There are several studies that measured position sense of the arms (9) and legs (7) by using different methodologic approaches. Walsh et al. (9), using a customized device, asked subjects to match the elbow joint angle of 1 arm with the elbow joint angle of the other. Using an isokinetic dynamometer, Paschalis et al. (7) manually positioned the knee joint at a reference angle and asked subjects to reproduce the reference position. Both studies found disturbed position sense after muscle-damaging exercise (7,9). However, to our knowledge, no study has compared position sense of the arms and legs in the same individuals either at rest or after any intervention (e.g., exercise).
Reaction time of elbows (5) and knee joints (3) has already been separately studied. In addition, it is known that reaction time of the arms to a visual or auditory stimulus is considerably faster compared with the legs (6). However, there are many situations in everyday life in which individuals have to react to stimuli originating from impulses other than visual or auditory; for example, when a person stumbles or tries to catch a “loosing” object. Recently, a new test was developed that detects voluntary movement by measuring reaction angle to release of the knee joint using an isokinetic dynamometer (7). Given that, during the measurement, the limb is attached to the dynamometer, the test is different from free fall tests or movement analyzed by video. In this test, the stimulus originates from the muscle itself, in contrast with the other frequently used tests (6) in which subjects have to respond to a visual or auditory stimulus.
To our knowledge, no study has directly compared position sense and joint reaction angle between the arms and legs. Therefore, the purpose of the present study was to compare position sense and joint reaction angle of the elbow and knee flexors. It was hypothesized that arms would exhibit better position sense and faster reaction angle compared with legs because of neuromuscular differences and distinct levels of participation in everyday life activities.
Experimental Approach to the Problem
The measurement set-up from Paschalis et al. (7) was adopted. The isokinetic dynamometer Cybex Norm (Ronkonkoma, NY, USA) was used. The subjects were told to lie supine (for elbow flexors) or prone (for knee flexors) on the dynamometer, and their position was recorded for the follow-up evaluation (Figure 1). The assessments were undertaken in the dominant arm (elbow flexors) and leg (knee flexors) in random order. The same procedures were followed over 3 consecutive days, and the average of the 3 measurements was used.
Twelve healthy, untrained men (23 ± 1 yr; 177 ± 1 cm; 78 ± 2 kg) were asked to abstain from any strenuous exercise 3 days before and during data collection. The subjects participated in low-intensity leisure activities (such as jogging, swimming, and dancing) 2 to 3 times per week for less than 3 hours per week and had no experience with eccentric exercise training for at least 6 months before the study. Untrained men were used to avoid any influence of exercise history in our measurements. In addition, the subjects were not taking any medication or dietary supplements during the study period. Subjects read and signed an informed consent form according to the standards of the ethics committee of the local university.
During determination of joint angle perception, the limbs were moved passively at a low speed from full extension to 90° flexion to familiarize the subjects with the range of motion. Then, the investigator positioned the limb at each of the 3 reference angles (30°, 45°, or 60° flexion from full extension, in random order), maintained it for 10 seconds, and returned the limb to the initial position (0°). Afterward, the subjects were asked to reproduce the reference position. Subjects actively moved their limb to the target angle, and they held it for approximately 2 seconds. Four efforts were performed, and the 2 best were recorded. The test-retest reliability of the position sense in absolute values was 0.93. The limb was moved from 0° to 90° and then to the target position and back to 0° before each of the 4 efforts.
The “actual angle” was calculated as the difference (degrees in actual values) between the chosen position and the reference angle and indicates whether subjects placed the knee joint in a flexed position or in an extended position relative to the reference angle. The “absolute angle” was the absolute value of the “actual angle” and represents an index of the magnitude of matching error.
Joint Reaction Angle to Release
The limb was passively positioned at 1 of the 3 different angles (20o, 40o, and 60o flexion from full extension) in random order. When muscles of the limb relaxed at the predetermined angle, the investigator let the limb to fall without warning. The angle the limb moved to before the subjects managed to stop the motion was recorded and considered as the joint reaction angle to release. The test-retest reliability of joint reaction angle to release was 0.95. Four trials were performed, and the 2 best were recorded.
The distribution of the dependent variables was examined by Shapiro-Wilk test and was found not to differ significantly from normal. Three-way analysis of variance (ANOVA) was used for the analysis of position sense (limb [arm and leg] × target angle [30o, 45o, and 60o] × angle type [actual and absolute]). Two-way repeated measures ANOVA was used for the analysis of reaction angle to release (limb [arm and leg] × angle [20o, 40o, and 60o]). To determine the meaningfulness of the differences in position sense and reaction angle between the limbs, effect size was calculated as the difference between arms and legs divided by the mean SD of arms and legs for each test. The test-retest reliability of position sense and joint reaction angle to release was determined by performing the intraclass reliability test. The intraclass correlation coefficient was calculated through a random-effect two-way ANOVA model from the values obtained from the 3 consecutive days in 3 subjects. The level of significance was set at p ≤ 0.05.
In regard to position sense (Figure 2), the main effects were significant for limb and angle type (p < 0.001 and p < 0.01, respectively) but not for target angle (p > 0.05). The two-way interactions were significant for limb × angle type (p = 0.012) but not for target angle × limb (p > 0.05) and target angle × angle type (p > 0.05). The three-way interaction target angle × limb × angle type was significant (p = 0.024). The effect sizes between arms and legs of position sense in absolute values were 0.78, 0.83, and 0.77 for 30°, 45°, and 60°, respectively. The effect sizes between arms and legs of position sense in actual values were −0.12, −0.01, and −0.35 for 30°, 45°, and 60°, respectively.
In regard to reaction angle to release (Figure 3), the main effect of limb was significant (p < 0.001) but not the main effect of angle (p > 0.05). The two-way interaction angle × limb was not significant (p > 0.05). The effect sizes between arms and legs of reaction angle values were 0.64, 0.66, and 0.71 for 20°, 40°, and 60°, respectively.
The better position sense of the arms could be partially explained by the higher density of muscle spindles, the better innervation ratio, and more frequent practice of the arm muscles in precise movements during everyday activities than leg muscles (4,8). Indeed, muscles of the arm have higher density of spindles (sensory receptors scattered throughout muscles) than muscles of the leg (8). It is also known that muscles of the arm have a lower innervation ratio (i.e., lower number of muscle fibers are innervated per motor neuron) than muscles of the leg (approximately 1:340 in the arms vs. 1:1900 in legs) (4).
It is known that reaction time of the arms in response to a visual or auditory stimulus is shorter than the leg (6). However, individuals frequently have to react to stimuli originating from impulses other than visual or auditory, when, for example, a person walks on slippery ground. By using the joint reaction angle to release test (7), which intends to measure muscle reaction time to a stimulus originating from the muscle itself, we found that the arms are faster than legs at all released angles. As was the case with the position sense, the faster response of the arms can be partly explained by the higher density of spindles (which also control the speed of muscle) and the lower innervation ratio of the arm muscles compared with legs. In addition, the longer conduction pathway needed for the action potential to reach the leg musclescompared with arm muscles may also contribute to the greater reaction angle of legs. Other potential morphologic differences between the muscles of the arms and legs that may have contributed to the observed differences include density of the Golgi tendon organs and the pressure receptors in the skin.
Different fiber types exhibit large differences in contraction speed (2). Elbow flexors contain less fast-twitch muscle fibers compared with the knee flexors (50% vs. 66%, respectively) (2). Judging solely from the fiber type composition of elbow and knee flexors, we would expect the legs to react more rapidly than the arms. However, the reverse was observed in the present experiment. Therefore, muscle fiber type composition does not appear to be a crucial determinant of the joint reaction speed.
Another possible reason for the difference between arms and legs in position sense and reaction angle is the more intensive use of leg muscles compared with arms in everyday life activities (e.g., walking, going upstairs and downstairs). It is known that intensive muscle use leads to fatigue. Therefore, it is possible that leg muscles are more fatigued in a “resting” state compared with arms. This, in turn, would lead to a slower shortening speed of a leg muscle fiber and slower rise of its tension rate (1). Consequently, the muscles of the lower limb, being in a less fresh state, cannot equal or control the tension compared with the fresh arm muscles in response to limb release and limb placement.
Although the tested joint angles for arms and legs were the same, these do not necessarily correlate with the ranges of movement of the muscles involved. In addition, comparison of movements of different joints is complicated by the different muscle groups acting around joints and, as a result, may contribute differently to the active positioning of a joint. Despite these differences in anatomic arrangements between joints confounding the interpretation of the present findings, they are part of the human musculoskeletal structure and cannot be eliminated during human physiologic experiments. A limitation of the present study is the absence of electromyographic measurements during the joint reaction angle to release that could have provided information on muscle activity onset.
To our knowledge, this is the first direct comparison between the arms and legs regarding position sense and joint reaction angle to release in humans. The main finding is that arms have better position sense and smaller reaction angle to release compared with legs. These results may be partly explained by the higher density of muscle spindles and the lower innervation ratio of the arms compared with legs. Fiber type composition of the arm and leg muscles does not appear to strongly affect reaction angle.
On the basis of the present study, we conclude that an imbalance of the determined relationship between arms and legs in position sense and reaction angle may indicate a neuromuscular disturbance to a sports medicine physician.
1. De Ruiter, CJ, Jones, DA, Sargeant, AJ, and De Haan, A. The measurement of force/velocity relationships of fresh and fatigued human adductor pollicis muscle. Eur J Appl Physiol Occup Physiol
80: 386-393, 1999.
2. Elder, GC, Bradbury, K, and Roberts, R. Variability of fiber type distributions within human muscles. J Appl Physiol
53: 1473-1480, 1982.
3. Escher, SA, Tucker, AM, Lundin, TM, and Grabiner, MD. Smokeless tobacco, reaction time
, and strength in athletes. Med Sci Sports Exerc
30: 1548-1551, 1998.
4. Feinstein, B, Lindegard, B, Nyman, E, and Wohlfart, G. Morphologic studies of motor units in normal human muscles. Acta Anat
23: 127-142, 1955.
5. Ives, JC, Kroll, WP, and Bultman, LL. Rapid movement kinematic and electromyographic control characteristics in males and females. Res Q Exerc Sport
64: 274-283, 1993.
6. Montes-Mico, R, Bueno, I, Candel, J, and Pons, AM. Eye-hand and eye-foot visual reaction times of young soccer players. Optometry
71: 775-780, 2000.
7. Paschalis, V, Nikolaidis, MG, Giakas, G, Jamurtas, AZ, Pappas, A, and Koutedakis, Y. The effect of eccentric exercise on position sense and joint reaction angle of the lower limbs. Muscle Nerve
35: 496-503, 2007.
8. Prochazka, A. Proprioceptive Feedback and Movement Regulation. Handbook of Physiology
. New York: Oxford University Press, 1996.
9. Walsh, LD, Hesse, CW, Morgan, DL, and Proske, U. Human forearm position sense after fatigue of elbow flexor muscles. J Physiol
558: 705-715, 2004.