Relationship Between Age and Spasticity in Children With Diplegic Cerebral Palsy

Abstract

Objective:

To examine the relationship between passive torque, reflex activity, co-contraction, and age during the assessment of spasticity of knee flexors and extensors in children with spastic diplegic cerebral palsy (CP).

Design:

Retrospective.

Setting:

Pediatric orthopedic hospital.

Participants:

Children (N=36) with spastic diplegic CP.

Interventions:

Not applicable.

Main Outcome Measures:

Spasticity of the knee flexors and knee extensors (as measured by peak passive torque, mean passive torque, reflex activity of the medial hamstrings, reflex activity of vastus lateralis, and co-contraction) was assessed during passive movements completed using an isokinetic dynamometer with concurrent electromyography.

Results:

A significant positive relationship was found between age and mean knee flexor passive torque (P<.05), while a significant negative relationship was found between age and mean percentage of the range of motion with co-contraction (P<.05).

Conclusions:

Our results suggest that passive stiffness may play a larger role in spasticity than reflex activity as children with spastic diplegic CP age. Additional research is needed to determine whether subject age could influence the effectiveness of interventions, such as serial casting or botulinum toxin, for spasticity in children with spastic diplegic CP.

SPASTICITY IS A COMMON impairment found in children with CP.¹ While alternative definitions for spasticity have been proposed, spasticity has been most commonly defined as a velocity dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks, resulting from hyperexcitability of the stretch reflex.² One method of measuring spasticity uses an isokinetic dynamometer to passively move a limb through a defined range of motion while the peak resistive torque is calculated. The peak resistive torque consists of a combination of reflexive components (such as co-contraction in agonist/antagonist muscle pairs and reflex activity in a stretched muscle) and nonreflexive components (such as increased muscle stiffness due to contracture or other factors in the absence of reflexive activity) of spasticity. The reliability of isokinetic dynamometry for the measurement of spasticity has been supported in children with CP.^3,4 An advantage of using isokinetic dynamometry to assess spasticity is that both the reflexive and nonreflexive components of spasticity can be quantified through concurrent EMG data acquisition. Our research team has reported increased co-contraction and reflex activity during the assessment of spasticity in children with CP compared with peers with typical development by using isokinetic dynamometry.^5,6

It is unknown whether aging influences spasticity in children with diplegic CP. While decreases in gait performance and passive range of motion of the lower extremity have been reported with aging in children with CP,^7,8 changes in spasticity in children with CP as they age have received limited attention in the literature. Hagglund and Wagner⁹ recently found that spasticity of the plantarflexors as measured by the Modified Ashworth Scale increased up to the age of 4 years and then decreased up to the age of 12 years in a sample of children with various subtypes of CP. However, their results should be interpreted cautiously because of limitations in the reliability of the Ashworth Scale.^10,11 An examination of the relationship between spasticity and aging using isokinetic dynamometry with EMG may provide insight into how spasticity changes as children age, which could be used for treatment planning in children with CP. The purpose of this study was to examine the relationship between passive resistive torque, reflex activity, coactivation, and age during the assessment of spasticity of knee flexors and extensors in children with CP by using isokinetic dynamometry and surface EMG.

METHODS

Subjects

A university-affiliated institutional review board approved this investigation. Written informed consent from each subject’s parents along with verbal assent was obtained from each subject before participation. Subjects met the following inclusion criteria: (1) a diagnosis of spastic diplegic CP; (2) between the ages of 7 and 14 years; (3) no documentation in the medical record of hip subluxation, hip dislocation, or significant scoliosis (curvature >40°); (4) ability to follow 1-step commands and to attend to tasks associated with data collection; (5) at least 1 year postsurgery to the lower extremities; (6) at least 6 months post-botulinum toxin injection; and (7) 10° or less of knee flexion contracture. Data collected from 36 children with CP (mean age ± SD, 10.8±1.8y; range, 7.9–14.0y; GMFCS level I, 10 children; GMFCS level II, 8; GMFCS level III, 18) were analyzed retrospectively (table 1).

Table 1:

Subject Demographics

GMFCS Level	Age (y)
III	11.6
I	13.3
III	9.3
I	11.5
III	9.9
III	12.7
III	11.4
III	10.3
III	12.6
II	8.5
III	10.9
III	10.3
III	9.1
I	11.6
II	10.2
III	11.2
I	9.0
III	8.2
III	8.0
I	12.5
II	10.7
I	8.0
II	12.7
II	13.6
II	10.0
I	8.3
II	10.9
III	7.9
III	8.6
I	11.9
II	14.0
I	11.7
III	11.9
III	13.7
III	12.8
I	11.5

Protocol

The methods used in this investigation have been described elsewhere but are summarized here.⁶ The limb and order of muscles were randomly selected for testing. Two surface EMG electrodes were placed on the child’s leg on the VL and MH. Signal conditioning electrodes^a with a parallel bar arrangement (contact area 1 × 10mm, 10-mm interelectrode distance) were used to record electrical activity of the muscles. These electrodes contained a built-in gain of 1000 volt/volt, a common mode rejection ratio of greater than 80dB at 60Hz, a noise level of 1.5μV root mean square, and a bandpass filter of 20 to 450Hz. Additional filtering and amplification of the EMG signal were completed using the Bagnoli 4-channel EMG system,^a which had a bandpass filter between 20 and 450Hz and a gain of 1000. EMG signals were sampled at a rate of 1.2kHz.

Subjects were seated on the isokinetic dynamometer^b in a position of 80° of hip flexion, 90° of knee flexion, and with the ankle unrestricted. The axis of the dynamometer was visually aligned with the knee joint’s axis. Distal attachment of the lower limb to the lever of the dynamometer was made approximately 3cm above the lateral malleolus. Baseline EMG data were collected for at least 5 seconds while subjects were instructed to relax as much as possible and not move. One set of 3 continuous passive movements at a velocity of 5°/s from 90° of knee flexion to 25° of knee flexion and back to 90° of knee flexion was collected for gravity correction calculation of the limb’s weight during data processing. EMG output from the amplifier, as well as force, velocity, and position data, was collected using a personal computer and custom software written in Labview 5.1^c for data analysis. EMG signals were displayed on a computer monitor for visual assessment. One set of 10 continuous passive movements from 90° of knee flexion to 25° of knee flexion was completed at 180°/s with a return speed of 5°/s to assess knee flexor spasticity. One set of 10 continuous passive movements from 25° of knee flexion to 90° of knee flexion was completed at 180°/s with a return speed of 5°/s to assess knee extensor spasticity. A 60-second rest break occurred after each movement velocity assessed.

A custom Matlab^d program was used for postprocessing and analysis of the torque and EMG data. The gravity-corrected knee flexor and knee extensor peak resistive torques were calculated for each repetition. Peak resistive torque was not calculated until the limb was moving at a constant velocity. The mean passive resistive torque for all movement repetitions at a given speed was quantified and labeled the passive torque. The maximum passive resistive torque of the 10 movement repetitions at each movement velocity was identified as the peak torque. EMG data were full-wave rectified and processed using a second-order Butterworth 10-Hz low-pass filter with phase correction to create a linear envelope. EMG onset and offset were then defined as muscle activity that was 3 SDs above baseline and occurred for a minimum of 50 milliseconds.¹² The number of movement repetitions and the percentage of each movement repetition with EMG classified as reflexive, or coactivation, were quantified. Reflexive muscle activity was defined as the presence of EMG activity of either the MH or VL during passive movement. Co-contraction was defined as simultaneous EMG activity of both the MH and VL during passive movement. Data were analyzed using SPSS version 14.0.^e Data were tested for normality and homogeneity of variance using the Shapiro-Wilks test and Levene’s test. All data were nonnormally distributed and thus were analyzed using Spearman rho correlations, and the level for statistical significance was set at P less than .05.

RESULTS

Table 2 presents Spearman correlation matrix of age, spasticity (peak and mean passive resistive torques), mean percentage of the range of motion with EMG activity of the MH (flexor reflexive activity), mean percentage of the range of motion with EMG activity of the VL (extensor reflexive activity), and co-contraction. A significant positive relationship was found between age and mean knee flexor passive torque (P<.05), while a significant negative relationship was found between age and mean percentage of the range of motion with co-contraction (P<.05). No other significant relationships were found between age, spasticity, and EMG activity.

Table 2:

Spearman Correlation Matrix of Age, Spasticity (as Measured by Mean and Peak Passive Torque of the Knee Flexors and Knee Extensors), Mean Percentage of the Range of Motion With EMG Activity of the MH, Mean Percentage of the Range of Motion With EMG Activity of the VL, and Co-Contraction

Assessment of Knee Flexor Spasticity
	Mean knee flexor passive torque	Mean % of MH activation	Mean % of VL activation	Mean % of co-contraction
Age	.418*	−.189	−.225	−.310
	Peak knee flexor passive torque	Mean % of MH activation	Mean % of VL activation	Mean % of co-contraction
Age	.313	−.247	−.283	−.378*
Assessment of knee extensor Spasticity
	Mean knee extensor passive torque	Mean % of MH activation	Mean % of VL activation	Mean % of co-contraction
Age	.263	.013	−.093	.057
	Peak knee extensor passive torque	Mean % of MH activation	Mean % of VL activation	Mean % of co-contraction
Age	.239	.200	.179	.215

^*P<.05.

DISCUSSION

The current investigation examined the relationship between passive resistive torque, reflex activity, co-contraction, and age during the assessment of spasticity of knee flexors and extensors in children with spastic diplegic CP by using isokinetic dynamometry and surface EMG. The significant positive relationship found between age and mean knee flexor passive torque conflicts with the findings of Hagglund and Wagner,⁹ who found that spasticity decreased as children with CP aged from 4 to 12 years. Differences between the Modified Ashworth Scale and isokinetic dynamometry may explain this disparity because the reliability of the Ashworth Scale has been questioned,^10,11 while the reliability of isokinetic dynamometry to measure spasticity in children with CP has been supported.^3,4 In addition, the choice of muscle groups (plantarflexors vs knee flexors and knee extensors) and age ranges may be contributing factors to the differences in results between studies. Our finding of no association between measures of knee extensor spasticity and aging also suggests that the relationship may be muscle specific. Additional research examining the effect of aging on spasticity of different muscle groups as measured by isokinetic dynamometry with EMG is necessary to determine whether our results are generalizable to other muscle groups. However, the use of isokinetic dynamometry to assess children with CP who are younger than 7 years will be limited because these children may be physically too small to be accurately positioned on the device.

To our knowledge, the finding of a negative relationship between co-contraction during the assessment of knee flexor spasticity and aging has not been reported in the literature. The literature on neural development and CP is currently emerging but has been focused mainly on animal research and human studies of neonates and infants.^13,14 The significance and etiology of a decrease in co-contraction during passive movements in children with CP as they age require additional research.

Our interpretation of the finding of increased passive torque along with decreased co-contraction with aging is that reflex activity, such as co-contraction during passive movements, plays a less important role in spasticity than nonreflexive passive stiffness as children with CP age. If reflex activity played an important role in spasticity as children with CP aged, then our investigation would have found a positive relationship between antagonistic muscle reflex activity and aging instead of our finding of no relationship between these variables. The implications of our findings may affect clinical decision making on the treatment of spasticity as children with spastic diplegic CP age. For younger children, interventions that attempt to decrease muscle reflex activity, such as treatment with botulinum toxin, may be most effective. A retrospective study by Desloovere et al¹⁵ found that children with hemiplegic or diplegic spastic CP treated with botulinum toxin at 5 to 10 years of age demonstrated improved gait kinematics compared with a control group treated without botulinum toxin. These authors concluded that joint contractures and bony deformities were prevented in children treated at a young age with botulinum toxin, which supports our hypothesis that subject age should be considered when deciding treatments for spasticity in children with CP. In contrast, we would speculate that interventions which attempt to address passive stiffness in older children with spastic diplegic CP, such as stretching or serial casting, may be more likely to cause a meaningful change in the child’s spasticity, especially because children with CP may demonstrate more limitations in passive range of motion as they age.^7,8 Clearly, additional research is necessary to support these hypotheses with respect to optimal decision making in children with spastic diplegic CP as they age.

However, our results should be interpreted cautiously because they may have been influenced by the methodology chosen for this study. The completion of multiple passive movements into flexion and extension may have affected these results because Nuyens et al¹⁶ have found that repeated passive motions may diminish reflex activity and torque during the assessment of spasticity. The effect of repeated passive motions during the assessment of spasticity in children with CP needs to be examined in future studies, especially because repeated passive motions may underestimate a child’s spasticity. Spasticity in children with CP may also vary according to the functional status of the child with CP. Ostensjo et al¹⁷ reported that children classified as GMFCS levels I and II had significantly less spasticity as measured by the Modified Ashworth Scale than children classified with GMFCS levels III and IV. Himmelmann et al¹⁸ also reported a positive correlation between GMFCS levels and Ashworth Scale in a large sample of children with diplegic CP. Therefore, the results of this investigation may be specific to the range of GMFCS levels in this study and our specific sample, which was primarily GMFCS level III. Additional research investigating the interaction between age and GMFCS level is recommended. Finally, our sample was limited to children with spastic diplegic CP, so it is unknown whether our findings are generalizable to children with different subtypes of CP.

CONCLUSIONS

Our results suggest that passive stiffness may play a larger role in spasticity than reflex activity as children with CP age. Additional research is needed to determine whether subject age could influence the effectiveness of interventions for spasticity in children with CP.

List of Abbreviations

CP

cerebral palsy

EMG

electromyography

GMFCS

Gross Motor Function Classification System

MH

medial hamstrings

VL

vastus lateralis