- Email: [email protected]
Effect of maintained stretch on the range of motion of the human ankle joint
Clinical
UT ERWORT El-LMAd
R FKirsch”,
P L Weiss2, R M Dannenbaum2,
‘Department of Biomedical Therapy, McGill University,
Engineering and *School Montreal, Canada
Vol. 10, No. 3, pp. 166-16X, 19% Copyright 0 19Y5 Elsevier Science Ltd Printed in Great Britain. All rights reserved 026%0033/95$10.00 + 0.00
Biomechnnic.s
R E Kearney’ of Physical
and Occupational
Summary The effectiveness of maintained stretch in expanding the range of motion of the human ankle joint was assessed in a population of normal adults. Controlled movements were imposed upon the ankle, and triceps surae and tibialis anterior electromyograms were monitored to ensure that only passive joint properties generated ankle torque. We found that a majority of subjects (7 of 12) showed evidence of muscle activity sufficient to distort a subjective assessment of changes in range of motion. For the remaining five subjects, a 60-s maintained stretch produced a small decrease in the torque subsequently generated but this effect was transient and largely by an imposed dorsiflexing movement, disappeared following 300 s of rest at a neutral position. This short-term effect is consistent with the viscoelastic properties of collagenous material stretched during such treatment and is unlikely to lead to long-term increases in range of motion. Relevance The results of these experiments indicate that subjective assessments of changes in joint range of motion may be distorted by voluntary and reflexive muscle activation. Moreover the presumed increases in range of motion produced by maintained stretch in our normal subjects were small and transient. These results suggest that future assessments of the long-term efficacy of this treatment must monitor muscle activity and take into account known viscoelastic properties of collagenous materials. Key words: Range of motion, ankle-joint viscoelasticity,
mechanics, shortening
C//n. Biomech.
April
1995,
Vol. 10: 166-168,
Introduction
Stretching of muscle and soft tissue is commonly used in rehabilitation to maintain or increase flexibility and range of motion (ROM) ; slow, maintained stretch to ‘pain-free’ end range is routinely used clinically and is widely accepted as the most effective way to produce a permanent change in the ROM’. Although widely accepted 1 the effectiveness of this treatment in increasing joint ROM has never been verified under controlled experimental conditions. The features of the applied stretch and the resulting joint torques are difficult to control and quantify in most clinical settings, and potentially substantial torques arising from -____
_-._--.----.-
Recei~vd: 9 July IU93 nccepted: 25 h&y 1994 Correspondence and reprint
. ----
~-
requests to: Dr Robert E Kearney, Department of Biomedical Engineering, McGill University, 3775 Oniversity St.. Room 307, Montrkal. Qu&bec H3A 284. Canada
reaction, maintained
stretch, collagen,
voluntary muscle activity or from stretch reflexes or shortening reactions could further distort the perception of treatment efficacy. The objective of this study was to quantify the ability of maintained stretch to increase the passive range of motion of the ankle joint in a group of healthy adults, using appropriate instrumentation and under well-controlled conditions. Methods
An electrohydraulic rotary actuator configured as a position servo was used to impose controlled angular displacements upon the ankle joint of each of 12 normal human subjects, ranging in age from 20 to 60 years and with no previous indication of reduced ROM. In each experiment the subject lay supine on a table with the left foot clamped to the actuator via a rigid glass-fibre boot (which did not restrict the freedom of ankle rotation) such that the actuator axis of rotation was coincident with that of the plantarflexion-
Kirsch
et al.: Effect of maintained
dorsiflexion axis of the ankle. A potentiometer (Beckman 6273-R5K) and a torque transducer (Lebow 2110-5K) built into the apparatus were used to measure angular position and joint torque respectively. Measurement resolution in each case was limited primarily by the quantization of our 16 bit A/D converter, being 0.006 N*m for torque and 0.002 rad for position. Note that plantarflexion torque increases in a negative direction by our laboratory convention. Surface electromyograms (EMG) were recorded differentially from the tibialis anterior (TA) and the triceps surae (TS) muscle groups to verify that muscle activity was absent. The EMG signals were amplified and high-pass filtered at 10 Hz (2 pole Butterworth). The position, torque, TA EMG, and TS EMG signals were low-pass filtered at 40 Hz to prevent aliasing and then sampled at 100 Hz; this sampling frequency was approximately 10 times higher than any movement or torque component observed in our experiments or during normal gait*, and thus was more than adequate to characterize joint mechanical properties and to detect any significant EMG activity. The voluntary, ‘pain-free’ ROM was determined for each subject by having them produce maximum dorsiflexion movements several times with the actuator disabled; the maximum resulting displacement was taken as the voluntary ROM . The neutral position (i.e. 0 rad) was defined as the joint position for which the passive torque was 0 Nm; typically this occurs at 15-20 degrees plantarflexion3. A triangular position command for the actuator was then constructed to move the joint from the neutral position to the voluntary ROM and back at a velocity of 0.03 rad s-l (1.7 deg s-t). The subject was asked to remain completely relaxed during all applied movements, and received visual feedback of TA and TS EMG via an oscilloscope to aid in this requirement. Each experimental trial consisted of three parts, as illustrated by the position recording of Figure l(a).
0.5 0.4 0.3 0.2 0.1 0
(b)
'r
Figure 1. Data recorded during a typical experimental trial of one subject. a Change in ankle joint position (relative to the neutral position) which was imposed in each trial. b Ankle torque produced by the stretch. Note that plantarflexion torques increase in a negative direction.
1
stretch
in the human
ankle joint
167
First a series of three triangular movements, each separated by 5 s, was applied from the neutral (0 rad) initial joint position. During each movement the joint was stretched to the dorsiflexion ROM (slightly more than 0.4 rad for this subject) and held there for 60 s before being returned to the neutral position. Finally a second series of three triangular movements was applied. Each experimental session consisted of two such trials separated by 300 s. Changes in ROM were evaluated by examining the passive torque generated by the fixed amplitude stretch; our analysis therefore consisted in extracting the peak torque response to each triangular movement and comparing these responses for different conditions.
Results
Seven subjects exhibited significant EMG activity in either the TA or TS muscles during the imposed movements which was deemed large enough (more than 2% of that recorded during a maximum voluntary contraction) to have generated an active contribution to the torque record. Our conclusions are therefore based upon the results obtained from the remaining five subjects whose EMG levels remained negligible throughout. Figure 1 illustrates the signals recorded during a typical experimental trial for one of these remaining subjects. The peak torque response (panel b) evoked by the imposed dorsiflexion movement (panel a) decreased slightly from the first to the second and third movement cycles. Joint torque decreased throughout the 60-s ‘hold’ phase, rapidly at first but then more slowly. The peak torques evoked by the subsequent series of three stretch cycles were clearly smaller in magnitude than prior to the maintained stretch. This behaviour was observed in all five subjects, as illustrated in Figure 2. To generate this plot the peak torque values for each movement cycle for each subject were normalized relative to the peak torque value of the first cycle (of the first trial) for that subject; this torque value was the largest of any other cycle for each subject. The peak torque values for each of the 12 movement cycles (6 movements in 2 trials) were then averaged across the five subjects; each point in Figure 2 is thus the average for that cycle across all subjects and the error bars (corresponding to 1 SD) indicate intersubject variability. The peak joint torque evoked by the triangular movements during trial 1 (open squares) decreased slightly but progressively prior to the ‘hold’ (cycles l-3), so that the value for cycle 3 was on average 5% lower than for cycle 1. The torque generated during the first cycle following the hold was decreased on average by an additional 7%) but torque actually increased (on average) somewhat during the two remaining stretches. The decrease in stretch-evoked torque following the 60-s hold, which would presumably correspond to an increase in ROM, was not maintained throughout the 300-s rest period between trials, however, as shown by
166
C/in.
Biomech.
1995;
10: No 3
+ +
,
I
1
1
2
3
I
Trial 1 Trial 2 _
I
I
I
4
5
6
Cycle number FigrtreP. Summary of changes in peak passive torque generated by the stretch for all five subjects. Each point is the average of the five subjects for a particular point during the experiment; prior to averaging each subject’s peak torques were normalized by the peak torque generated by the first stretch. Thus 1 .O is the largest piantarflexion torque and smaller values denote smaller plantarflexion torques. Error bars indicate subject-tosubject variability. The open squares and light line indicate points from the first experimental trial, while the filled squares and heavy line are the points from the second triai. The vertical dashed line indicates that the 60-s ‘hold’ occurred between cycles 3 and 4 in each trial.
the filled squares in Figure 2. The peak torque evoked by the first movement in trial 2 returned, on average, to 95% of the value for the first movement of the first trial. Subsequent effects of the triangular movements and the hold during trial 2 were quite similar (although ‘a bit larger) to those of trial 1. The maximum decrease in peak torque from its initial value (on average across subjects), seen for the first post-hold cycle of trial 2, was approximately 15%. Discussion This study represents the first attempt to define and quantify the effects of a specific stretch protocol on the range of motion of a group of healthy adults. Our results indicate that a single 60-s maintained stretch of the ankle joint produced transient decreases in joint torque, and presumably concomitant increases in ROM. This effect largely disappeared after only 300 s of rest, however, indicating that a single 60-s stretch produced no long-term changes in ROM. It should be noted that these effects were observed with precisely controlled movements and with careful monitoring of the subjects’ EMGs. The latter precaution was particularly important since data from seven of 12 subjects was excluded because of significant EMG activity (usually arising from shortening reactions) in some or all of the movement cycles. The joint torque resulting from this
muscle activity could significantly distort subjective assessments of joint ROM. Our results therefore do not support the hypothesis that maintained stretch effects a permanent change in ankle ROM’. Rather they are consistent with the passive mechanical behaviour of isolated collagenous materials such as tendons, ligaments, fascia, and skin4. Such materials become transiently more compliant following repeated or maintained extension; the gradual decrease in peak torque observed in this study probably resulted from the preconditioning of various connective tissues in the joint and stretched muscles. As illustrated here and elsewhere4 the preconditioning process is easily reversible, so long-term changes in range of motion are unlikely to result from this mechanism. Our findings do not necessarily mean that maintained stretch has no therapeutic value in expanding joint ROM. We did not examine the impact of repeated treatments over long periods of time or the effects of stretching beyond the subjects’ voluntary ROM as is often done in the clinic. Moreover, we examined normal, healthy adult subjects; more substantial changes might occur if this treatment were applied to subjects who have pathology or to a geriatric population.
Couch&on We conclude that the effects of maintained stretch treatment should be assessed only under rigidly controlled experimental conditions, since voluntary and reflexive changes in muscle activation may skew subjective assessments of joint ROM. The viscoelastic properties of collagenous material stretched during such treatment will produce a transient decrease in the resistance of the joint to movement which must be taken into account when assessing long-term changes in joint mobility. Future studies should focus on the systematic identification of movement parameters which will result in maximum long-term increases in ROM.
References 1 Joynt RL. Therapeutic exercise.In: DeLisa JA, ed. Rehabilitation
Medicine
Principles
and Practice.
JB Lippincott, Philadelphia, 1988 2 Antonsson, EK, Mann, RW. The frequency content of gait. J. Biomech. 1985;18: 39-47 3 WeissPL, Kearney RE. Hunter IW. Position dependence of ankle joint dynamics- I. Passivemechanics.J Biomech 1986; 19: 727-35 4 Fung YC. Biomechanics:
Mechanical
Properties
Tissues.Springer-Verlag, New York, 1981
of Living
UT ERWORT El-LMAd
R FKirsch”,
P L Weiss2, R M Dannenbaum2,
‘Department of Biomedical Therapy, McGill University,
Engineering and *School Montreal, Canada
Vol. 10, No. 3, pp. 166-16X, 19% Copyright 0 19Y5 Elsevier Science Ltd Printed in Great Britain. All rights reserved 026%0033/95$10.00 + 0.00
Biomechnnic.s
R E Kearney’ of Physical
and Occupational
Summary The effectiveness of maintained stretch in expanding the range of motion of the human ankle joint was assessed in a population of normal adults. Controlled movements were imposed upon the ankle, and triceps surae and tibialis anterior electromyograms were monitored to ensure that only passive joint properties generated ankle torque. We found that a majority of subjects (7 of 12) showed evidence of muscle activity sufficient to distort a subjective assessment of changes in range of motion. For the remaining five subjects, a 60-s maintained stretch produced a small decrease in the torque subsequently generated but this effect was transient and largely by an imposed dorsiflexing movement, disappeared following 300 s of rest at a neutral position. This short-term effect is consistent with the viscoelastic properties of collagenous material stretched during such treatment and is unlikely to lead to long-term increases in range of motion. Relevance The results of these experiments indicate that subjective assessments of changes in joint range of motion may be distorted by voluntary and reflexive muscle activation. Moreover the presumed increases in range of motion produced by maintained stretch in our normal subjects were small and transient. These results suggest that future assessments of the long-term efficacy of this treatment must monitor muscle activity and take into account known viscoelastic properties of collagenous materials. Key words: Range of motion, ankle-joint viscoelasticity,
mechanics, shortening
C//n. Biomech.
April
1995,
Vol. 10: 166-168,
Introduction
Stretching of muscle and soft tissue is commonly used in rehabilitation to maintain or increase flexibility and range of motion (ROM) ; slow, maintained stretch to ‘pain-free’ end range is routinely used clinically and is widely accepted as the most effective way to produce a permanent change in the ROM’. Although widely accepted 1 the effectiveness of this treatment in increasing joint ROM has never been verified under controlled experimental conditions. The features of the applied stretch and the resulting joint torques are difficult to control and quantify in most clinical settings, and potentially substantial torques arising from -____
_-._--.----.-
Recei~vd: 9 July IU93 nccepted: 25 h&y 1994 Correspondence and reprint
. ----
~-
requests to: Dr Robert E Kearney, Department of Biomedical Engineering, McGill University, 3775 Oniversity St.. Room 307, Montrkal. Qu&bec H3A 284. Canada
reaction, maintained
stretch, collagen,
voluntary muscle activity or from stretch reflexes or shortening reactions could further distort the perception of treatment efficacy. The objective of this study was to quantify the ability of maintained stretch to increase the passive range of motion of the ankle joint in a group of healthy adults, using appropriate instrumentation and under well-controlled conditions. Methods
An electrohydraulic rotary actuator configured as a position servo was used to impose controlled angular displacements upon the ankle joint of each of 12 normal human subjects, ranging in age from 20 to 60 years and with no previous indication of reduced ROM. In each experiment the subject lay supine on a table with the left foot clamped to the actuator via a rigid glass-fibre boot (which did not restrict the freedom of ankle rotation) such that the actuator axis of rotation was coincident with that of the plantarflexion-
Kirsch
et al.: Effect of maintained
dorsiflexion axis of the ankle. A potentiometer (Beckman 6273-R5K) and a torque transducer (Lebow 2110-5K) built into the apparatus were used to measure angular position and joint torque respectively. Measurement resolution in each case was limited primarily by the quantization of our 16 bit A/D converter, being 0.006 N*m for torque and 0.002 rad for position. Note that plantarflexion torque increases in a negative direction by our laboratory convention. Surface electromyograms (EMG) were recorded differentially from the tibialis anterior (TA) and the triceps surae (TS) muscle groups to verify that muscle activity was absent. The EMG signals were amplified and high-pass filtered at 10 Hz (2 pole Butterworth). The position, torque, TA EMG, and TS EMG signals were low-pass filtered at 40 Hz to prevent aliasing and then sampled at 100 Hz; this sampling frequency was approximately 10 times higher than any movement or torque component observed in our experiments or during normal gait*, and thus was more than adequate to characterize joint mechanical properties and to detect any significant EMG activity. The voluntary, ‘pain-free’ ROM was determined for each subject by having them produce maximum dorsiflexion movements several times with the actuator disabled; the maximum resulting displacement was taken as the voluntary ROM . The neutral position (i.e. 0 rad) was defined as the joint position for which the passive torque was 0 Nm; typically this occurs at 15-20 degrees plantarflexion3. A triangular position command for the actuator was then constructed to move the joint from the neutral position to the voluntary ROM and back at a velocity of 0.03 rad s-l (1.7 deg s-t). The subject was asked to remain completely relaxed during all applied movements, and received visual feedback of TA and TS EMG via an oscilloscope to aid in this requirement. Each experimental trial consisted of three parts, as illustrated by the position recording of Figure l(a).
0.5 0.4 0.3 0.2 0.1 0
(b)
'r
Figure 1. Data recorded during a typical experimental trial of one subject. a Change in ankle joint position (relative to the neutral position) which was imposed in each trial. b Ankle torque produced by the stretch. Note that plantarflexion torques increase in a negative direction.
1
stretch
in the human
ankle joint
167
First a series of three triangular movements, each separated by 5 s, was applied from the neutral (0 rad) initial joint position. During each movement the joint was stretched to the dorsiflexion ROM (slightly more than 0.4 rad for this subject) and held there for 60 s before being returned to the neutral position. Finally a second series of three triangular movements was applied. Each experimental session consisted of two such trials separated by 300 s. Changes in ROM were evaluated by examining the passive torque generated by the fixed amplitude stretch; our analysis therefore consisted in extracting the peak torque response to each triangular movement and comparing these responses for different conditions.
Results
Seven subjects exhibited significant EMG activity in either the TA or TS muscles during the imposed movements which was deemed large enough (more than 2% of that recorded during a maximum voluntary contraction) to have generated an active contribution to the torque record. Our conclusions are therefore based upon the results obtained from the remaining five subjects whose EMG levels remained negligible throughout. Figure 1 illustrates the signals recorded during a typical experimental trial for one of these remaining subjects. The peak torque response (panel b) evoked by the imposed dorsiflexion movement (panel a) decreased slightly from the first to the second and third movement cycles. Joint torque decreased throughout the 60-s ‘hold’ phase, rapidly at first but then more slowly. The peak torques evoked by the subsequent series of three stretch cycles were clearly smaller in magnitude than prior to the maintained stretch. This behaviour was observed in all five subjects, as illustrated in Figure 2. To generate this plot the peak torque values for each movement cycle for each subject were normalized relative to the peak torque value of the first cycle (of the first trial) for that subject; this torque value was the largest of any other cycle for each subject. The peak torque values for each of the 12 movement cycles (6 movements in 2 trials) were then averaged across the five subjects; each point in Figure 2 is thus the average for that cycle across all subjects and the error bars (corresponding to 1 SD) indicate intersubject variability. The peak joint torque evoked by the triangular movements during trial 1 (open squares) decreased slightly but progressively prior to the ‘hold’ (cycles l-3), so that the value for cycle 3 was on average 5% lower than for cycle 1. The torque generated during the first cycle following the hold was decreased on average by an additional 7%) but torque actually increased (on average) somewhat during the two remaining stretches. The decrease in stretch-evoked torque following the 60-s hold, which would presumably correspond to an increase in ROM, was not maintained throughout the 300-s rest period between trials, however, as shown by
166
C/in.
Biomech.
1995;
10: No 3
+ +
,
I
1
1
2
3
I
Trial 1 Trial 2 _
I
I
I
4
5
6
Cycle number FigrtreP. Summary of changes in peak passive torque generated by the stretch for all five subjects. Each point is the average of the five subjects for a particular point during the experiment; prior to averaging each subject’s peak torques were normalized by the peak torque generated by the first stretch. Thus 1 .O is the largest piantarflexion torque and smaller values denote smaller plantarflexion torques. Error bars indicate subject-tosubject variability. The open squares and light line indicate points from the first experimental trial, while the filled squares and heavy line are the points from the second triai. The vertical dashed line indicates that the 60-s ‘hold’ occurred between cycles 3 and 4 in each trial.
the filled squares in Figure 2. The peak torque evoked by the first movement in trial 2 returned, on average, to 95% of the value for the first movement of the first trial. Subsequent effects of the triangular movements and the hold during trial 2 were quite similar (although ‘a bit larger) to those of trial 1. The maximum decrease in peak torque from its initial value (on average across subjects), seen for the first post-hold cycle of trial 2, was approximately 15%. Discussion This study represents the first attempt to define and quantify the effects of a specific stretch protocol on the range of motion of a group of healthy adults. Our results indicate that a single 60-s maintained stretch of the ankle joint produced transient decreases in joint torque, and presumably concomitant increases in ROM. This effect largely disappeared after only 300 s of rest, however, indicating that a single 60-s stretch produced no long-term changes in ROM. It should be noted that these effects were observed with precisely controlled movements and with careful monitoring of the subjects’ EMGs. The latter precaution was particularly important since data from seven of 12 subjects was excluded because of significant EMG activity (usually arising from shortening reactions) in some or all of the movement cycles. The joint torque resulting from this
muscle activity could significantly distort subjective assessments of joint ROM. Our results therefore do not support the hypothesis that maintained stretch effects a permanent change in ankle ROM’. Rather they are consistent with the passive mechanical behaviour of isolated collagenous materials such as tendons, ligaments, fascia, and skin4. Such materials become transiently more compliant following repeated or maintained extension; the gradual decrease in peak torque observed in this study probably resulted from the preconditioning of various connective tissues in the joint and stretched muscles. As illustrated here and elsewhere4 the preconditioning process is easily reversible, so long-term changes in range of motion are unlikely to result from this mechanism. Our findings do not necessarily mean that maintained stretch has no therapeutic value in expanding joint ROM. We did not examine the impact of repeated treatments over long periods of time or the effects of stretching beyond the subjects’ voluntary ROM as is often done in the clinic. Moreover, we examined normal, healthy adult subjects; more substantial changes might occur if this treatment were applied to subjects who have pathology or to a geriatric population.
Couch&on We conclude that the effects of maintained stretch treatment should be assessed only under rigidly controlled experimental conditions, since voluntary and reflexive changes in muscle activation may skew subjective assessments of joint ROM. The viscoelastic properties of collagenous material stretched during such treatment will produce a transient decrease in the resistance of the joint to movement which must be taken into account when assessing long-term changes in joint mobility. Future studies should focus on the systematic identification of movement parameters which will result in maximum long-term increases in ROM.
References 1 Joynt RL. Therapeutic exercise.In: DeLisa JA, ed. Rehabilitation
Medicine
Principles
and Practice.
JB Lippincott, Philadelphia, 1988 2 Antonsson, EK, Mann, RW. The frequency content of gait. J. Biomech. 1985;18: 39-47 3 WeissPL, Kearney RE. Hunter IW. Position dependence of ankle joint dynamics- I. Passivemechanics.J Biomech 1986; 19: 727-35 4 Fung YC. Biomechanics:
Mechanical
Properties
Tissues.Springer-Verlag, New York, 1981
of Living