If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Pang MY, Mak MK. Influence of contraction type, speed, and joint angle on ankle muscle weakness in Parkinson's disease: implications for rehabilitation.
Objective
To compare the ankle muscle strength and torque-angle relationship between individuals with Parkinson's disease (PD) and participants without impairments.
Design
Cross-sectional, exploratory study.
Setting
Motor control laboratory in a university.
Participants
Convenience sample of community-dwelling individuals with PD (n=59) recruited from a PD self-help group and age-matched participants without impairments (n=37) recruited from community older adult centers.
Interventions
Not applicable.
Main Outcome Measure
Peak torque and angle-torque profile during concentric and eccentric contraction of ankle dorsiflexors and plantarflexors at 2 different angular speeds (45 and 90°/s).
Results
The PD group displayed lower muscle peak torque values than participants without impairments in all test conditions. Generally, concentric strength was more compromised, with a greater between-group difference (Cohen d=1.29–1.60) than eccentric strength (Cohen d=.81–1.37). Significant group by angular speed interaction was observed in ankle plantarflexion concentric peak torque (P<.001), indicating that muscle weakness was more pronounced when the angular speed was increased. The group by joint angle interaction in concentric contraction of ankle plantarflexors at 90°/s was also significant (P<.001), revealing that the between-group difference in torque values became increasingly more pronounced when the joint was moving toward the end range of the ankle plantarflexion. This exaggerated ankle plantarflexor muscle weakness at the end range was significantly correlated with clinical balance measures (P<.05).
Conclusions
Muscle weakness in PD is influenced by contraction type, angular speed, and joint range. Exaggerated weakness is found in concentric contraction of ankle plantarflexors, particularly when the angular speed is high and the muscle is in shortened lengths.
have shown that more impaired leg extensor strength is significantly related to a longer time taken to complete the Timed Up & Go test. It is thus important to address lower extremity muscle weakness in the PD population.
Weakness in major muscle groups that control the ankle joint, namely the ankle dorsiflexors and plantarflexors, warrants particular attention, because these muscles play an important role in regulating important bodily function such as balance and gait.
Moreover, previous studies did not systematically investigate the relationship between muscle strength and the joint angle. While it is known that force production in normal muscle is influenced by muscle length,
the torque-angle profile in different types of contraction (ie, eccentric vs concentric) and its relationship to contraction speed has not been systematically studied in PD. Examining the torque-angle relationship is clinically relevant, because it helps to identify the joint range at which the torque production is the most deficient, and thus provides important information for the design of an optimal muscle-strengthening protocol for individuals with PD.
The objectives of this study were to compare the isokinetic ankle muscle strength and torque-angle relationship between people with PD and participants without impairments, and to assess the relationship between muscle weakness and balance ability. Several research hypotheses were generated based on previous findings in PD and other patient populations. First, some research evidence has suggested that torque generation may be more compromised at higher contraction speeds in individuals with PD.
Second, studies in older adults and individuals with neurologic pathologies have shown that eccentric muscle strength is better preserved than concentric muscle strength.
The present study was thus designed to test the following hypotheses: (1) there would be a significant group × speed interaction, with the PD group showing more strength deficits at higher movement velocities; (2) eccentric muscle strength would be more compromised than concentric muscle strength in individuals with PD; and (3) there would be a significant group × joint angle interaction, with the PD group showing more strength deficits in the inner range of the muscle.
the comparison of peak torque values of isokinetic ankle dorsiflexion between individuals with PD and participants without impairments yielded effect sizes varying from 0.8 to 2.4 (t test) (ie, large effect sizes), depending on the type of contraction and angular speed. This study involved 2 factors (group and angular speed) for each type of contraction (concentric and eccentric). Based on a 2-way analysis of variance, with a large effect size of .50 and a power of .90, the minimum sample size required would be 64 (32 people with PD and 32 participants without impairments).
A convenience sample of people with PD and participants without impairments was recruited from a local PD patient self-help group and community centers, respectively. The inclusion criteria for participants with PD were: diagnosis by a neurologist using the United Kingdom Parkinson's Disease Society Brain Bank Criteria,
disease duration of >1 year, aged 50 years or older, community-dwelling, able to ambulate for at least 10m with or without walking aids independently, and able to follow simple verbal commands. The exclusion criteria were: significant musculoskeletal conditions that would interfere with testing, diagnosis of other neurologic diseases, and other serious illnesses that precluded participation. The eligibility criteria for the control group were identical as stated above, except that participants without impairments did not have any history of PD. The study was approved by the university's human research ethics review committee. All participants gave written informed consent. All experimental procedures were conducted in accordance with the Declaration of Helsinki.
A total of 59 individuals with PD and 37 participants without impairments were enrolled in the study. Participant characteristics are listed in table 1. Relevant demographic information (eg, medical history) was obtained by interview. The Motor Examination of the Unified Parkinson Disease Rating Scale (UPDRS) III was administered by an experienced clinical researcher in order to assess the degree of motor impairment.
UPDRS Development Committee Unified Parkinson's disease rating scale.
in: Fahn S. Marsden C.D. Calne D.B. Goldstein M. Recent developments in Parkinson's disease. Vol. 2. Macmillan Health Care Information,
Florham Park1987: 153-163
Ankle muscle strength was quantified by an isokinetic dynamometer,a which is capable of maintaining a constant movement speed during testing. Only the more affected side was assessed, because more exaggerated muscle weakness was typically found in the more impaired leg.
Participants were placed in a prone position, and a footplate was attached to their feet while the thigh and lower leg were stabilized by straps. Participants performed isokinetic concentric and eccentric contraction of the ankle between a range of 10° dorsiflexion and 25° plantarflexion at 2 discrete angular velocities of 45°/s and 90°/s. Previous studies employed angular velocities between 30°/s and 180°/s.
To assess the influence of angular speed on torque generation, a higher angular velocity of 90°/s was also chosen. Both velocities were tolerated by individuals with PD, as determined by our pilot study. The order of the testing conditions was randomized. Participants were allowed to practice each type of contraction at their submaximal effort twice, followed by the test trial where 3 maximum concentric or eccentric contractions were performed. Each participant was closely monitored by the researcher during the data collection process, in order to ensure that the participants performed the required movements as instructed. Participants were given a 3-minute rest period between each mode of contraction. The torque (Nm) profiles of the 3 test trials were averaged for further analysis. The variables of interest included the peak concentric and eccentric joint torques, and the torques recorded at different joint angles (in 5° intervals of ankle dorsiflexion/plantarflexion).
Balance Measurements
Stance stability was evaluated by the 1-leg standing (OLS) test.
Each participant was instructed to stand on their more affected leg with eyes open, with hands placed on the hips, and maintain this position for as long as possible. The OLS time was measured using a stopwatch. A practice trial was given before the actual recording.
The limit of stability (LOS) test was performed to assess dynamic postural control, using the Smart Balance System.
The system consists of dual force plates that are connected to a computer system and a screen display placed in front of the participant. Each participant stood on the forceplates while wearing a harness to prevent falls. The theoretical LOS of each participant (ie, the maximum range in which the center of gravity [COG] can be moved safely without changing the base of support) was automatically computed by the Smart Balance System, based on the assumption that movement about the ankle while standing on a fixed surface behaves as an inverted pendulum.
On the screen display, there were 8 target boxes placed at 100% of the theoretical LOS (forward, backward, left, right, forward right, forward left, backward right, and backward left), a center box, and a cursor representing the participant's COG.
In the starting position, the participant was required to maintain the COG cursor within the center box. During the LOS test, the participant was instructed to move the COG cursor toward a highlighted target box as quickly and accurately as possible, and maintain the cursor within the target box. The participants were given a maximum of 8 seconds to complete the movement toward the target. The endpoint excursion is the distance traveled by the COG on the primary attempt to reach the designated target and is expressed as a percentage of the LOS. The endpoint is defined as the point at which the initial movement toward the target stops and subsequent corrective movement begins (fig 1).
Fig 1LOS test. Each participant was asked to move the COG as quickly and accurately as possible toward a second target located at the perimeter of the LOSs. The endpoint excursion refers to the distance traveled by the COG on the initial attempt to reach the target.
All analyses were conducted using SPSS 17.0.c Two-way analysis of covariance (ANCOVA) models with mixed design (within-subject factor: angular speed; between-subject factor: patient group; covariate: age) were used to compare the peak torque values in each of the following types of muscle contraction at 45°/s and 90°/s: concentric ankle dorsiflexion, eccentric ankle dorsiflexion, concentric ankle plantarflexion, and eccentric ankle plantarflexion. Post hoc analysis was performed when appropriate. For analysis of the torque-angle profiles, 2-way ANCOVA models with mixed design (within-subject factor: joint angle; between-subject factor: group; covariate: age) were then used, followed by post hoc analysis as necessary. In isokinetic testing, acceleration and deceleration phases may occur.
Our pilot data indeed showed the existence of the acceleration and deceleration phases within the first and last 2° to 5° of movement, respectively. Therefore, we discarded the data in the first and final 5° of each movement. Only the torque profiles between the range of 5° dorsiflexion and 20° plantarflexion were analyzed. For ANCOVA, effect sizes were expressed in partial eta-squared (large=.14, medium=.06, small=.01).
Finally, Pearson correlation analysis was performed to determine the degree of association of muscle torques with balance parameters measured. A significance level of P<.05 was set.
Results
There were no significant differences in any demographic variables between the 2 groups (see table 1).
Peak Torque
The peak torque data are displayed in table 2. ANCOVA revealed a significant group × speed interaction in ankle plantarflexion concentric strength only (F=11.201, P=.001, effect size in partial η2=.11), indicating exaggerated muscle weakness in concentric ankle plantarflexion among the PD patients when the angular speed was increased from 45°/s to 90°/s. Significant main effects of group for all muscle torque variables measured was found (P<.001), with the PD group consistently showing lower peak torque values than the participants without impairments (partial η2=.16–.38). Significant main effect of speed was detected in concentric ankle plantarflexion only (F=6.665, P=.011, partial η2=.07). In addition, the effect sizes for the concentric conditions were substantially greater than those for the corresponding eccentric conditions, except for ankle dorsiflexion at 45°/s.
Table 2Comparison of the Peak Muscle Torque Between PD and Control Groups
Angle-Torque Profiles Using Absolute Torque Values
When the absolute torque values were used to generate the angle-torque profiles, it was found that the group × angle interaction effect was statistically significant for all test conditions (partial η2=.04–.19) (not shown). The main effects of angle and group were also significant for all test conditions (P<.05). The significant group × angle interaction indicated that the relationship between torque production and joint range demonstrated in the PD group is different from that in participants without impairments.
Angle-Torque Profiles Using Relative Torque Values
To further explore the relationship between muscle weakness and joint angle, the torque value attained at a particular joint angle was expressed as a percentage of the peak torque, and the angle-torque profiles were then compared. For ankle dorsiflexion strength (fig 2), the group × angle interaction was not significant for all test conditions (P>.05) (see figs 2A–D). The main effect of group was significant for concentric contraction at 90°/s (see fig 2B) and eccentric contraction (see figs 2C and D) at both speeds (P<.05, partial η2=.06–.09), but was not statistically significant for concentric contraction at 45°/s (P>.05) (see fig 2A). The main effect of angle was only significant for eccentric contraction (P<.001, partial η2=.05–.06) (see figs 2C and D), but not for concentric contraction (P>.05) (see figs 2A and B), regardless of angular speed.
Fig 2Ankle dorsiflexion angle-torque profiles. On the horizontal axis, negative values represent ankle dorsiflexion, whereas positive values represent ankle plantarflexion. (A) Concentric contraction at 45°/s, (B) concentric contraction at 90°/s, (C) eccentric contraction at 45°/s, and (D) eccentric contraction at 90°/s. The error bars represent one SE of the mean. Between-group difference: *P<.05.
For ankle plantarflexion (fig 3), the group × angle interaction was significant for concentric contraction at both angular speeds (see figs 3A and B) and eccentric contraction at 90°/s (P<.05, partial η2=.03–.19) (see fig 3D). The group × angle interaction for eccentric plantarflexion at 45°/s did not quite reach statistical significance (P=.080). The main effect of group was significant for all test conditions (P<.05, partial η2=.07–.34) (see figs 3A–D). The main effect of angle was significant for eccentric ankle plantarflexion (P<.05, partial η2=.49–.56) (see figs 3C and D), but not for concentric ankle plantarflexion (P>.05) (see figs 3A and B), regardless of angular speed.
Fig 3Ankle plantarflexion angle-torque profiles. (A) Concentric contraction at 45°/s, (B) concentric contraction at 90°/s, (C) eccentric contraction at 45°/s, and (D) eccentric contraction at 90°/s. The same convention was used as in figure 1. Between-group difference: *P<.05; †P<.001.
Concentric ankle plantarflexion strength at 90°/s at 20° of ankle plantarflexion was selected for subsequent correlation analysis, because it showed the most pronounced deficit (see fig 3B). The results showed that it was significantly correlated with OLS time (r=.306, P=.022), and endpoint excursion (LOS test) (r=.388, P=.003).
Discussion
This novel study found that the PD group displayed significant weakness in both ankle dorsiflexors and plantarflexors. The weakness is particularly apparent during concentric contraction of the plantarflexors, when the angular speed is high and the joint is moving toward the end range of ankle plantarflexion.
Ankle Muscle Weakness in PD
Our findings of muscle weakness in both concentric ankle dorsiflexion and plantarflexion in the PD group are consistent with those reported by previous studies.
were the only studies that demonstrated eccentric ankle dorsiflexors weakness in individuals with PD. Our study extended their finding by showing that both eccentric ankle dorsiflexion and plantarflexion joint torques were compromised in PD.
Influence of Speed
Our results proved our hypothesis that individuals with PD have more muscle strength deficits at higher movement speeds. With an increase in angular speed, the difference in peak torque generated during concentric ankle plantarflexion contraction between the patients and participants without impairments became more conspicuous. Pedersen and Oberg
reported that the deficits in peak torque generation were more apparent at angular speed of 180°/s than 30°/s for concentric ankle dorsiflexion. Our lack of group × speed interaction effect in concentric ankle dorsiflexion could be because of the narrower range of angular speeds used in our study (ie, from of 45°/s to 90°/s), compared with those used by Pedersen and Oberg.
However, these investigators compared the joint torque at much higher velocities, at 90°/s and 150°/s.
The impairment in torque generation at higher velocities is considered to represent bradykinesia, a symptom that reflects the dysfunction of central mechanisms. Hallett and Khoshbin
demonstrated that bradykinesia resulted from the inability to activate the appropriate muscle to generate force at a sufficient rate. Individuals with PD may have selective decrease in the number of and atrophy of fast-twitch type II muscle fibers, and hence they have more weakness during movements at faster speeds.
Our results support our hypothesis that concentric strength is more impaired than eccentric strength in individuals with PD. By comparing the effect sizes (see table 2), we found that eccentric strength was better preserved than concentric strength, particularly for ankle plantarflexors. Pedersen et al
reported a similar phenomenon in ankle dorsiflexors. The more pronounced deficits in concentric strength in the PD population appear to be consistent with those reported in older people and individuals with other neurologic conditions (eg, cerebral palsy, stroke).
The mechanisms underlying this phenomenon are not entirely clear. Presumably, physical inactivity may lead to reduction in both concentric and eccentric strength. However, a sedentary lifestyle may also contribute to a decrease in contractile element and an increase in connective tissue content in the muscle, which may alter the mechanical stiffness of the muscle. During eccentric contraction, the stiffness of the lengthening muscle would contribute to tension development. The rigidity may also contribute to the force development, because the muscle is lengthening during eccentric contraction. It is possible that the reduction in strength because of eccentric contractile inactivity may be partially compensated by gains in mechanical stiffness.
Further studies are necessary to examine the muscle structural changes in PD.
Influence of Joint Angle
In analyzing the angle-torque profiles, we used the relative torque values in conjunction with the absolute values. The results revealed substantial difference in peak torque values between the PD and control groups (see table 2). The peak torque value for a particular type of contraction also varied across the subjects within each of the 2 groups, as reflected by the SDs (see table 2). Additionally, the peak torque values also differed considerably among the different types of contraction (see table 2). By expressing each individual's torque values generated at various joint angles as a percentage of his or her own peak torque value, the variability of the peak torque arising from different sources (ie, between-group, within-group, between-contractions) could be taken into account when analyzing the angle-torque profiles. Using the relative torque values can thus facilitate the comparison of angle-torque profiles between the PD and control groups across the different test conditions. Previous studies in other patient populations have also used a similar approach in the analysis of muscle torque data.
In analyzing the absolute torque values, the group × angle interaction was statistically significant for all test conditions, clearly indicating that the relationship between torque generation and joint angle demonstrated in the PD group is distinct from that in the control group. In subsequent analysis using relative torque values, the significant group × angle interaction was found only in ankle plantarflexion. In particular, the PD group exhibits exaggerated muscle weakness in the inner range of concentric ankle plantarflexion, which is more apparent at higher speeds (see fig 3B). At 90°/s, the relative ankle plantarflexion joint torque in the PD group is at its maximum at the outer range of the muscle (ie, when the ankle is in a dorsiflexed position) but is substantially reduced and reached its minimum at the inner range (ie, when the ankle is in a plantarflexed position). Our findings thus support our hypothesis that muscle weakness is more pronounced in the inner range of the muscle. Our results are also consistent with the exaggerated weakness found in shorter elbow flexors and extensors
among patients with stroke. The length-dependent deficits might be because of impaired motor unit rate coding at shorter muscle lengths. It has been shown in neurologically intact participants that the twitch duration is reduced during voluntary isometric contraction at a shorter muscle length.
Participants without impairments could increase the motor unit firing rates to maintain the joint torque. People with PD may have central deficits in maintaining or sustaining the motor units firing rate for a long period of time,
resulting in exaggerated weakness of ankle plantarflexors at shorter lengths.
We note that the length-dependent deficit was only present in ankle plantarflexors, but not ankle dorsiflexors. Individuals with PD are known to walk with shuffling gait and with toes touching the ground first instead of the heel. Kinematic and kinetic analysis found that these patients had decreased ankle plantarflexion excursion
The disuse of the ankle plantarflexors, especially in the inner range, during daily functional activities might result in more deficits in this range.
The length-dependent weakness found here could not be explained by the reduced effort in anticipation of reaching the end of movement. First, this finding is not observed in the participants without impairments (ie, significant group × angle interaction) (see fig 3B). Second, this pattern of muscle weakness is not found in the ankle dorsiflexors (see fig 2) or eccentric contraction of ankle plantarflexors (see fig 3D).
Clinical Implications
Muscle weakness in individuals with PD may have important functional implications. Indeed, we found a significant relationship between ankle plantarflexor muscle weakness and balance ability. Addressing muscle weakness is thus an important area in fall management of these patients. The findings in this study would be useful in guiding the design of a muscle-strengthening program for the PD population. For example, because muscle strength is more compromised at higher velocities and inner range of movement, strength training may focus on higher-velocity movements that force the patients to use the inner range of the target muscle group. It has been reported that after speed-specific and angle-specific isometric and concentric training in unimpaired individuals, the strength gain was more apparent at the trained speeds and joint range.
People with PD have more impairment in concentric than eccentric ankle muscle strength. On the one hand, it is important to target this deficit in concentric contraction in the strength-training program. On the other hand, the strengthening program may exploit their better-preserved eccentric strength to maximize functional capacity. Indeed, Dibble et al
have shown that high-force eccentric training resulted in better outcomes in muscle volume, walking endurance, and stair descent in people with PD. Further study is required to investigate the effects of different exercise protocols on the torque-angle relationship and function in people with PD.
Study Limitations
This was a cross-sectional study that compared the muscle strength between people with PD and age-matched participants without impairments. It could not provide information on the temporal changes of muscle strength in people with PD. A prospective study would be required to examine the degree of muscle strength after PD as time progresses. All assessments were performed within 1 hour during the on phase of the medication cycle. Further study should investigate the muscle strength profiles during the off phase. Additionally, our participants in the PD group were all community-dwelling, ambulatory individuals. The results cannot be generalized to those who are institutionalized or wheelchair bound. Finally, all participants in the PD group were recruited from a local PD patient self-help group, which held regular meetings for its members. The PD group in our study may thus be more physically and socially active than their nonmember counterparts.
Conclusions
Individuals with PD demonstrate significant ankle muscle weakness. The deficit in ankle muscle force production is influenced by contraction type, angular speed, and joint range. The results are useful in guiding the design of a muscle strength-training program for individuals with PD.
in: Fahn S. Marsden C.D. Calne D.B. Goldstein M. Recent developments in Parkinson's disease. Vol. 2. Macmillan Health Care Information,
Florham Park1987: 153-163
Supported by The Hong Kong Polytechnic University (grant no. G-SAA6 ), and the SK Yee Medical Foundation (grant no. ZH61 ).
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.
Reprints are not available from the author.
In-press corrected proof published online on Aug 2, 2012, at www.archives-pmr.org.
Significant difference between PD patients and control subjects (P<.001).
00409-1&id=gr1.jpg){kind=link}
00409-1&id=gr2.jpg){kind=link}
00409-1&id=gr3.jpg){kind=link}