metronome. They differed with respect to angular velocity and movement frequency as shown in Table 2. Successive fatigue tests were separated by at least 20 ...
Applied Physiology European Journal of
Eur J Appl Physiol (1989) 59:80-88
and Occupational Physiology © Springer-Vertag 1989
Influence of angular velocity and movement frequency on development of fatigue in repeated isokinetic knee extensions* Svend Erik Mathiassen National Institute of Occupational Health, Department of Physiology, Division of Applied Physiology, S-17184 Solna, Sweden
Summary. Six sedentary students, six orienteers, and six soccer players were each subjected to 15 tests, comprising 120 s of repeated, maximal isokinetic knee extensions. The tests differed with respect to movement velocity (30 ° .s -1, 120 ° .s -1, and 300 ° . s - l ) , and movement frequency (5 at each velocity). At a certain velocity, a rectilinear relationship was found between muscular performance intensity (expressed either as average power output or as exercise time ratio) and development of fatigue (expressed either as an absolute or as a fractional decline in work output). Significant inter-velocity differences existed between the slopes of these lines at some combinations of performance and fatigue expressions. Only tendencies towards a difference in x-intercept values were found. This x-intercept value can be taken as a measure of the greatest attainable intensity level of performance without the development of fatigue. This suggestion is valuable both in basic physiological research, and as a possible criterion for optimization of muscular performance. At a given exercise time ratio, increasing movement velocity produced increasing fatigue. However, at a given muscular power output -- above 15 W approximately -- fatigue developed to a greater extent at the low velocity than at the two higher ones, which did not differ significantly. Substantial individual variation was seen in the positions of the low-, medium-, and high-velocity lines. These variations did not depend on the training background. This implies that the validity of using single-velocity, single-frequency tests in determining isokinetic endurance is doubtful. * This work was done as an MD dissertation at the August Krogh Institute, University of Copenhagen, Denmark. All measurements were made at the Institute of Physical Education, University of Odense, Denmark.
Key words: Endurance -- Fatigue -- Skeletal muscle -- Isokinetics -- Dynamic
Introduction Isokinetic methods have been used extensively since the early seventies in attempts to measure and understand the development of muscular fatigue. Fatigue, in this context, is defined as the inability to maintain a given, maximal strength level when performing repeated, maximal isokinetic contractions. Only one fatigue test regimen has been commonly used: 50 or 100 contractions at a frequency of 50.min -1, movement velocity 180 ° . s - i (Thorstensson and Karlsson 1976). It is not known whether the results of using this "Thorstensson test" predicts the performance capacity of a person at other movement velocities and frequencies. In general it would seem that there has been little work done on the influence of velocity and frequency of movement on the development of fatigue. The "Thorstensson test" has some drawbacks, making its exact reproduction uncertain: (1) the muscular effort required to lift the moved extremity (e.g. the lower leg) against gravity is not quantified (Winter et al. 1981); (2) the possible influence on the test result of impact-related torque spikes is not taken into account (Sapega et al. 1982); and (3) the prescribed movement frequency is so high, that it can be attained only with great difficulty when the correct movement range (e.g. knee extension from 90 ° to 0 ° knee angle) is to be completed in every single movement. The purpose of this study was to investigate the development of muscle fatigue in repeated
S. E. Mathiassen: Factors influencing fatigue in isokinetic movements
81
isokinetic knee extensions at different angular velocities and at different movement frequencies, with special attention to the training history of the individuals studied.
Table 2. Frequency of movement (J) and exercise time (ET) in each of the fatigue tests at three velocities given in the chronological order used
Materials and methods
f (min -1)
ET (%)
f (min-')
ET (%)
f (min-')
ET (%)
Six sedentary students (SE), 6 orienteering runners (OR) a n d 6 soccer players (FO), all males, volunteered for the study after being informed of the purpose a n d possible discomforts of the experiments. Table 1 presents the average values of the physical characteristics of the groups. The SE group did not do any regular physical training, and had not done any for at least 5 years, except for cycling (at most 10 km daily, 5 days each week). The OR group were elite at regional level, each having at least 8 years' running experience, and training at least 6 h each week, including competitions. The FO group were likewise of good regional standard, with a history of not less than 8 years' training of at least 6 hours each week, including matches, excluding the months of November and December. All experiments were conducted with an isokinetic dynamometer (Cybex II, Lumex Inc., New York), equipped with a strain-gauge lever arm (Cykob, Stockholm), and an angle potentiometer. The electrical outputs of the lever arm, a n d of the angle potentiometer were calibrated prior to every test session, using known weights and angles. The subjects were seated in the experiment chair with a 90 ° hip flexion, and with their thighs placed horizontally (i.e. a horizontal femur). To obtain this position the front part of the chair was raised 15 ° . The subjects were strapped across the waist and across b o t h thighs a n d were allowed to hold on to the chair. The rotation axis of the right knee joint was aligned with the pivot axis of the Cybex lever arm and the lower leg was fixed to the lever arm cushion at a distance of 20% body height from the knee axis. To ensure that the lower leg and lever arm were parallel, the lever arm cushion was displaced 6 cm anteriorly of the lever arm. All knee extensions started at a knee angle of 90 °. This was ensured by a "brake cushion" posterior to the lever arm. The subjects were instructed to exert maximal effort throughout the entire range of movement to 0 ° knee angle (fully stretched leg). The simultaneous outputs from the angle poten-
12 20 8 16 24
15 25 10 20 30
20 40 10 30 50
10 20 5 15 25
4 6 3 5 7
20 30 t5 25 35
Day 2 120 ° .s -1
Day 3 300 ° .s -1
Day 4 30 ° . s - I
Exercise time, defined as E T % = (time actually spent in knee extensions). 100/(total test time)
tiometer and the strain-gauge lever arm were recorded on a two-channel digital storage oscilloscope (Gould OS 4000, Gould Advance, Essex, UK) for further analysis. Inspection of the oscilloscope record served as a check on the proper execution of the complete knee extension. The subjects were tested on four occasions within 2 weeks, each test session was separated from the others by at least 48 h. The experiments were carried out in March (SE and OR), and April (FO), which is the early competitive season in orienteering as well as in soccer. The subjects were asked to avoid any training within 12 h prior to a test session. The 1st test day was used to familiarize the subjects with the Cybex equipment, to determine anthropometric data including thigh circumference and gravitational torque of the lower leg, and to measure the maximal isokinetic strength of the subjects at a number of movement velocities, among them 30 ° .s -~, 120 ° -s -1, and 300 ° .s -1. On each of the remaining 3 test days, five fatigue tests were performed at one of the movement velocities. Each of these 3 x 5 tests comprised 120 s of regularly repeated maximal knee extensions regulated by a metronome. They differed with respect to angular velocity and movement frequency as shown in Table 2. Successive fatigue tests were separated by at least 20 min. The three velocities were chosen to cover the full range of angular velocities offered by the Cybex equipment. On each test day, the actual
Table 1. Personal data of subjects (n = 6 in all three groups) Groups
SE OR FO
mean SD mean SD mean SD
Age (year)
Height (cm)
Weight (kg)
TC (cm)
IS30 (J)
IS120 (J)
IS300 (J)
22.2 + 1.6 27.2 +3.4 24.0 +4.1
177.5 __.7.5 177.5 +3.7 181.8 +5.2
66.5 V +6.0 68.8 F +4.2 80.4 s° ±6.2
49.4 v ±2.2 49.8 v +2.0 54.8 s° ±2.5
187.1F ±24.8 208.1 +32.7 262.0 s ±42.5
166.1V ± 10.3 174.7 +28.0 212.2 s +24.5
114. 4 ± 14.7 102.3 v _ 15.3 125.6 ° + 12.0
TC = thigh circumference, measured mid-way between the trochanter major and the epicondylus lateralis femoris in relaxed standing IS30, IS120, IS300 = maximal isokinetic work output in one complete knee extension from 90 ° knee angle to 0 ° ; movement velocities 30 ° • s - i, 120 ° . s - 1, and 300 ° • s - J respectively F significantly different (p 0.05) in slope values were found between velocities, or between groups. The MNI-values are identical to those in Table 4. This leaves the impression that the ET%-
30 o. s 1: SLOPE: 1.35. 10`2 MNI: 1,21 • 103
, 60 '
300 o . s-l: SLOPE: 0.55 - ~0-2 MI'~: 0.48, 1{}
/
50'
;
40"
-""
/
,~:
/
)*
120 * . s ' l :
/
t
MNI:
2
1,52 1. • 10
lO
o q, . / ) ' / / 0
2
,
.
4
6
.
AVERAGE POV~%R OUTPUT A T ONSET OF TEST
. R
10
.
12
/QUADRICEPSAREA
1~
, !6
103(w'r62)
Fig. 2. Average power output at onset of test/quadriceps area (APO/QA) vs work output decline% (WOD%). Mean slopes (m~.W -1) and mean maximal non-fatiguing intensity level [MNI (W.m-2)] at the three velocities tested. All 18 subjects are included. Extrapolation beyond the investigated range in APO/QA is shown as broken lines
W O D / Q A relationship is characterized by a bunch of parallel lines, displaced still further to the left with increasing velocity. Figure 3 shows this relationship for all subjects.
Discussion
Methodological considerations The usual parameter for stating maximal strength in isokinetic studies is torque -- either angle-specific or peak. The alternative choice of work output in this study was motivated by the virtues of this parameter being less sensitive to the "torqueovershoot'-oscillations of the Cybex system (Sa-
Table 5. The average power output/quadriceps area vs work output decline% relationship. Mean values of slope and maximal non-fatigue intensity level (MNI). Definitions as in Table 4 Group
Movement velocity 30° .s -a
120° .s -~
Slope (m 2.W-1) M (~ F S
300° .s-I
MNI (W. m-Z)
Slope (m2. W-~)
MNI (W. m-2)
Slope (m2.W-1)
MNI (W-m-2)
1.68.103 M H F S
0.59.10_2 L H F S
1.88.103 L H F S
0.52.t0-2 (D M F S
0.95.103
L M F S
OR
1.12.t0_2
FO
1.37.10_ z ( ~ ( ~ OS
0.66.103
M H OS
0.57.10-2 (D H O S
0.59.103
L H O®
0.55.10-2 C ) M O S
0.17.103
L M O S
SE
t.57.10_ 2 ( ~ ) OF
1.30.103
M H OF
0.69.10_z (~)H OF
2.09.103
L H O (~)
0.58.10-2 (D M OF
0.33.103
L M OF
SD
0.21" 10 -2
0.32.103
0.07.10 -2
0.40.103
0.04.10 -2
0.51.103
S. E. Mathiassen: Factors influencing fatigue in isokinetic movements
85
Table 6. The exercise time % vs work output decline/quadricepts area relationship. Mean values of slope, and maximal non-fatigue intensity level (MNI). Definitions as in Table 4 Group
Movement velocity 30 ° . s - '
120 ° .s -1
Slope (J. m - z)
MNI (go)
OR
237
20.2
FO
302
M H F S MH O S
SE
334
M H OF
15.0
SD
35
7.8
Slope (J. m - 2) M (~) F S MH O S
296
M H OF
409
L F L O
337
3.7
7.6
L H OF
7.3
43
< 121
5
120 o . s-l: SLOPE: 347 MNI: 5,7 300 *
. s'l:
°
SLOPE: 317 MNI:
8
1.2
..-" .,
6SLOP~: 291 MNI I 4 3 4"
©
o
0
'"
0
I
10
20 EXERCISE
30
2.1
1.6
• 10 3
0.05) at the two highest velocities (Table 5, Fig. 2), whereas the low-velocity line is significantly steeper. MNIvalues do not depend significantly on movement velocity. The idea of a velocity-dependent "power-output:fatigue" relationship is of obvious importance in, for example, job designing. From the results in this study, it may be suggested that, in this sense, the most "favourable" movement velocity might depend on the required muscular performance intensity. However, it is still not known whether a certain external power output implies equal internal metabolic turnover in movements of different velocity. Therefore, the metabolic "power-output:fatigue" relationship might differ from the biomechanical relationship described in this study.
86
S.E. Mathiassen: Factors influencingfatigue in isokinetic movements
In the relation ET% vs W O D / Q A no velocitydependence of slope values was found, but the lines tended to differ in respect to their MNI-values. If this result can be confirmed in future studies, it opens up an easy way to predict fatigue development from the demands of muscular performance. Results given in the literature that are directly comparable to those presented in this study are sparse. Barnes (1981) used test regimens at both 120 ° .s -1 and 300 ° .s -1, but with a test duration of only 10 s. Clarke and Manning (1984) tested at 120 °. s-a as well as at other test velocities, but did not publish their results. The figures of Komi and Viitasalo (1977) presenting data from an experiment at 30 ° -s -1 and 23 ET%, show a Force Decline% of approximately 15% (this study: 13.0 WOD% at 25 ET%). However, the test movement was a whole-leg press, involving other muscle synergies besides the knee extensors. This complicates comparisons, because the actual speed of shortening of a muscle not only depends on the movement velocity of the corresponding body segment, but also on the mechanical lever arm of that muscle (Hinson et al. 1979). Comparisons of isokinetic experiments done on different muscle groups are thus futile. Furthermore, it is as yet an unsolved question whether torque (or force) and work output decline in parallel during the execution of repeated, fatiguing movements. Orr and Green (1981) reported no development of fatigue in tests at 300 ° .s -~, lasting 200 min, at 6 ET% and 10 ET%. These results are not in agreement with the findings in the present study (SE subject: 13.5 WOD% at 6 ET%; 23.6 WOD% at 10 ET%). Whether decrease in muscular performance in repeated movements is dependent on movement velocity, has been investigated only by Barnes (1981), who used different ET% at different test velocities, thus excluding conclusions about the influence of the velocity alone. Inter-study comparisons between tests at different velocities are complicated by the wide range of ET% used. Although obvious reservations exist for comparison of the results of the present study with findings from tests made at other velocities, the great amount of information from "Thorstensson tests" makes such comparisons relevant. A typical result from a person, with intermediate fibre type distribution, performing a "Thorstensson test" (180 ° .s -~, 42 ET%) would be a decline in peak torque of 55% (Inbar et al. 1981 ; Komi and Tesch 1979; Larsson and Karlsson 1978; Nilsson et al. 1977; Thorstensson and Karlsson 1976). The test duration -- 60 s or 120 s -- is of minor impor-
tance (Clarkson et al. 1982; Komi and Tesch 1979; Thorstensson and Karlsson 1976). Assuming similarity between peak-torque-decline% and WOD%, this 55% equals, after correction for antigravity work, a WOD% of approximately 48%. A typical sedentary subject of the present study would have performed an A P O / Q A of approximately 1 2 5 0 0 W . m -2 at the onset of a "Thorstensson test" (using the results of day 1). This A P O / Q A indicates (Table 5) a WOD% of 70%75% at a movement velocity between 120°.s -1 and 300 ° •s - 1 This is 25% more than the 48% estimated from the "Thorstensson test" result above. This might be viewed as a substantiation of the suggestion that a "Thorstensson test" is actually executed at a considerably lower ET% than claimed.
Individual factors A finding of the present study was that fatigue development (i.e. the slope of the ET% vs WOD% relationship) does not depend on the initial nonfatigued level of strength. This is not in agreement with the results of several other authors (Clarkson et al. 1982; Costill et al. 1979; Patton et al. 1978). That initial strength per se is not a valid indicator for isokinetic endurance is, however, shown by the finding of Larsson and Karlsson (1978) that an age-related decrease in maximal isokinetic strength does not affect the relative torque decline in a "Thorstensson test". The present study did not show any clear influence of the training background on isokinetic fatigue development, although the OR group tended to be the least susceptible to fatigue. Probably, the OR group had a greater aerobic capacity (Vo2max) for leg activity than other two groups, the FO group having the greatest anaerobic potential. Evidence is found in the literature to show that neither /)'o. . . . nor mitochondrial enzyme activity influence the result in a "Thorstensson test" (Brown and Wilkinson 1983; Costill et al. 1979; Kaiser 1984; Orlander et al. 1979). However, the results in the present study indicate that a correlation might exist at movement velocities other than 180 °. s-1, depending on the choice of parameters for muscular performance and fatigue. In accordance with the results in this study, no significant correlations are to be found in the literature relating fatigue development and indices for anaerobic capacity, such as score in a Wingate test (Inbar et al. 1981), performance in sprint-running and different types of jump (Brown and Wil-
S. E. Mathiassen: Factors influencing fatigue in isokinetic movements
kinson 1983), and activity of glycolytic enzymes in homogenates of muscle biopsies (Costill et al. 1979; Larsson and Karlsson 1978). However, it should be stressed that expectations of isokinetic capacity based on performance in "every-daylife" are complicated by the fact that isokinetic movements are very rare outside the laboratory. Biomechanical and physiological evaluations of the validity of isokinetic observations in non-isokinetic situations have not been done so far. Great individual variations were found among subjects in the present study, not only in susceptibility to fatigue in general (Tables 4, 5, 6), but also in the mutual relationship between the development of fatigue at the three velocities. This indicates that factors among individuals other than those investigated here influence isokinetic endurance. One such factor could be the fibre type distribution of the quadriceps femoris muscle, as shown in several studies using "Thorstensson tests" (Larsson and Karlsson 1978; Nilsson et al. 1977; Tesch 1980; Thorstensson and Karlsson 1976). Conclusions
1. Velocity of movement is an important, independent factor influencing development of muscular fatigue in isokinetic work. 2. A linear relationship exists between intensity of muscular performance and development of fatigue in repeated, isokinetic knee extensions. 3. Neither maximal isokinetic strength nor training background have any substantial influence on development of isokinetic fatigue. 4. A single-velocity, single-frequency test does not give a valid determination of isokinetic endurance in general. Acknowledgements. I wish to thank John Henning for his cooperation, and to Preben K Pedersen and Kurt Jorgensen for constructive reviews.
References Barnes WS (1981) Isokinetic fatigue curves at different contractile velocities. Arch Phys Med Rehabil 62:66-69 Braune W, Fischer O (1892) Bestimmung der Trfigheitsmomente des menschlichen Krrpers und seiner Glieder. Abh Mat Phys Klasse Krn Sfichs Ges Wis 18:408-492 Brown SL, Wilkinson JG (1983) Characteristics of national, divisional, and club male alpine ski racers. Med Sci Sports Exerc 15:491-495 Campbell RC (1979) Statistics for biologists. 2nd edn. Cambridge University Press, London Clarke DH, Manning JM (1984) Properties of isokinetic fatigue at varying movement speeds in adult males. Med Sci Sports Exerc 16:141
87
Clarkson PM, Johnson J, Dextradeur D, Leszczynski W, Wai J, Melchionda A (1982) The relationships among isokinetic endurance, initial strength level, and fiber type. Res Q Exerc Sports 53:15-19 Costill DL, Coyle EF, Fink WF, Lesmes GR, Witzmann FA (1979) Adaptations in skeletal muscle following strength training. J Appl Physiol 46:96-99 Dempster WT (I955) Space requirements of the seated operator. WADC Technical Report 55-159. Wright-Patterson Air Force Base, Ohio Fugl-Meyer AR, Gerdle B, L~ngstrrm M (1985) Characteristics of repeated isokinetic plantar flexions in middle-aged and elderly subjects with special regard to muscular work. Acta Physiol Scand 124:213-222 Henning J (1985) Evaluation of isokinetic concentric muscular strength in the knee extensors of young men (in danish). Dissertation, Laboratory of Theory of Gymnastics, August Krogh Institute, Copenhagen Hinson MN, Smith WC, Funk S (1979) Isokinetics: a clarification. Res Q Exerc Sports 50:30-35 Inbar O, Kaiser P, Tesch P (1981) Relationships between leg muscle fiber type distribution and leg exercise performance. Int J Sports Med 2:154-159 Jacobs I, Kaiser P, Tesch P (1981) Muscle strength and fatigue after selective glycogen depletion in human skeletal muscle fibers. Eur J Appl Physiol 46:47-53 Kaiser P (1984) Physical performance and muscle metabolism during fl-adrenergic blockade in man. Acta Physiol Scand [Suppl 536]: 1-53 Komi PV, Tesch P (1979) EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol 42:41-50 Komi PV, Viitasalo JT (1977) Changes in motor unit activity and metabolism in human skeletal muscle during and after repeated eccentric and concentric contractions. Acta Physiol Scand 100:246-254 Larsson L, Karlsson J (1978) Isometric and dynamic endurance as a function of age and skeletal muscle characteristics. Acta Physiol Scand 104:129-136 Nilsson J, Tesch P, Thorstensson A (1977) Fatigue and EMG of repeated fast voluntary contractions in man. Acta Physiol Scand 101:194-198 Orlander J, Kiessling K-M, Larsson L (1979) Skeletal muscle metabolism, morphology, and function in sedentary smokers and non-smokers. Acta Physiol Scand 107:39-46 Orr G, Green H (1981) Muscle fiber recruitment, glycogen depletion, and fatigue of Vastus Lateralis during high velocity isokinetic exercise. Med Sci Sports Exerc 13:94 Patton RW, Hinson M, Arnold BR, Lessard B (1978) Fatigue curves of isokinetic contractions. Arch Phys Med Rehabil 59: 507-509 Sapega AA, Nicholas JA, Sokolow D, Saranifi A (1982) The nature of "torque overshoot" in Cybex isokinetic dynamometry. Med Sci Sports Exerc 14:368-375 Schantz P, Randall-Fox E, Hutchinson W, Tyden A, Astrand P-O (1983) Muscle fiber type distribution, muscle crosssectional area, and maximal voluntary strength in humans. Acta Physiol Scand 103:40-46 Tesch P (1980) Fatigue patterns in subtypes of human skeletal muscle fibers. Int J Sports Med 1:79-81 Thorstensson A, Karlsson J (1976) Fatiguability and fibre type composition of human skeletal muscle. Acta Physiol Scand 98:318-322 Winter DA, Wells RP, Orr GW (1981) Errors in the use of isokinetic dynamometers. Eur J Appl Physiol 46:397-408 Accepted January 16, 1989
88
S.E. Mathiassen: Factors influencing fatigue in isokinetic movements
Appendix A
Ta was found as:
T.=l.a
The corrected instantaneous torque value (Tc) was calculated according to the formula: To = Tm + Tg + T~
Tg was found as: Tg = TI- cos 0
where Tm torque output, measured by the strain-gauge lever arm Tg gravitational torque of lower leg and lever arm (see below) T, acceleration/deceleration torque of lower leg and lever arm (see below) where Tl
torque of horizontally placed lower leg and lever arm (measured) angular position, measured by the angle potentiometer
where I
moment of inertia of lower leg and lever arm with respect to the knee joint axis (see below) a instantaneous angular acceleration, found by double differentiation of the relationship time vs 0 I was estimated, using the regression equation: I=0.0016.D+0.0467 where D product of height 2 (m z) and weight (kg) of the subject This equation was based on data from Braune and Fischer (1892), Dempster (1955), and on own observations The total work output (WO) of one knee extension was found as:
WO =2"(Tc.AO)
where AO smallest digitally measurable increment in angular position (approximately 10 -2 rad) 2" is taken from 0 to ~/2
