Muscle force production and functional performance in spastic cerebral palsy: Relationship of cocontraction☆☆☆★★★
Received: July 29, 1999; Accepted: October 19, 1999;
Abstract
Damiano DL, Martellotta TL, Sullivan DJ, Granata KP, Abel MF. Muscle force production and functional performance in spastic cerebral palsy: relationship of cocontraction. Arch Phys Med Rehabil 2000;81:895-900. Objective: To determine cocontraction's relation to strength and motor function in children with spastic cerebral palsy (CP). Design:Prospective evaluation with a convenience sample of 10 subjects. Setting: Pediatric rehabilitation center at a tertiary care hospital. Patients: Ten ambulatory children with spastic CP, mean age 5 to 14yrs. Main Outcome Measures: A single comprehensive assessment of hamstring and quadriceps muscle strength; gait analysis while monitoring electromyographic (EMG) activity in those muscles; administration of the Gross Motor Function Measure (GMFM); heart-rate monitoring during quiet rest versus gait to compute an energy expenditure index (EEI). Cocontraction ratios and magnitudes were determined for the gait and strength testing trials using the EMG data. Results: Cocontraction ratios during strength tests correlated directly with those during free gait. Cocontraction magnitude and total EMG magnitude had an inverse relationship to EEI; children with more muscle activity in the agonist and antagonist tended to be more energy efficient. Knee extensor muscle strength correlated positively with the GMFM and gait velocity. Neither cocontraction ratio nor magnitude during gait was related to strength. Conclusions: Children with CP used a similar muscle activation strategy across two different motor tasks. Strength and cocontraction were uniquely related to different aspects of motor function. Further research is needed to quantify more precisely cocontraction and force to EMG relations in this population. © 2000 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
SPASTIC CEREBRAL PALSY (CP), the result of an insult to an immature central nervous system, has a primary adverse effect on motor coordination. The incidence of CP is approximately 2 to 3 cases per 1000 live births, with the spastic form of the disorder the most prevalent.1 Movement patterns in CP are typically characterized by slow, stiff movements.2However, research increasingly shows that spasticity is not the sole culprit in producing the motor dysfunction of CP;3transformation of the muscle fibers and abnormal dynamic muscle activation patterns, such as excessive cocontraction and diminished agonist force, may play a more significant role.4, 5, 6
Muscle cocontraction can be defined as the simultaneous activation of agonist and antagonist muscle groups crossing the same joint and acting in the same plane.7 Mechanically, this activation pattern increases joint stiffness while limiting agonist force production.8 Despite its inherent inefficiency, cocontraction is a common motor control strategy primarily activated when a person needs increased joint stability or improved movement accuracy.9 In any given task, however, a point can be reached at which excessive use of cocontraction begins to impair movement.10 Increased cocontraction has been shown qualitatively during gait in children with CP.2 Although researchers have not yet precisely quantified cocontraction during movement in persons with CP, the degree of antagonistic coactivity has been estimated mathematically in this population using several different computational methods. Some authors have computed ratios of antagonist-to-agonist activity between opposing muscles11 or within the same muscle across tasks.12, 13 Some have based the ratios on antagonist-to-total (agonist + antagonist) activity.14 Still others have computed the area of overlap of the electromyographic (EMG) values over a specified time period for the two opposing muscle groups.15 Because the amplitude of the EMG signal cannot be equated directly with the absolute level of muscle force, EMG amplitude is typically scaled to activation levels recorded during an isometric maximum or a specified steady-state submaximum, if available, or to the maximal or mean value of the EMG signal during a given task. However, the use and method of normalization may have a differential effect on the results obtained when comparing muscles with varying degrees of strength (or weakness) within a person or across subject groups. Because of this potential for distortion in cocontraction, several authors16, 17have chosen not to scale the EMG values, and, based on their results, some16 have suggested that antagonist restraint, or cocontraction, may not be excessive in patients with spastic disorders.
Increased cocontraction is presumably detrimental to motor performance in CP; however, few researchers have documented the precise negative functional implications of using it excessively. Recently, Unnithan and colleagues15showed a direct relationship in CP between greater cocontraction magnitudes, scaled to individual EMG maximal values during either gait or an isometric maximum, and higher energy costs during treadmill walking. Increased antagonist restraint (cocontraction) may also possibly explain the pervasive weakness documented in spastic CP.6 Quantification of cocontraction during knee flexion and extension maximum voluntary contractions (MVC) revealed that children with CP had significantly higher cocontraction ratios than normal during knee extension, but not flexion.18 In a similar study on adults with hemiparesis, cocontraction during an isometric maximal effort was strongly and inversely related to ankle dorsiflexion strength (r = −.91) in the affected extremity, but was not related to plantarflexor strength.14
The primary goal of the present investigation was to decipher the role of muscle cocontraction in force production and functional motor performance in children with spastic CP. In particular, we correlated muscle activation strategies of the knee joint musculature across two tasks, maximal force production and over-ground walking. We also explored the interrelation between cocontraction and weakness, their relation to temporal-spatial and kinematic gait parameters, and to the Gross Motor Function Measure (GMFM) and the energy expenditure index (EEI).
We hypothesized that the motor control strategy used across tasks would be consistent within individuals with CP, ie, those who had greater cocontraction during force production would also have higher levels of cocontraction during walking. We also hypothesized that the level of cocontraction and the degree of weakness would be directly related. Based on the strong relation we found previously between muscle weakness in CP and motor function,19 we believed that excessive cocontraction would adversely affect functional performance—children with greater cocontraction would have poorer motor performance. A second goal was to understand better the potential for variable results and clinical interpretations attributable to scaling and computational methods.
Methods
Subjects
Ten children with spastic CP ranging in age from 5 to 14 years (mean, 9.2yrs) agreed to participate in the study (table 1), and informed consent was obtained from each child's parent or legal guardian.
| Participant | Age (yrs) | Type CP | Ambulatory Level | Surgical History |
|---|---|---|---|---|
| 1 | 8 | H | I | — |
| 2 | 6 | D | III | GS, ST stapling |
| 3 | 12 | D | I | Hamstrings |
| 4 | 9 | D | III | SDR |
| 5 | 9 | D | II | — |
| 6 | 10 | D | II | IL, Add, Hamstrings |
| 7 | 6 | D | III | Add, Gracilis, Hamstrings, GS |
| 8 | 5 | D | I | — |
| 9 | 14 | D | III | Add, Hamstrings, PTT |
| — | ||||
| 10 | 6 | D | III | |
Ambulatory levels: I, independent ambulator; II, can ambulate independently, but requires assistance some of the time; III, requires assistance to ambulate at all times.
Abbreviations: H, hemiplegia; D, dyslegia; GS, gastrocnemius-soleus recession; ST, subtalar; IL, iliopsoas; PTT, posterior tibialis transfer.
Procedures
Each child underwent bilateral isometric strength testing of the quadriceps and hamstring muscles, and had 3-dimensional gait assessment at a freely selected speed and a fast speed. Muscle activity was monitored by means of surface EMG with the same electrodes in place for the strength-testing and gait trials. Each child was administered the GMFM. Reliable EEI data were obtained on only 5 children; 2 refused, 1 was unable to cooperate, and 2 were unable to complete the assessments.
After the skin was prepared by vigorous rubbing with an alcohol pad, bipolar surface EMG electrodes were placed over the muscle bellies of the rectus femoris and the biceps femoris bilaterally with interelectrode distance not greater than 1cm (to minimize cross-talk). The EMG data for the ipsilateral quadriceps and hamstrings muscle groups were used to compute cocontraction values for each leg.
Isometric strength tests were performed in the order of right and left quadriceps, then right and left hamstrings. All tests were performed with the subject in a supported sitting position with the hips and knees in 90° of flexion. A hand-held dynamometera was used for all strength measurements. A curved, padded bar on the end of the dynamometer was placed perpendicular to the long axis of the tibia as far distally as possible without constraining the anterior tibialis or tendoachilles tendons. The bar was placed against the anterior shin for the knee extension tests, and against the posterior part of the calf for the knee flexion tests. The examiner held the device rigidly in place while the child was encouraged to push “as hard as possible” for approximately five seconds and then instructed to relax. Two trials were performed for each muscle test, and the peak force was recorded for each exertion from the dynamo meter display. The children performed two maximal exertions, resting between exertions to minimize fatigue effects. The trial in which they produced the higher force was used in the analyses. To maximize reliability, the same examiner performed all strength assessments, and a “make” test was performed in which the examiner held the device rigidly in place during the test, keeping the test position constant while the child exerted force. The dynamometer was linked to the computer so that the time course of the force output from the dynamometer was captured concurrently with the EMG output.
After taking the strength measurements, the investigator placed 15 retroreflective markers over the following anatomic locations on the pelvis and lower extremities, as specified for the Vicon Clinical Manager software,b version 1.21: the anterior superior iliac spines, the sacrum, the lateral aspect of the knee joints, the lateral malleoli, the heels, the 2nd metatarsals, and the lateral aspects of the thigh and calf segments. Three-dimensional gait analyses were performed using a six-camera Vicon 370 system.b Children were asked to walk barefoot down a 12-meter carpeted walkway at a freely selected speed, then were instructed to walk “as fast as possible without running.” Although each child performed multiple walking trials, only a single trial was analyzed for each condition—the median trial with respect to velocity at the free speed and the fastest of the fast walking trials. The temporal-spatial data were processed using Vicon Clinical Manager software,b version 1.21. The measurements analyzed from the 3-D gait assessment were free and fast velocity, stride length, cadence, and sagittal plane total knee excursion during free speed gait across a gait cycle (from right-foot contact to the next consecutive right-foot contact).
The GMFM was administered to quantify further the functional capabilities of the child on a variety of motor tasks typically encountered in daily life. The tasks ranged from rolling to running and jumping with the difficulty level such that a normally developing 5-year-old would score 100%.20 The EEI was obtained on the 5 children who agreed to and were able to complete this portion of the assessment. Each child was fitted with a Polartek heart rate monitor,c rested quietly for 5 minutes, then walked at a self-selected (free) speed for 5 minutes. Heart-rate data were recorded during the last minute of the resting period and the last minute of the walking period to ensure steady-state heart-rate values for each condition. Lap times and distance traveled during walking were recorded to compute mean velocity. The EEI estimates relative energy expenditure by comparing the differences in steady-state heart rate during rest and during a walking trial divided by the child's mean velocity.21 Both the GMFM and the EEI have been shown to be reliable and valid measures of motor performance in CP.20, 21
Data analysis
EMG and force data for the strength and gait trials were collected concurrently and were filtered with a low-pass root mean squared (RMS) filter with a cut-off frequency of 5Hz. EMG data for the gait trials were also filtered with a low-pass RMS filter, with a cut-off frequency of 15Hz. To estimate the degree of cocontraction at maximal force production during the isometric tests, a window of 90% to 100% MVC was chosen, and the corresponding EMG values were extracted to compute the cocontraction values. We chose this window, rather than the point of maximal force, to minimize the effects of (1) data spikes that are commonly seen at or near the point of maximal effort, and (2) the present but variable time shift in force maximum compared with EMG maximum that is caused by electromechanical delay. The cocontraction during strength tests and gait was quantified in two ways. First, a cocontraction ratio (CCR) was computed by dividing the minimal EMG value by the maximal EMG value to obtain a ratio at each time point, and then computing a mean ratio for the entire trial. In the strength trials, the maximal value was always the agonist, and the minimum was always the antagonist. In the gait trials, however, the minimal and maximal values alternated between the hamstrings and quadriceps throughout the cycle. Second, we approximated the cocontraction magnitude (CCM) by computing the mean value of the area of overlap (the EMG minimum) of the linear envelopes of the two EMG signals. Before computing CCM, we scaled the EMG values to their respective isometric maximums, then weighted them by the degree of weakness in each muscle, compared with normal pediatric strength values.6 This computing sequence reduced the effect of strength differences from normal values and from between-muscle groups on the cocontraction computation; we could more realistically estimate the force negated by antagonistic activity during gait because the activation level of a weaker agonist was not distorted.
We also computed ratios of antagonist-to-agonist muscle activity within a muscle group across the two isometric tests. For example, the hamstring activity during the quadriceps maximum test was divided by its activity during the hamstring maximum test. This method has been used as a cocontraction value in previous studies.12, 13
We used a general linear model in the statistical analyses. Simple linear regression was used to assess the correlation of cocontraction values during strength tests to those during gait, and to assess the relation between cocontraction and strength, respectively, to the remaining functional assessment measures. Repeated-measures analysis of variance (ANOVA) was used to assess differences between cocontraction ratios and magnitudes across tasks or muscle groups within subjects. A probability value below.05 indicated statistical significance.
Results
The mean and standard deviation for each analyzed para meter are shown in table 2.
| Parameter | Mean | SD |
|---|---|---|
| CCR | ||
| Knee extension (H/Q EMG) | .53 | .27 |
| Knee flexion (Q/H EMG) | .43 | .26 |
| Free gait (Min/Max EMG) | .55 | .11 |
| Fast gait (Min/Max EMG) | .54 | .15 |
| CCM | ||
| Knee extension (Min EMG)* | 32.1 | 18.7 |
| Knee flexion (Min EMG)* | 16.7 | 11.9 |
| Free gait (Min EMG)* | 37.8 | 16.8 |
| Fast gait (Min EMG)* | 57.8 | 23.7 |
| Antagonist/Agonist Ratios | ||
| Hamstring | 1.17 | 1.34 |
| Quadriceps | .28 | .13 |
| % Normal strength | ||
| Knee extensor | .63 | .27 |
| Knee flexor | .38 | .26 |
| Gait Parameters | ||
| Free velocity (m/sec) | .58 | .18 |
| Fast velocity (m/sec) | .95 | .34 |
| Cadence, free speed (steps/min) | 105.5 | 30.8 |
| Stride length, free speed (m) | .69 | .15 |
| Participant Age (yrs) | 9.2 | 3.0 |
| GMFM Total | 77.4 | 13.6 |
| EEI | 2.6 | 1.12 |
| *EMG scaled to isometric maximum, then weighted by % normal strength. | ||
Abbreviations: H, hamstrings; Q, quadriceps; Min, minimal value; Max, maximal value.
Table 3 lists the significant bivariate correlations (p <.05) obtained when comparing cocontraction and strength values to each other and to the other functional parameters measured.
| Comparison | r |
|---|---|
| Knee Extension CCR with Free Gait CCR | .60 |
| Knee Extension CCR with Fast Gait CCR | .35 |
| Knee Flexion CCR with Free Gait CCR | .46 |
| Knee Flexion CCR with Fast Gait CCR | .44 |
| Free Gait CCR with Fast Gait CCR | .68 |
| Free Gait CCM with Fast Gait CCM | .72 |
| Free Gait CCM with EEI | −.71 |
| Fast Gait CCM with EEI | −.90 |
| Free Gait EMG Max with EEI | −.59 |
| Fast Gait EMG Max with EEI | −.74 |
| Free Gait EMG Max to EMG Min (CCM) | .79 |
| Knee Extensor Strength with Free Gait Velocity | .68 |
| Knee Extensor Strength with GMFM | .57 |
| Comparison | F | p |
|---|---|---|
| Knee Flexion CCR vs Free Gait CCR | 6.1 | .02 |
| Knee Flexion CCR vs Fast Gait CCR | 5.0 | .04 |
| Knee Flexion CCM vs Free Gait CCM | 30.3 | <.01 |
| Knee Flexion CCM vs Fast Gait CCM | 71.1 | <.01 |
| Knee Extension CCM vs Fast Gait CCM | 15.5 | <.01 |
| Free Gait CCM vs Fast Gait CCM | 23.7 | <.01 |
| Antagonist/Agonist Ratio, Knee Flexors vs Extensors | 7.85 | .01 |
The CCM was also significantly lower for the hamstrings strength test than for free or fast gait; lower for the quadriceps test than for fast gait; and lower for free speed gait as compared with fast (table 4). This last result indicates that while theproportion of antagonistic coactivity did not increase across gait speeds, the actual amount of muscle activity exerted to oppose the agonist was greater when the child walked faster.
Neither cocontraction ratios nor magnitudes were related to normalized knee flexor or extensor strength expressed as a percentage of pediatric norms. Weaker children did not use more, or less, cocontraction. Thus, our findings did not support the second hypothesis, which proposed that such a relationship existed. Nor was cocontraction related to functional level as indicated by the GMFM total score or temporal-spatial gait parameters, or to knee excursion during gait. CCM during free speed gait was most strongly related to the EEI. Additionally, a strong relationship was found between the EEI and maximal EMG during gait (table 4), indicating that the level of agonist muscle activity is also related to energy expenditure.
Muscle strength was related to different aspects of performance than cocontraction, although the strength-to-function relation existed only for the quadriceps muscles, not the hamstrings. Knee extensor strength correlates positively with parameters such as the GMFM and gait velocity, as shown in earlier investigations.19 In contrast to EMG values, normalized strength is not related to EEI.
Antagonist-to-agonist ratios computed within muscles during the strength tests were markedly greater in both magnitude and variability for the hamstrings than for the quadriceps. In fact, the mean for the hamstrings ratio was greater than 1.0, with 6 of the 10 subjects showing greater antagonist than agonist activity in the hamstrings in at least one extremity.
Discussion
Without being able to place force transducers in the muscles, precise quantification of cocontraction in a dynamic situation is an elusive goal. All reported values of cocontraction during movement, regardless of computation method, are at best estimations, each accompanied by a set of physiologic assumptions—some of which may not be valid in persons with abnormal muscle coordination. For example, in our scaling method, we assumed that a linear relationship existed between force and EMG in the hamstrings and quadriceps muscles. Woods and Bigland-Ritchie22 supported this assumption, and showed that in large muscles, at isometric force levels greater than 30% to 40%, the force-EMG relationship does indeed appear to be linear. Although we used this assumption in the present study, these relationships have neither been validated nor refuted in persons with spasticity, and more research in this area is needed.
EMG magnitude can be compared reliably in a relative manner within muscles with the same electrodes in position during the different conditions, while scaling methods or mathematical models for estimating absolute magnitude, as discussed, are more problematic. For these reasons, CCRs are reported more commonly than the amount of antagonistic force. The limitation of reporting ratios alone is that the ratio of cocontraction at low and high force levels could be equivalent, while the actual amount of restraint could vary drastically with very different functional implications for the individual. Also, a high ratio may not signify an excessive contribution by an antagonist, but rather an agonist deficiency. Research suggests that the primary reason for muscle weakness in CP is not antagonist restraint from spasticity or cocontraction, but a primary inability of the agonist to produce sufficiently high force levels.3, 18 Researchers16 examining patients with adult-onset spasticity have reinforced this finding of agonist deficiency and have not documented increased antagonistic restraint.16High ratios of cocontraction in the CP population may also have been mistakenly computed and interpreted.
To estimate magnitudes, some authors scale EMG values to those obtained during an isometric exertion. In a person with normal strength and the ability to isolate effectively the agonist during a maximal effort, this procedure may provide a reasonable approximation, but we believe it is not appropriate in persons with CP who are grossly unable to produce a sufficient level of muscle activation to achieve a normal level of force, a factor that can also vary across muscles within the same individual. Scaling EMG activity to an isometric value alone in the CP population would tend to exaggerate the contributions of weaker, more spastic, or less effective muscles. For example, in the present investigation, the hamstrings had significantly lower isometric force relative to normal strength values than the quadriceps. The hamstring muscles in CP show clear differences from the quadriceps muscles6 and from hamstring muscles in children without CP18—their antagonist activity is relatively greater than their agonist activity. Consequently, when we scaled EMG values of the hamstrings and quadriceps during gait to an isometric maximum in the present investigation, the adjusted hamstring activity consistently surpassed the quadriceps activity, so that the cocontraction value was nearly equivalent to the mean quadriceps value across the cycle. We then decided to weight the scaled values by the degree of weakness in each muscle group. Once this was done, the area of overlap during gait (the CCM) was represented by alternating contributions of both muscles, rather than by a single muscle. This method could also provide a rough estimate of the actual negative contribution of the antagonist to net force output (fig 1).
Fig. 1
The EMG data for the hamstrings and quadriceps muscle groups and the area of overlap (shaded) across a gait cycle for a single subject. (A) The area of overlap when each muscle is unscaled and expressed in millivolts; (B) the same data scaled to the respective percentage of the maximal voluntary contraction value for each muscle; (C) the same data scaled to the percentage maximal voluntary contraction as in (B) and then weighted by the estimated degree of weakness. Rquad, right quadriceps; Rham, right hamstring.
The main goals of the present study were to examine further the relationships between muscle cocontraction and isometric strength in CP, and to elucidate more precisely the role of cocontraction in functional motor performance in CP. As hypothesized, the proportion of cocontraction was related across tasks—children who had higher ratios during the strength tests also tended to have higher ratios during gait. The relationship of quadriceps strength to functional gait velocity and the GMFM in CP has been shown previously,19 and was further supported by our present results. The lack of a relationship between cocontraction and level of function as measured primarily by the GMFM and free speed gait velocity is somewhat surprising. One might suspect that because excessive cocontraction is considered an abnormal muscle activation pattern, children with greater neurologic involvement would have more cocontraction. This was not the case in the present investigation, and was perhaps related to our sample's limited spectrum of functional involvement. However, similar findings have been reported in patients with other central nervous system disorders.23
The lack of a relation between strength and cocontraction appears to be counterintuitive, because antagonist activity could diminish the net force contribution of the agonist, unless the agonist simultaneously increased its activity. However, in a previous study18 quantifying the effect of cocontraction on force production in CP, weakness in the knee flexors was not from a difference in the levels of cocontraction. In the knee extensors of children with CP who had normalized torque values 47.5% of age-matched peers, the cocontraction levels were only 7.3% greater than normally seen. This finding indicated that cocontraction had a minimal effect at best on the ability to produce maximal torque in the knee extensors, with virtually no effect on knee flexor torque. Muscle force and muscle cocontraction can be regulated independently in persons with normal motor control,24 and these results suggest that this ability is also possible, and even probable, in persons with CP.
Cocontraction increases joint stiffness, which makes movement more laborious.24 However, even if cocontraction levels were excessively high, force production levels could be maintained if the agonist force increased concurrently with antagonist restraint. As a consequence, the joint would become incrementally stiffer, which may or may not be desirable depending on the task. In contrast to Unnithan's15 finding of a direct relationship between cocontraction and oxygen cost, in this sample those who had higher unscaled EMG values whether in the agonist or antagonist (CCM) tended to be moreenergy efficient. The differences between studies could be attributed to a spurious effect in our small sample size or to the different study design or scaling methods used. In their study the cocontraction magnitude was based on percentage of the isometric or gait maximum, and they compared treadmill walking, looking across rather than within populations. Another possible explanation is that since the EEI measures energy cost rather than expenditure, the true relationship of antagonist muscle activity to energy output may have been obscured by the velocity denominator in the EEI or by the colinearity of muscle force and velocity.
The greater magnitude of cocontraction seen during free and fast speed gait as compared with the hamstring strength testing may be explained in part by the inherently different nature of each task. Walking necessarily uses both muscle groups, sometimes acting in concert at different times, independently or in opposition, while in maximal isometric exertions, the most effective strategy would be to activate the agonist muscle group primarily, if not exclusively. Also, the postural requirements for maintaining upright stability are greater while walking, so more cocontraction would be expected even in a normal motor control system.9
Conclusions
Quadriceps weakness and the magnitude of muscle activity, both antagonist and agonist, appeared to be most strongly related to functional parameters in children with CP. However, these functional relations did not overlap. Compared with their normally developing peers, children with CP more commonly use cocontraction during over-ground and treadmill walking and during isometric maximal exertions. Potential negative effects of excessive cocontraction include greater total muscle activation during net force production and altered movement quality and quantity (because of increased joint stiffness). Nonetheless, cocontraction may still be a useful compensatory strategy in CP to increase joint stability, to limit degrees of freedom, or to allow the motor system to respond more readily to perturbations. More research is needed on the neurophysiologic bases of excessive cocontraction in CP, the effect of therapeutic interventions such as surgery, bracing, or exercise, and the functional benefits of reducing cocontraction in persons with CP. The precise relation between force and EMG measures is fundamental to understanding normal and abnormal motor function, yet is still a subject of much debate and controversy; it, too, warrants extensive further investigations.
References
1Rosen, MG and Dickinson, JC. The incidence of cerebral palsy. Am J Obstet Gynecol. 1992; 167: 417–423
2Berger, W, Quintern, J, and Dietz, V. Pathophysiology of gait in children with cerebral palsy. Electroencephalogr Clin Neurophysiol. 1982; 53: 538–548
3Sahrmann, SA and Norton, BJ. The relationship of voluntary movement to spasticity in the upper motor neuron syndrome. Ann Neurol. 1977; 2: 460–465
4Ito, J, Akiko, A, Tanaka, H, Tasaki, T, Cho, K, and Yamazaki, R. Muscle histopathology in spastic cerebral palsy.Brain Dev. 1996; 18: 299–303
5Dietz, V and Berger, W. Cerebral palsy and muscle transformation. Dev Med Child Neurol. 1995; 37: 180–184
6Wiley, ME and Damiano, DL. Lower extremity strength profiles in spastic cerebral palsy. Dev Med Child Neurol. 1998; 40: 100–107
7Olney, SJ. Quantitative evaluation of cocontraction of knee and ankle muscles in normal walking. in:Biomechanics IX-A. Human Kinetics, Champaign (IL); 1985: 431–435
8Winter, DA. Biomechanics and motor control of human movement. in: John Wiley:, New York; 1990: 114–117
9Damiano, DL. Reviewing muscle cocontraction: is it a developmental, pathological, or motor control issue.Phys Occup Ther Pediatr. 1993; 12: 3–20
10Campbell, SK. Framework for the measurement of neurologic impairment and disability. in: Contemporary management of motor control problems. Foundation for Physical Therapy, Alexandria (VA); 1991: 146–147(Proceedings of the II Step Conference)
11Myklebust, BM, Gottlied, GL, and Agarwal, GC. Stretch reflexes of the normal infant. Dev Med Child Neurol.1986; 28: 440–449
12Baratta, R, Solomonow, M, Zhou, BH, Letson, D, Chuinard, R, and D'Ambrosia, R. Muscular coactivation: the role of antagonist musculature in maintaining knee stability. Am J Sports Med. 1988; 16: 113–122
13Knutsson, E and Martensson, A. Dynamic motor capacity in spastic paresis and its relation to prime mover dysfunction, spastic reflexes and antagonist co-activation. Scand J Rehab Med. 1980; 12: 93–106
14Levin, MF and Hui-Chan, C. Ankle spasticity is inversely correlated with antagonist voluntary contraction in hemiparetic subjects. Electromyogr Clin Neurophysiol. 1994; 34: 415–425
15Unnithan, VB, Dowling, JJ, Frost, G, and Bar-Or, O. Role of cocontraction in the O2 cost of walking in children with cerebral palsy. Med Sci Sports Exerc. 1995; 28: 1498–1504
16Fellows, SJ, Kaus, C, and Thilmann, AF. Voluntary movement at the elbow in spastic hemiparesis. Ann Neurol.1994; 36: 397–407
17Peat, M, Hyman, IC, Dubo, MD, Winter, DA, Quanbury, AO, Steinke, T et al. Electromyographic temporal analysis of gait: hemiplegic locomotion. Arch Phys Med Rehabil. 1976; 57: 421–425
18Ikeda, AJ, Abel, MF, Granata, KP, and Damiano, DL. Quantification of cocontraction in spastic cerebral palsy.Electromyogr Clin Neurophysiol. 1998; 38: 497–504
19Damiano, DL and Abel, MF. Functional outcomes of strength training in spastic cerebral palsy. Arch Phys Med Rehabil. 1998; 79: 119–125
20Russell, D, Rosenbaum, P, Cadman, D, Gowland, C, Hardy, S, and Jarvis, S. The gross motor function measure: a means to evaluate the effects of physical therapy. Dev Med Child Neurol. 1991; 31: 351–352
21Rose, J and Gamble, JG. The energy expenditure index: a method to quantitate and compare walking energy expenditure for children and adolescents. J Pediatr Orthop. 1991; 11: 571–578
22Woods, JJ and Bigland-Ritchie, B. Linear and non-linear surface EMG/force relationships in human muscles.Am J Phys Med. 1983; 62: 287–299
23Bohannon, RW and Andrews, AW. Relationship between impairments and gait performance after stroke. A summary of relevant research. Gait Posture. 1996; 3: 236–240
24Basmajian, JV and DeLuca, CT. Muscles alive: their functions revealed by electromyography. in: 5th ed. Williams & Wilkins, Baltimore; 1985: 223–228
Nenhum comentário:
Postar um comentário