structure. It is possible that long-term training maintains normal muscle architecture with firmer fasciae and connective tissue septa but less “infiltrated” connective tissue and fat, thus counteracting the age-related loss of contractile tissue.
REFERENCES 1 . Alnaqeeb MA, Al Zaid NS, Goldspink G: Connective tissue changes and physical properties of developing and ageing skeletal musc1e.J Anal 1984;139:667-689. 2. Aniansson A, Gustaffson E: Physical training in elderly men with special reference to quadriceps muscle strength and morphology. Clin Physiol 1981; 1:87-98. 3. Aniansson A, Hedberg M, Henning G-B, Grimby G: Muscle morphology, enzymatic activity and muscle strength in elderly men: a follow-up study. Muscle Nerve 1986;9:585591.
532
Ultrasound Imaging of Muscle
4. Aniansson A, Sperling I., Rundgren A, Lehnberg E: Muscle function in 75-year-old men and women. A longitudinal study. Scand J Rehabil Med 1983;9:92- 102. 5. Bartrum RJ, Crow HC: Gray Scale Ultrasound: A Manual for Physictans and Technical Personnel. Philadelphia, W.B. Saunders, 1977. 6. Borkan G, Hults D, Gerzof S, Robbins A, Silber C: Age changes in body composition revealed by computer tomography, J Gerontol 1983;38:673-677. 7. Denis C, Chatard J-C, Dormois D, Linossier M-T, Geyssant
MUSCLE & NERVE
June 1991
A, Lacour J-R: Effects of endurance training on capillary supply of human skeletal muscle in two age groups (20 and 60 years). J Physiol 1986;81:379-383. 8. Dons B, Bollerup K, Bonde-Petersen F, Hancke S: T h e cffects of weight-lift ing exercise related to muscle fibre composition and muscle cross-sectional area in humans. k;ur J Appl l’lzysiol 1979;40:95- 106. 9. Era P: Posture control in the elderly. I n t J Trchno/ A g n g 1988;1: 166- 179. 10. Ferrucci JT: Medical progress. Body ultrasonography. N EnglJ Med 1979;300:538-542. 11. Forst R, Casser H-R, Zilkens K-W: Computergestutzte Skeiettmuskelsonogrammauswertung bei NeuromuskuIaren Erkrankungen, in Stuhler T, Feige A (eds): Ultrashalldiagnostik des Bewegungsapparatus. Berlin, Heidelberg, Springer-Verlag, 1987, p p 89- 102. 12. Heckmatt J , Dubowitz V: Real-time ultrasound imaging of muscles. Muscle Nerve 1988;11:56-65. 13. Heckmatt J , Dubowitz V, Leeman S: Detection of pathological change in dystrophic muscle with B-scan ultrasound imaging. Lancet 1980;1:1389- 1390. 14. Heikkinen E, Arajarvi R-L, Era P, Jylha M, Kinnunen V, Leskinen A-L, Masseli E, Pohjolainen P, Rahkila P, Suominen H , Turpeinen P, Vaisanen M, Osterback L: Functional capacity o~ men born in I906-19I0, 1926-30 and 104650. A basic report. ScaiidJ Soc Med 1984;33(suppl):1-93. 15. Hicks JE, Shawker T H , J o n e s BL, Linzer M, Gerber LH: Diagnostic ultrasound: Its use in the evaluation of muscle. Arch Phys Med Rehahil 1984;65:129- 131. 16. Ikai M, Fukunaga ‘I: Calculation of muscle strength per unit cross-sectional area of human muscle by means o f ultrasonic measurement. Int Z Angew Physiol Eznschl Arboitsphysiol I 968;26:26- 32. 17. Kovanen V, Suominen H: Effects of age and life-time physical training on fibre composition of slow and fdSt skelArch 1987;408:543-55 I . etal muscle in rats. Pfl-’fliiger’.~ 18. Kovanen V, Suominen H , Pelionen L: Effects of aging and life-long physical training on collagen in slow and fast skeletal muscle in rats. A morphometric and immunohistochemical study. Cell Tzssue RPS1987;248:247-255. 19. Larsson L: Morphology and functional characteristics of the ageing skeletal muscle in man. Arta P h y . d Scnnd 1978;(supp1)457. 20. Larsson L: Physical training effects on musclc morphology in sedentary males at different ages. Med Scz Sport.\ Exerc 1982;14:203-206.
Ultrasound Imaging of Muscle
2 1. Lexell J, Henriksson-Larsen K: Distribution of different fiber types in human skeletal muscles: effects of aging studied in whole muscle cross section. M t d t NPRP 1983;6:588-595. 22. Lexell J , Taylor C , SjBstrom M: What is the cause of ageing atrophy? Total number, size and proportion of different fiber types studied in whole vastus lateralis muscle from 15- to 83-year-old men. J Neuro/ Sci 1988;84:275294. 2 3 . MacLennan W, Hall M, Timothy 1, Robinson M: Is weakness in old age due to muscle wasting? Agr and A p n g 1980;9:188- 192. 24. Moritani T, DeVries H : Potential for gross muscle hypertrophy in older men. J Gerontol 1980;35:672-682. 25. Penman KA: Human striated muscle ultrastructural changes accompanying increased strength without hypertrophy. Res Quart 1970;41:418-424. 26. Suominen H , Heikkinen E, Liesen H , Michel D, Hollmann W: Effects of 8 weeks’ endurance training on skeletal muscle metabolism in 56-70-year-old sedentary men. Eur J Appl Physzol 1977;37:173- 180. 27. Suominen H , Heikkinen E, Parkatti T: Effects c:f 8 weeks’ physical training on musclc and connective tissue o f the i n . vastus lateralis in 69-year-old men and wonlen. 1 (;erontol 1977;32:33-37. 28. Suominen H, Rahkila P, Era P, .Jaakkola I*, Heikkinen E: Functional capxity in middle-aged male endurance and power athletes, in Harris R, Harris S (eds): Physicul Actiuity, Aging and Sport 1. Scientqzc and Mrdiral Research. Albany, Center for the Study of Aging, 1989, pp 2 13-2 18. 29. Weiss LW: T h e use of B-mode ultrasound for measuring the thickness of skeletal muscle at t w o upper leg sites.] Orthoped Sports Phys The?-1984;6: 163- 167. YO. Weiss I.W, Clark FC: Ultrasonic measurement of upperarni skeletal muscle thickness. ,J Sports M d Phy.5 F2tnr.s.s 1985;27:128- 133. 31. Young A , Hughes P, Russel M , Parker M, Nichols 1’: Measurement of quadriceps muscle wasting by ultrasonogi-aphy Rheumatol Rrhnbzl 1980;19: 141- 148. 3 2 . Young A, Stoke M, Crowe M: Size and strength of quadriceps muscle o i old and young women. Eur J Clin I n ~ c l 1984; 14:282-287. 33. Young A, Stoke M, Crowe M: T h e size and strength o f the quadriceps muscles of old and young men. Clin I’hyszol 1985:5: 145- 154. ,
I
MUSCLE & NERVE
June 1991
533
ULTRASOUND IMAGING OF THE QUADRICEPS MUSCLE IN ELDERLY ATHLETES AND UNTRAINED MEN SARIANNA SIPILA, MSc, and HARRI SUOMINEN, PhD
W i t h increasing age, human muscles undergo both structural and functional changes, the most striking of which are reduction in muscle mass and strength."42"922 T h e decline of muscle strength is, to some extent, due to muscle atrophy, which is caused by a loss of muscle and by a reduction, especially in type 2, in fiber size.3~19322 A marked reduction in muscle strength accompanied by only a marginal reduction in muscle mass with aging may involve a replacement of muscle tissue by connective tissue. 1,18323 Borkan et a16 also found an age-related infiltration of fat into the lean tissue of the leg muscles. A reduction in physical activity is one reason for the impaired functional capacity of elderly people. Several studies have shown the positive effects of 2 to 6 months of physical training on eldAccording to Kovanen erly muscles.2~7~9~10~20~26~27 et all7 and Suominen et a1," lifelong physical
training counteracts some age-related changes in muscles, although endurance-type training also exerts some effects similar to age-associated changes, such as slowing of muscle fibers.17 Ultrasound (US) imaging has been used for noninvasive, quantitative characterization of muscle tissue. Muscle mass has been assessed from the ~s~8 1,16,31,32 .1 or from the thickness2Y330 of the muscle. US imaging has mostly been used in studies with young or middle-aged subjects, whereas relatively few authors have reported on US imaging of muscles in elderly T h e purpose of this study was to determine the thickness, cross-sectional area, and internal structure of the quadriceps muscle using two different US imaging techniques, and to relate these properties to muscle strength and amount of training in elderly athletes and untrained men. MATERIALS AND METHODS
From the Department of Health Sciences, Unlversity of Jyvaskyla, Jyvaskyla, Finland Acknowledgments: Supported by grants from the State Council for Research in Sport and Physical Education of the Ministry of Education, Finland. The authors thank Mrs. Merja Perhonen and Mr. Pertti Pykala for their skillful technical assistance. Address reprint requests to s. Sipila, Department of Health Sciences, University of Jyvaskyla Seminaarinkatu 15, SF-40100 Jyvaskyla, Finland. Accepted for publication May 8, 1990 CCC 0148- 639)(/91/060527-07 504.00 0 1991 John Wiley & Sons, Inc.
Ultrasound Imaging of Muscle
This study was part of a larger study on health and functional capacity among Finnish male endurance athletes (long-distance runners, orienteers, and cross-country skiers; n = 67) and power athletes (sprinters, jumpers, and throwers; n = 31) aged 70 to 90 years. Most had a lifelong training history and were still active in competitive sports. A random sample of 70- to 81-year-old men (n = 43) from the population register of the rural municipality of Jyvaskyla served as an
Subjects.
MUSCLE & NERVE
June 1991
527
untrained control group. From these groups, 14 endurance athletes, 7 power athletes, and 11 untrained men were randomly selected for the US measurements. T h e physical characteristics of the subjects are shown in Table 2. The athletes had been training regularly for 30 to 70 years, except for one endurance athlete who had been training for 15 years. When a weighted sum index was computed on the basis of the relative energy expenditure and efficiency from selfreported running (1.0), cross-country skiing (0.7), cycling (0.3), walking (0.6), and swimming (4.0), the annual training distance of the endurance athletes averaged 1839 km (range 785 to 3700 km). The power athletes did not usually log training kilometers. Two of them participated in endurance events and recorded their training kilometers. The indices for those two men were 520 and 1426. The power athletes had, however, participated in 3 to 7 training sessions (mean 4.6) per week for track and field, gymnastics, etc. Of the untrained subjects, only one man had participated in more intensive physical activity during the preceding year. His training kilometers yielded an index of 550. T h e other 5 men who reported their training kilometers during the preceding year all had indices below 100. A written informed consent was obtained from all subjects before the laboratory examination. In the clinical examination, no neurological problems were perceived in any of the subjects. Isometric Strength. Maximal isometric quadriceps muscle strength was measured in the right leg in a sitting position on a custom-made dynamometer chair which was a modification of that described by Heikkinen et al.'" T h e earlier apparatus was improved so that knee extension strength could, if required, be measured at different knee joint angles. T h e knee was set at an angle of 60" (full extension O0) and the ankle was attached above the malleolus to a belt-strain-gauge system. The subjects were not encouraged during the trials. The best performance of three trials was taken as the result.
Ultrasound Imaging. T h e US imaging was obtained from the right m. quadriceps by selecting, as the scanning site, the midpoint between the great trochanter and the lateral joint line of the knee. The mid-thigh was located by means of Hexible tape. During scanning, the subjects lay supine
528
Ultrasound Imaging of Muscle
with the leg extended and relaxed on the examination table (Fig. 1). T h e thickness of the muscle was assessed with a real-time scanner (Aloka SSD-280 LS) fitted with a 7.5 MHz linear array transducer. Both longitudinal and transverse scans were performed. The cross-sectional area (CSA) was assessed with a compound US scanner (Aloka SSU-190) fitted with a 5 MHz transducer. The outline of the quadriceps was traced visually using electronic calipers. Both the muscle thickness and CSA were measured by the scanner's own computer. T h e axial and lateral resolution of the former transducer was set at 0.4 mm and 0.9 mm ( - 3 dB), and that of the latter at 0.4 mm and 1.6 mm, respectively. Gain settings and near- and far-field scales were set equally for all subjects. T o avoid tissue compression, a generous amount of gel was used under the probe. The positioning and orientation of the probe was altered until the best bone echo was achieved. The probe was then assumed to be at right-angles to the femur. T h e compound US scans were photographed with a Polaroid camera and the real-time scans were printed on film paper by an Aloka Ultrasono SSZ-95 printer. From the US scans, four different characteristics were evaluated on a four-point scale by two blind observers (Table 1). T h e scales were developed from those reported by Forst et aL9 T o assess the reproducibility of the method, both observers evaluated the pictures twice. Every picture was thus evaluated four times. T h e sum of the four observations was taken as the result for each variable. T h e coefficients of variation for the measure-
FIGURE 1. Position of probe and leg during compound scanning.
MUSCLE & NERVE
June 1991
Table 1. Four-point scales for four different features of US scans and abbreviations used in tables. Discerning fasciae (real-time = RFa, compound = CFa): 1. Fasciae not discernible 2. Fasciae poorly discerned 3. Fasciae moderately discerned 4 Fasciae sharply discerned Intramuscular septa formation (real-time = RS, compound =
CS): 1. 2. 3. 4.
Table 3. Intra- and inter-observer correlation coefficients of analysis of internal muscle structure ( P < 0,001) Variable RFa RS RM RFe CFa
cs
No septa formation Poor septa formation Moderate septa formation Sharp septa formation
CM CFe
Intramuscular echo intensity (real-time = RM, compound = CM): 1 . Echo poor intramuscular structure 2. Small echogenicity 3. Moderately increased echogenicity 4. Strongly increased echogenicity Discerning femur (real-time = RFe, compound 1 . Femur not discernible 2. Femur poorly discerned 3. Femur moderately discerned 4. Femur sharply discerned
=
CFe):
ments repeated over consecutive days were 4.1% for thickness in the transverse scans, 4.8% in the longitudinal scans, and 4.3% for CSA. T h e correlation coefficients were 0.96, 0.95, and 0.99, respectively. The intra- and inter-observer correlation coefficients for the variables representing the internal muscle structure are shown in Table 3. Standard procedures were used to calculate means and standard deviations (SD). The statistical significance of the differences between the two groups of subjects were calculated by Student’s t test (two-tailed) for independent samples and between three groups of subjects by the analysis of variance. T h e differences between the athletes and the untrained men in the Statistical Methods.
Intra-observer
Inter-observer
0.932 0.879 0.836 0.977 0.899 0.853 0.916 0.844
0.830 0.863 0.811 0.939 0.895 0.838 0.713 0.770
internal muscle structure were also calculated by the Mann- Whitney U test. Because there were no differences in the statistical significance between Student’s t test and the Mann-Whitney test, only the results of the t test are reported. Pearson’s coefficient of correlation was used to express the correlations between duplicated measurements and between different variables. Spearman’s coefficient of correlation was used to calculate intraand inter-observer correlations in the evaluation of the internal muscle structure. RESULTS
There were no significant differences between the athletes and the untrained men in age, height, weight, length of the femur, or the content of subcutaneous fat in the middle thigh. Isometric knee extension strength was significantly higher in the power athletes than in the untrained men. No significant differences existed between the power and the endurance athletes, or between the endurance athletes and the untrained men in isometric muscle strength (Table 2). The mean values for thickness and CSA of the quadriceps in the endurance athletes and in the power athletes did not differ from those obtained in the untrained men (Table 4).
Table 2. Physical characteristics of elderly athletes and untrained men (mean Variable Age (years) Height (cm) Weight (kg) Length of femur (cm) Thickness of thigh subcutaneous tissue (mm) Isometric strength of quadriceps muscle (N)
Power athletes (n = 7) Endurance athletes (n
*
77.1 3.5 169.0 & 6.0 69.8 9.3 43.6 ? 3.1 4.3 1.0
* *
74.2 ? 3.0 171.0f 7.3 68.7 2 8.6 43.8 f 2.2 4.7 f 2.3
409* t 53
362t 2 76
=
* SD)
14) All athletes ( n = 21) Untrained men ( n
*
75.2 3.4 170.3 f 6.8 69.1 ? 8.2 43.7 2.5 4.6 1.9
* 377$ * 71
=
11)
73.4 f 2.4 167.6 ? 4.3 74.7 ? 11.9 43.1 ? 2.6 5.8 f 2.1 300
2
83
*P = 0.007, fP = 0.065, #P = 0.010 (difference between the athletes and the untrained men).
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
529
~~
Table 4. Quadriceps muscle thickness and CSA of elderly athletes and untrained men (mean k SD) Variable
Power athletes ( n
=
7)
Endurance athletes (n = 14)
Thickness (cm)
2.73
k
0.37
2.77
CSA (cm’)
52.5
k
10.8
52.7
? 2
The differences in the internal muscular structure between the athletes and the untrained men were more pronounced in the compound scans compared to the real-time scans (Table 5 ) . T h e fasciae and femur were better discerned and septa formation was more pronounced in the athletes
All athletes (n = 21)
0.39
2.76
8.8
52.7
k k
Untrained men ( n = 11)
0.37
2.80
9.2
48.4 2 11.1
k
0.56
than in the untrained men. Intramuscular echo intensity was, on the other hand, smaller in the athletes than in the untrained men (Fig. 2). Isometric quadriceps muscle strength correlated significantly with the CSA of the same muscle group, but not with muscle thickness (Table 6).
FIGURE 2. Transverse and longitudinal real-time (a, b) and compound (c, d) US scans of the mid-thigh in an untrained man (a, c) compared with those of a highly trained endurance athlete (b, d). The untrained man shows a marked increase in the echo reflected from intramuscular tissue (fat) and virtually complete loss of the echo reflected from the connective tissue septa and femur.
530
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
~
Table 5. Scores for muscle structure of elderly athletes and untrained men (mean Variable
Power athletes (n = 7)
Endurance athletes (n = 14)
All athletes (n = 21)
11.4 t 4.4 11.7 5 4.2* 9.0 t 2.4* 13.4 t 3.7t 12.1 t 2.3* 11.7 ? 2.1$ 8.0 t 2.9$ 12.3 t 2 . l t
11.1 t 2.8* 11.1 2 2.4' 7.9 2 2.6$ 12.4 t 3.2t 11.2 2.6$ 11.1 t 2.5$ 8.7 t 2.5$ 12.1 t 2.7t
11.2 t 3.3* 11.3 2 3.0t 8.9 t 2.5$ 12.7 t 3.3$ 11.5 t 2.54 11.3 t 2.3$ 8.5 t 2.6$ 12.1 t 2.5$
RFa RS RM RFe CFa
cs CM CFe
*
?
~
~~
SD). Untrained men (n
=
11)
8.4 t 3.7 7.9 t 3.3 11.7 t 2.6 7.7 t 4.1 6.9 t 2.7 6.4 t 2.1 13.8 t 2.2 7.9 t 3.1
'P < 0 05, t P < 0.01, #P < 0.007 (difference between the athletes and the untrained men).
Muscle strength also correlated with the variables of the internal muscle structure evaluated from the compound scans. T h e index of training kilometers correlated with the CSA, strength, and, most particularly, with the internal structure of the muscle group. The greater the number of training kilometers logged, the more discernible were the fasciae, connective tissue septa, and femur, and the smaller was the intensity of the intramuscular echo. Taking the athletes separately, the correlation coefficients were lower, but still in the range 0.19 to 0.62. DISCUSSION
The results of the present study support earlier results on the reliability of US imaging as a research method for the quantitative characterization of skeletal muscle. The coefficient of variation for repeated measurements has been reported as 4%31 and the correlation coefficient 0.98 to 0.99,s,29,30both of which are comparable with our measurements. On the basis of the intra- and inter-observer correlation coefficients, the reproduc-
ibility of the internal muscle structure evaluation developed for this study is also good. Certain special problems occur when quantifying aging muscles with US scanning. In animal studies, the results show an age-related increase in the intramuscular connective tissue content of endomysial" and perimysial origin. Borkan et al' found fat infiltration in the lean tissue when studying the legs of elderly sedentary men. These two age-related phenomena may mask atrophy of the muscle tissue. In this study, US scans of the untrained men and the athletes were different. In the scans of the untrained men, increased intramuscular echogenicity with decreased echoes reflected from larger connective tissue septa and the femur are obviously due to the increase in the proportion of intramuscular connective tissue and fat.5,10,15The reason for these changes in US scans may be multiple interfaces, which are present in lipomatous tissue and collagen. The sound beam is attenuated more rapidly because of these interfaces. A decreased sound beam results in decreased echoes
'
Table 6. Correlation coefficients between training kilometers. quadriceps muscle strength, thickness, CSA, and internal structure variables in elderly men (n = 32, 21 for TRKM) Variable TRKM Strength Thickness CSA RFa RS RM RFe CFa
Strength
Thickness
CSA
RFa
RS
RM
RFe
CFa
cs
CM
CFe
0.455* 1.000
-0.092 0.201 1.000
0.368* 0.488t 0.6954 1.000
0.609t 0.144 -0.240 -0.191 1.000
0.632$ 0.246 -0.275 -0.148 0.898$ 1.000
-0.7404 -0.209 0.018 -0.055 -0.672$ -0.659$ 1.000
0.670$ 0.230 -0.437 -0.233 0.803$ 0.877$ -0.7274 1.000
0.6554 0.332* -0.290 -0.062 0.526$ 0.605$ -0.601$ 0.7854 1.000
0.717$ 0.455t -0.186 -0.01 7 0.528$ 0.595$ -0.6694 0.749$ 0.937$ 1.000
-0.672$ -0.417t 0.048 -0.185 -0.343* -0.4184 0.648$ -0.616$ -0.851$ -0.891 $ 1.000
0.664$ 0.298' -0.259 -0.067 0.569$ 0.615$ -0.5584 0.7644 0.8574 0.849% -0.6974 1.000
cs
CM CFe
Ultrasound Imaging of Muscle
MUSCLE & NERVE
June 1991
531
reflected from deeper structures such as larger septa and bone. Heckmatt et a112,13found similar changes in US muscle scans of children with muscular dystrophy. In the needle muscle biopsy, they also found adipose and connective tissue infiltration. Muscle strength correlated with the CSA of the muscle. N o significant correlation was found between strength and muscle thickness. This was not surprising because the measurement of quadriceps muscle thickness in the central mid-thigh excludes both the vastus lateralis and medialis. Strength also correlated with the internal muscle structure evaluated from the compound scans. we is^'^ criticized muscle mass assessment on the basis of CSA because compound scanning causes some distortion of the muscle tissue. Although the compound technique is more sensitive to sources of error, and the lateral resolution of the transducer used in compound scanning is inferior to that used in real-time scanning, the compound technique gives more information about large muscle groups than the real-time technique. Muscle mass and muscle strength decline with age due to muscle fiber loss2o721 and reduction in type I1 fiber It is likely that many elderly people make only few rapid, intensive movements in their daily lives. This slowing of movements does not recruit fast type I1 fibers which may induce atrophy of such fibers. Endurance trainin and aging show similar effects on fiber types.lF The basic reason for slowing due to endurance training is, however, conversion of type I1 fibers to type I and type IIB to type IIA. Intensive strength training may counteract these age-related changes, although training-induced changes in muscle CSA are not so evident in elderly as in younger people. A training-induced increase in muscle strength in elderly people is not necessarily accompanied by an increase in CSA. Moritani et a12* and Penman25 ascribed a greater importance to neural factors compared to hypertrophy in ac-
counting for increased muscle strength with training in elderly men. The groups in this study showed no differences in quadriceps thickness and CSA. The athletes had been training for several years and were still active in competitive sports. The power athletes were, however, mainly jumpers and sprinters. They also participated in endurance-type training, but none of them had included intensive strength training in their training programs. It is clearly a disadvantage for jumpers to have too heavy a muscle mass. Endurance training, which does not hypertrophy muscle fibers, together with the characteristics of the power athlete group (no weight lifters or throwers) may thus explain the insignificant differences between the groups in quadriceps thickness and CSA. It was also difficult to measure the quadriceps of the untrained men because of the poor echoes reflected from the fasciae and bone. The differences between the groups in quadriceps muscle strength might also have been more pronounced had the training of the elderly power athletes borne a closer resemblance to that of their younger counterparts. The cross-sectional nature of the present study has also to be taken into account when interpreting the results. T h e results of the present study, however, indicate that long-term physical training has beneficial effects on muscle functioning in elderly people. Both endurance and power athletes seem to have better muscle “quality” than untrained men. The results also indicate that the more training involved, the higher the muscle strength, the bigger the cross-sectional area, and the better the internal muscular structure. It is possible that long-term training maintains normal muscle architecture with firmer fasciae and connective tissue septa but less “infiltrated” connective tissue and fat, thus counteracting the age-related loss of contractile tissue.
REFERENCES 1 . Alnaqeeb MA, Al Zaid NS, Goldspink G: Connective tissue changes and physical properties of developing and ageing skeletal musc1e.J Anal 1984;139:667-689. 2. Aniansson A, Gustaffson E: Physical training in elderly men with special reference to quadriceps muscle strength and morphology. Clin Physiol 1981; 1:87-98. 3. Aniansson A, Hedberg M, Henning G-B, Grimby G: Muscle morphology, enzymatic activity and muscle strength in elderly men: a follow-up study. Muscle Nerve 1986;9:585591.
532
Ultrasound Imaging of Muscle
4. Aniansson A, Sperling I., Rundgren A, Lehnberg E: Muscle function in 75-year-old men and women. A longitudinal study. Scand J Rehabil Med 1983;9:92- 102. 5. Bartrum RJ, Crow HC: Gray Scale Ultrasound: A Manual for Physictans and Technical Personnel. Philadelphia, W.B. Saunders, 1977. 6. Borkan G, Hults D, Gerzof S, Robbins A, Silber C: Age changes in body composition revealed by computer tomography, J Gerontol 1983;38:673-677. 7. Denis C, Chatard J-C, Dormois D, Linossier M-T, Geyssant
MUSCLE & NERVE
June 1991
A, Lacour J-R: Effects of endurance training on capillary supply of human skeletal muscle in two age groups (20 and 60 years). J Physiol 1986;81:379-383. 8. Dons B, Bollerup K, Bonde-Petersen F, Hancke S: T h e cffects of weight-lift ing exercise related to muscle fibre composition and muscle cross-sectional area in humans. k;ur J Appl l’lzysiol 1979;40:95- 106. 9. Era P: Posture control in the elderly. I n t J Trchno/ A g n g 1988;1: 166- 179. 10. Ferrucci JT: Medical progress. Body ultrasonography. N EnglJ Med 1979;300:538-542. 11. Forst R, Casser H-R, Zilkens K-W: Computergestutzte Skeiettmuskelsonogrammauswertung bei NeuromuskuIaren Erkrankungen, in Stuhler T, Feige A (eds): Ultrashalldiagnostik des Bewegungsapparatus. Berlin, Heidelberg, Springer-Verlag, 1987, p p 89- 102. 12. Heckmatt J , Dubowitz V: Real-time ultrasound imaging of muscles. Muscle Nerve 1988;11:56-65. 13. Heckmatt J , Dubowitz V, Leeman S: Detection of pathological change in dystrophic muscle with B-scan ultrasound imaging. Lancet 1980;1:1389- 1390. 14. Heikkinen E, Arajarvi R-L, Era P, Jylha M, Kinnunen V, Leskinen A-L, Masseli E, Pohjolainen P, Rahkila P, Suominen H , Turpeinen P, Vaisanen M, Osterback L: Functional capacity o~ men born in I906-19I0, 1926-30 and 104650. A basic report. ScaiidJ Soc Med 1984;33(suppl):1-93. 15. Hicks JE, Shawker T H , J o n e s BL, Linzer M, Gerber LH: Diagnostic ultrasound: Its use in the evaluation of muscle. Arch Phys Med Rehahil 1984;65:129- 131. 16. Ikai M, Fukunaga ‘I: Calculation of muscle strength per unit cross-sectional area of human muscle by means o f ultrasonic measurement. Int Z Angew Physiol Eznschl Arboitsphysiol I 968;26:26- 32. 17. Kovanen V, Suominen H: Effects of age and life-time physical training on fibre composition of slow and fdSt skelArch 1987;408:543-55 I . etal muscle in rats. Pfl-’fliiger’.~ 18. Kovanen V, Suominen H , Pelionen L: Effects of aging and life-long physical training on collagen in slow and fast skeletal muscle in rats. A morphometric and immunohistochemical study. Cell Tzssue RPS1987;248:247-255. 19. Larsson L: Morphology and functional characteristics of the ageing skeletal muscle in man. Arta P h y . d Scnnd 1978;(supp1)457. 20. Larsson L: Physical training effects on musclc morphology in sedentary males at different ages. Med Scz Sport.\ Exerc 1982;14:203-206.
Ultrasound Imaging of Muscle
2 1. Lexell J, Henriksson-Larsen K: Distribution of different fiber types in human skeletal muscles: effects of aging studied in whole muscle cross section. M t d t NPRP 1983;6:588-595. 22. Lexell J , Taylor C , SjBstrom M: What is the cause of ageing atrophy? Total number, size and proportion of different fiber types studied in whole vastus lateralis muscle from 15- to 83-year-old men. J Neuro/ Sci 1988;84:275294. 2 3 . MacLennan W, Hall M, Timothy 1, Robinson M: Is weakness in old age due to muscle wasting? Agr and A p n g 1980;9:188- 192. 24. Moritani T, DeVries H : Potential for gross muscle hypertrophy in older men. J Gerontol 1980;35:672-682. 25. Penman KA: Human striated muscle ultrastructural changes accompanying increased strength without hypertrophy. Res Quart 1970;41:418-424. 26. Suominen H , Heikkinen E, Liesen H , Michel D, Hollmann W: Effects of 8 weeks’ endurance training on skeletal muscle metabolism in 56-70-year-old sedentary men. Eur J Appl Physzol 1977;37:173- 180. 27. Suominen H , Heikkinen E, Parkatti T: Effects c:f 8 weeks’ physical training on musclc and connective tissue o f the i n . vastus lateralis in 69-year-old men and wonlen. 1 (;erontol 1977;32:33-37. 28. Suominen H, Rahkila P, Era P, .Jaakkola I*, Heikkinen E: Functional capxity in middle-aged male endurance and power athletes, in Harris R, Harris S (eds): Physicul Actiuity, Aging and Sport 1. Scientqzc and Mrdiral Research. Albany, Center for the Study of Aging, 1989, pp 2 13-2 18. 29. Weiss LW: T h e use of B-mode ultrasound for measuring the thickness of skeletal muscle at t w o upper leg sites.] Orthoped Sports Phys The?-1984;6: 163- 167. YO. Weiss I.W, Clark FC: Ultrasonic measurement of upperarni skeletal muscle thickness. ,J Sports M d Phy.5 F2tnr.s.s 1985;27:128- 133. 31. Young A , Hughes P, Russel M , Parker M, Nichols 1’: Measurement of quadriceps muscle wasting by ultrasonogi-aphy Rheumatol Rrhnbzl 1980;19: 141- 148. 3 2 . Young A, Stoke M, Crowe M: Size and strength of quadriceps muscle o i old and young women. Eur J Clin I n ~ c l 1984; 14:282-287. 33. Young A, Stoke M, Crowe M: T h e size and strength o f the quadriceps muscles of old and young men. Clin I’hyszol 1985:5: 145- 154. ,
I
MUSCLE & NERVE
June 1991
533
