Eagleton, Melissa MPT; Iams, Amelia MPT; McDowell, Jennifer MPT; Morrison, Rachel MPT; Evans, Constance L. MA, PT
Strength Issues in Cerebral Palsy
One of the accepted definitions of strengthening is an increase in the force generation capability of muscle tissue due to physiologic differences. In adolescents, it is more appropriate to relate differences in strength to force generation than to muscle hypertrophy. Damiano and Abel 1 examined strengthening in children and adolescents with cerebral palsy (CP) and demonstrated a direct correlation between increased strength and improved functional ability, especially during gait. Berger et al 2 and Olney et al 3 used biomechanical and electromyographic (EMG) analyses of CP gait to demonstrate that the primary limiting factor in ambulation for this population is muscle weakness. In the absence of normal muscle strength, adolescents with CP must find other ways to create stability around joints. EMG patterns during walking reveal that adolescents with CP exhibit cocontraction defined as the simultaneous activity of agonistic and antagonistic muscles around a joint. 4 Cocontraction may be advantageous during gait because it may be used to achieve joint stability, especially at the ankle. However, when coupled with inadequate force production frequently seen in the gait of individuals with CP, cocontraction has been linked to inefficient energy expediture. 4
Berger et al 2 compared EMG activity of leg muscles during locomotion in children who were healthy with that of children with supraspinal lesions. They observed that children with supraspinal lesions in most instances had reduced EMG activity of the anterior tibialis and gastrocnemius muscles. In addition, children with CP showed stereotypical cocontraction in the antagonistic leg muscles just before and during the stance phase of the gait cycle. The amount of cocontraction depended on the severity of the motor lesion. 2 Muscle imbalances between antagonistic muscles of the lower extremity are a common finding in patients with CP. Without intervention, imbalances may become more pronounced over time, causing further weakness, muscle atrophy, joint contractures, and possible joint deformities. 5 The balance of complementary muscle groups is essential to overcome asymmetric muscular development and promote efficient functioning in individuals with CP. 6
CP Gait Analysis
Inefficient ambulation is one of the most significant functional limitations in patients with CP. Five parameters define normal gait:7 (1) stance phase stability, (2) swing phase clearance, (3) foot pre-position in terminal swing, (4) adequate step length, and (5) energy conservation. In adolescents with CP, several of these parameters may be altered. Specific gait abnormalities associated with spastic CP vary with the type and degree of impairment. The most commonly observed characteristics of gait in adolescents with spastic CP include limited hip and knee range of motion, excessive hip adduction and internal rotation, anterior pelvic tilt, pelvic obliquity, and persistent plantar flexion at the ankle. 8,9 Common characteristics of CP gait include decreased heelstrike, decreased walking speed, decreased stride length and/or cadence, and decreased ability to increase velocity on demand compared with adolescents without impairments.
When considering abnormalities of gait in the population with CP, it is important to realize that abnormalities do not occur in isolation but interact with each other. Primary deviations are directly related to impairments of the central nervous system, and secondary deviations are compensations for primary deviations. An example of this interaction is the primary deviation of cospasticity (defined as the abnormal state of rate-dependent contraction of the agonist and antagonist muscles around a joint) of the rectus femoris and hamstrings that are commonly associated with a “stiff-knee” gait, manifested as decreased knee flexion in swing. The stiff-knee gait may be associated with difficulty with foot clearance during swing phase and may be compensated for by hip hike or circumduction.
The gait deviations described above are remarkably similar to the characteristics of early gait patterns in children who are healthy. 10 In both instances, shorter stride and step lengths, increased time in stance phase, wider base of support, decreased velocity, and increased hip and knee flexion throughout gait in both stance and swing are observed. 2,8,11 The EMG patterns are also similar in both instances. Increased cocontraction is displayed at the joints when compared with the normal mature walking pattern.
Physical Therapy Intervention in CP
Physical therapy intervention for individuals with CP regardless of age is focused on promoting the optimal level of function. During the preschool period, the child might receive intensive therapy to treat functional limitations, specifically addressing impaired force generation and skeletal alignment. 11 Helping the child to develop motor patterns and the use of direct training are two commonly used methods of intervention. The treatment of school-aged children and adolescents with CP is focused on maintaining the patient’s level of function in the face of changes occurring with growth and maturation. Physical therapists also address the individual’s increasing need to develop life skills and to form strategies to overcome environmental and societal barriers. 11
In the traditional treatment of children with CP, strengthening programs have been avoided for the reasons identified by Damiano et al:12 (1) the possibility of increasing spasticity and thereby exacerbating further muscle contractures and joint stiffness, (2) adolescents with CP lack the isolated muscle control needed to increase strength in targeted muscles, and (3) weakness is not considered to be the primary limiting factor of motor function.
Strength Training in CP
Classifying patients with CP as hemiplegic, diplegic, and quadriplegic describes the distribution pattern of spasticity but, more relevant to this study, directly implicates weakness as one of the hallmark clinical characteristics of this population. 1 Several researchers have been able to demonstrate a correlation between generalized muscle weakness and decreased performance in gait 13 and, more specifically, the inefficiencies and decreased stride length in CP gait. 14
A number of researchers provide evidence that supports the benefits of strengthening programs in this patient population. Damiano et al 12 focused on increasing quadriceps muscle strength in children with CP and were able to demonstrate the capacity to significantly increase strength of the quadriceps muscle in all the children in the study, with the majority attaining normal strength values. The greatest gains occurred at the points in the range that were initially the weakest. In addition, strengthening the knee extensors decreased the amount of knee flexion during stance and increased stride length during gait. These findings were thought to be due to the improved balance of the agonist-antagonist relationship at the knee. 12 This may be explained by the increased capacity of the strengthened quadriceps muscle to contract in end range. Horvat 6 also demonstrated the effectiveness of strengthening a patient with spastic CP through a progressive resistance training program using free weights and weight machines. Strength was assessed on a Cybex II Isokinetic System. Strength, muscular endurance, and range of motion gains were noted on both sides of the body indicating a general improvement, with the greatest improvements observed in the involved lower extremity. 6
Muscle strength has been highly correlated with gait parameters, especially with gait velocity, and should therefore be considered during gait analysis. After muscle strengthening, free walking velocity has been shown to increase as a result of increased cadence and/or increased step length. 1
The normal gait pattern is generally considered to be the most energy-efficient means of ambulation. Because there is a dependent relationship between energy conservation and the other parameters of normal gait, improving gait parameters should result in an increased level of energy conservation. In addition, compensatory gait patterns may be eliminated by resolution of the primary deviations. 7
Spasticity and the lack of motor control associated with CP have long been recognized as sources of inefficiencies, especially during gait. It has been reported that the energy demands of walking for adolescents with CP are 1.5 to three times that of healthy adolescents, depending on the degree and type of involvement. 7 Energy expenditure during gait can be reduced in adolescents with CP in at least three ways. These include reduction of cospasticity, restoration of stance phase stability, and elimination of foot drag, all of which can be improved with strengthening of the trunk and lower extremity. 7
A study by Kramer and MacPhail 14 used the Energy Expenditure Index (EEI), which is defined as walking heart rate minus resting heart rate (RHR), divided by walking speed, expressed in beats per meter 15 as a measure to quantify the energy consumed during walking in children with CP. Their results established correlations between knee extensor strength and walking efficiency. These correlations suggest that knee extensor strength and velocity of movement are important components of energy-efficient ambulation. 14
In summary, it can be hypothesized that weakness in adolescents with CP is a major contributor to gait deviations and inefficiencies during ambulation. Although questions still exist regarding the effectiveness of strength training, a number of studies have demonstrated that strength training can be implemented without increasing spasticity. Research by Bohannon 13 supports the concept that patients with brain lesions, whose muscles function differently than expected under normal circumstances, can still achieve strengthening. Similarly, Damiano et al 12 suggest doing resistance exercise in a more functional position as suggested by the principle of specificity of training. This warrants the investigation of the possible benefits of a functional strength-training program specific to muscles used in gait, which is the focus of this study.
Community Fitness Programs
Darrah et al 16 report that exercise programs for individuals with a disability can have psychological effects that result in improved attitude and confidence toward their disability. The study consisted of a comprehensive fitness program conducted in a community facility to encourage participants to view exercise as a social rather than a medical activity. The results demonstrated a dramatic difference in their subjects’ self-perceptions of physical appearance. In addition, the social relationships fostered during the study were maintained among the subjects and also resulted in continuation of the program in another facility by the majority of the participating families. The authors concluded that encouraging exercise as part of self-care rather than medical care in individuals with CP resulted in internal motivation to continue exercising.
The purpose of this study was to implement a generalized strengthening program for adolescents with CP addressing the trunk and lower extremity muscles and to examine whether differences in step length, cadence, velocity, the distance walked in three minutes, and energy expenditure would result. In addition, the preference for completing the program in a community fitness atmosphere as opposed to during standard physical therapy appointments was examined.
The implementation of a six-week trunk/lower extremity strengthening regimen in adolescents with CP will (1) increase gait velocity, cadence, and step length; (2) increase the distance walked in a three-minute time frame; and (3) improve energy efficiency during gait as measured by a decreased EEI. In addition, patients will report a preference for completing their fitness programs in a community atmosphere as opposed to the standard physical therapy clinic.
Twenty adolescents with various presentations of CP were contacted; 13 agreed to participate in the study, and seven completed the study in its entirety. The effects of strength training on gait were examined in these seven adolescents with CP, aged 12 to 20 years, who resided in the Winchester, VA, and Brunswick, ME, areas. Exclusion criteria included subjects who had a history of Botox injections, dorsal rhizotomy, or use of a baclofen pump due to the lack of knowledge regarding their effects on the capacity for strength training. In addition, participants with current cardiopulmonary problems were excluded from the study. All subjects were able to ambulate independently 150 ft on a level surface with or without an assistive device and follow multistep commands and were medically cleared by a physician for a strength-training program.
The institutional review board of Shenandoah University Health Professions Council approved this study. The parents and adolescents gave informed consent before data collection.
Timex wristwatches with a digital readout to 0.01 seconds were used to measure time for ambulation trials and heart rate.
Standard and double stethoscopes.
The rater measuring the subjects in Brunswick, ME, used a standard stethoscope. A double stethoscope was used to measure the subjects in the Winchester, VA, area. The stethoscopes were used to measure 10-second resting and postexercise heart rates.
A treadmill was used for three-minute walk test at a self-selected velocity for subjects tested in Virginia. A comfortable walking velocity was determined by the subject as the researcher slowly increased the speed of the treadmill. Rose et al 9 have shown that self-selected velocity on the ground was comparable with self-selected velocity on the treadmill.
Standard Cloth Measuring Tape
The tape was used to measure the distances for ambulation trials.
A pretest-posttest group design was used to analyze the effects of a strength-training program on five dependent measures: gait velocity, step length, cadence, distance ambulated in a three-minute walk, and EEI. All subjects were instructed to maintain their typical activity levels in addition to their exercise routine throughout the study. Pretest data were collected no more than two weeks before implementation of the strengthening program. After the collection of the pretest data, the exercise program was begun. Six data points for each of the following dependent variables were collected on various days over a two-week period: (1) heart rate before and after a three-minute walk to calculate EEI, (2) three-minute walk distance on level surface or treadmill at a self-selected velocity, (3) visual analysis of cadence and calculation of step length during six-meter ambulation trials, and (4) timed six-meter walk for velocity. Data collection procedures were repeated post-testing no more than two weeks after completion of the training program.
RHR was measured after having the subject sit for two minutes with as little movement as possible. After measurement of RHR, subjects completed a three-minute walk test on level ground or on the treadmill. Subjects who were unfamiliar with the treadmill were given a three-minute orientation period at the beginning of the data collection session to familiarize themselves with walking on the treadmill. Postexercise heart rate was obtained immediately after cessation of the test. This information was used to calculate EEI (walking heart rate minus RHR, divided by walking speed in meters). The distance ambulated within three minutes was recorded to the nearest tenth of a meter.
To obtain measures of velocity, cadence, and step length, each subject was asked to walk a 10-meter distance quickly and safely at a self-selected velocity. Only the middle six meters were used for data collection to account for periods of acceleration and deceleration. While timing the six-meter walk, the number of steps taken by the subject was counted. Step length was determined by dividing the six meters ambulated by the number of steps taken by the subject. Each subject consistently demonstrated approximately the same step length bilaterally.
The intervention consisted of a strength-raining program performed three times per week for six weeks. All subjects were asked to exercise with a partner who could help with equipment set up and exercise log recording. Subjects who missed five or more days of the training program were not included in the study. The training program was focused on trunk, hip, knee, and ankle flexors and extensors and hip abductors using free weights and weight machines. In instances when the available equipment was inadequate or weight increments were too large, exercises using cuff weights, Theraband, or the subject’s body weight were substituted. One to two therapists involved in the study met the subject and his/her partner on the first day of the exercise program at the workout location. The subject and partner were given a visual demonstration, and then the subject performed the activity. This was done for each of the exercises, and verbal feedback was given as needed. Subjects were instructed to progress the exercise load (originally established as 80% of each subject’s one repetition maximum) and repetitions (originally eight to 10) as follows: (1) increased repetitions of the original load, (2) increased load and decreased repetitions to the initial number of repetitions, and then (3) increased repetitions at the increased load. This suggested progression was indicated on the exercise log sheets. To address flexibility, the subjects were provided with a list of stretches that included illustrations. They were asked to perform several of the stretches before and after their strengthening workout. Subjects spent an average of 40 minutes to one hour to complete a workout session.
Subjects were asked to record any muscle soreness that occurred while exercising or other complaints on their log sheets. During the first week of the training program, subjects were telephoned to inquire whether any problems were encountered during their exercise sessions. The rate of progression of load and/or repetitions was contingent on weekly feedback from the subject throughout the program during scheduled weekly checks by one of the therapists. Subjects completed the program either at a fitness center or at a school gym.
A pilot study was conducted using a healthy adolescent within the specified age range. This was done to ensure that the exercise protocol was appropriate for the population of adolescents with CP who would be carrying out the strengthening program. The adolescent and his/her partner also provided feedback on the ease of following the progression protocol as it was explained on the exercise log.
All data collected were analyzed using a group design. An intraclass correlation coefficient (ICC)2,1 was run on the repeated measures of the dependent variables from two subjects between the two raters in Winchester, VA. A coefficient of variation was calculated to test the reliability of the dependent measures between the Maine rater and one rater in Virginia for the remaining five subjects. Nonparametric tests were run to analyze the data because of the smaller than expected number of participants in this study. Wilcoxon signed-rank tests were run between pre- and postintervention data for each of the dependent measures. The strength of the relationship between independent and dependent variables was analyzed using η2. 17,18 η2 is defined as the portion of explained variance between the variables. This statistic removes variance secondary to individual differences and ranges in value from 0 to 1, with 1 demonstrating the strongest relationship. Data specific to each subject were plotted on individual graphs with trial on the X axis and the dependent measure on the Y axis. An α level of 0.05 was used for analyzing data. It was believed that a 95% confidence level would be sufficient to draw conclusions on the effects of strength training over multiple independent variables.
The ICC2,1 for heart rate was 0.95, and for stopwatch timing, it was 0.93, indicating excellent reliability between two raters. Coefficients of variation, as shown in Table 1, were run for heart rate measurements at rest and after exercise as well as for stopwatch timing to test for the intrarater reliability of two raters.
Differences in strength before and after intervention could not be measured by hand-held dynamometer for five of the seven subjects due to equipment failure. Lacking an objective measure, these data were excluded from our results.
The results of the Wilcoxon signed-rank tests comparing the preintervention measures with postintervention measures indicate a significant increase in the group’s gait velocity (p = 0.05), step length (p = 0.05), and cadence (p = 0.05) as well as in the distance walked in three minutes (p = 0.05), as seen in Table 2. A significant decrease was observed in EEI (p = 0.05) in the group, also shown in Table 2. Of the five dependent measures, η2 values demonstrated adequate strength (70% or greater) for distance ambulated during three-minute walk, step length, and velocity, as shown in Table 3.
Figures 1–5 illustrate the mean differences in all dependent variables assessed in the form of bar graphs. Although our statistics involve the group, it is interesting to observe that subjects 6 and 7 demonstrate the greatest differences in each of the five dependent variables (Figures 1–5).
According to Bohannon, 13 velocity, cadence, appearance, and independence level of gait are directly related to muscle strength. Studies addressing gait efficiency and speed in adolescents with CP have shown varied results when increases in lower extremity muscle strength are demonstrated. The purpose of this study was to examine the use of strength training as a possible therapeutic intervention to improve gait in adolescents with CP.
The results of this study demonstrate a significant increase in gait velocity, step length, and cadence. It has previously been hypothesized that step length and cadence are linearly related to velocity in adolescents without CP; our results suggest that this may also be true in adolescents with CP. In 1995, Damiano et al 5 found that children with CP who participated in a quadriceps muscle–strengthening program increased gait velocity through an increase in stride length without an increase in cadence. Visual analysis of our five figures reveals that six of the seven subjects demonstrated some increase in step length and velocity, but only four of seven demonstrated an observable increase in cadence. Further research with a larger sample size is suggested for more conclusive evidence that step length and cadence are linearly related to gait speed in this population.
The observed increase in step length may be explained by improved quadriceps activation and, consequently, increased hamstring inhibition. Inhibition of the hamstrings through the swing phase of gait to initial contact may allow increased hip flexion and therefore improved step length.
Overall improved trunk and lower extremity strength may have resulted in increased core stability and balance during gait, allowing increased comfort during walking and therefore increased speed. The results of this study indicate an increase in distance traversed during a three-minute walk test. This increase in distance can be correlated with the significant increases in gait velocity, step length, and cadence. In a 1994 study by Kramer and MacPhail, 14 the researchers concluded that treatment techniques such as strength training affected gross motor ability and may have had a carry-over effect on walking efficiency. This conclusion was based on the relationship found between the EEI and strength during comfortable walking. According to the authors of that study, a significant increase in EEI would be expected if a significant increase in velocity had been found. 14 In this study, we found a decrease in EEI in four of the seven subjects as their velocity increased after intervention. Our results are in accord with the EEI results reported by Damiano and Abel 1 in their study of children with CP in a strengthening program. The EEI values in our study were variable across subjects, as can be seen in Figure 5, similar to the mean values for EEI in the Damiano and Abel study. 1 Increases in EEI can be explained by increased distance walked during a fixed time period or an increased velocity. A decrease in EEI may result from increased mechanical efficiency during ambulation. In some cases in which EEI increased, the efficiency of energy expenditure may have been counteracted by an increase in gait velocity, resulting in no observable difference.
A possible explanation for the increases in gait velocity, step length, and cadence with a decrease in EEI after a strengthening program may be a result of strengthening causing a difference in the adolescents’ gait patterns. This could be considered a phase shift in their motor control ability. The increase in strength may have helped to increase their stability and in some cases allowed the child to increase his/her step length. Another explanation is that the adolescents had an improved sense of well-being resulting in improved confidence and quality of movement. There may also have been increased stance stability and/or improved muscle health secondary to increased movement allowing them to adapt better to the demands of walking. The combination of any of the above factors may have resulted in the observed ability to ambulate faster with less energy expenditure.
Based on subjective feedback from participants and their caregivers, it was determined that all seven subjects enjoyed the strengthening program in a fitness center more than during typical physical therapy. It is also worth noting that during their exercise sessions, neither the subjects nor their caregivers reported any exacerbation of spasticity. This further supports the notion that strength training is an enjoyable treatment option for this patient population.
There were several limitations of this research project. One was the time constraint for the strength training program. Six weeks has been reported as the minimal time necessary to show increases in strength or muscle hypertrophy for adults. 19 Although the length of time for adolescents with CP to show improvements in strength has not been determined, it seems reasonable that it might be a similar or greater length of time. It is also important to note that the lack of a second hand-held dynamometer prevented our ability to report accurately any differences in muscle strength after the intervention. This lack of data prevents us from definitively reporting whether any significant gains in lower extremity and trunk strength were achieved. The therapists’ manual muscle testing suggested that an increase occurred in most subjects. Another limitation involves the fact that the training equipment was designed and intended for adults; some of the subjects were too small to use the equipment and had to instead use Theraband for resistance.
Using the treadmill for the three-minute walk test for two of the subjects limited our ability to demonstrate improvements in distance ambulated. The possibility exists that subjects’ strength could have increased over the course of the study, regardless of intervention. We did not use a control group in this study. Due to the staggered entrance of the subjects into the training program, those entering in the winter months may have been less active in general compared with those entering in fall and spring, affecting the amount of overall improvement that they may have achieved. Months with inclement weather may have posed problems related to transportation and therefore affected the number of times the subjects were able to get to the fitness center. The most influential limitation to our study was subject compliance. This significantly limited our group size because we experienced nearly a 50% decrease in the number of participants who started the study to those who completed the postintervention data collection.
Future research is needed to determine the appropriate duration required for a strength training program to achieve increased muscle strength in adolescents with CP. Assessing the long-term effects of a strengthening program to determine the duration of its influence is also strongly suggested. It would be interesting to further investigate the relationship, if any, between the dependent variables on a larger sample size and possibly plotting these variables weekly over the course of the exercise program in addition to before and after intervention. In addition, it would be interesting to look at the relationship between muscle strengthening and range of motion at associated joints. This may help to determine whether a relationship does exist between increased quadriceps strength and increased hip flexion during gait and step length. Another area for future study may be an investigation of joint angles and the effects of strength training on body alignment during ambulation.
Strengthening exercise is an important therapeutic intervention used by physical therapists in a wide range of practice settings. The effects of a strengthening program on healthy adult subjects are well documented, but there is little information on the effects of a strengthening program in adolescents with CP. Further, it is not well documented whether there is a relationship between trunk and lower extremity strengthening and ambulation ability.
The results of this study suggest that a six-week strengthening program was effective in increasing the distance walked in three minutes, gait velocity, step length, and cadence of our group and resulted in an overall decrease in EEI. All subjects in this study reported a preference for exercising in a community fitness atmosphere as opposed to exercising during routine physical therapy appointments. Confirming these findings with a larger sample size and objective measures to determine any difference in trunk and lower extremity strength may offer a new evidence-based method of intervention for adolescents with CP.
1. Damiano DL, Abel MF. Functional outcomes of strength training in spastic cerebral palsy. Arch Phys Med Rehabil. 1998; 79: 119–125.
2. Berger W, Quintern J, Dietz V. Pathophysiology of gait in adolescents with cerebral palsy. Electroencephalogr Clin Neurophysiol. 1982; 53: 538–548.
3. Olney SJ, MacPhail HEA, Hedden DM, et al. Work and power in hemiplegic cerebral palsy gait. Phys Ther. 1990; 70: 431–437.
4. Unnithan VB, Dowling JJ, Frost G, Volpe Ayub B, Bar-Or O. Cocontraction and phasic activity during GAIT in adolescents with cerebral palsy. Electromyogr Clin Neurophysiol. 1996; 36: 487–494.
5. Damiano DL, Kelly LE, Vaughn CL. Effects of quadriceps femoris muscle strengthening on crouch gait in adolescents with spastic diplegia. Phys Ther. 1995; 75: 658–671.
6. Horvat M. Effects of a progressive resistance training program on an individual with spastic cerebral palsy. Am Corr Ther J. 1987; 41: 7–11.
7. Gage JR. Gait analysis: an essential tool in the treatment of cerebral palsy. Clin Orthop. 1993; 288: 126–134.
8. Johnson DC, Damiano DL, Abel MF. The evolution of gait in childhood and adolescent cerebral palsy. J Pediatr Orthop. 1997; 17: 392–396.
9. Rose SA, Ounpuu S, DeLuca PA. Strategies for the assessment of pediatric gait in the clinical setting. Phys Ther. 1991; 71: 961–980.
10. Berger W. Characteristics of locomotor control in adolescents with cerebral palsy. Neurosci Biobehav Rev. 1998; 22: 579–582.
11. Campbell SK. Physical Therapy for Children. Philadelphia: WB Saunders; 1995: 489–524.
12. Damiano DL, Vaughan CL, Abel MF. Muscle response to heavy resistance exercise in adolescents with cerebral palsy. Dev Med Child Neurol. 1995; 37: 731–739.
13. Bohannon RW. Is the measurement of muscle strength appropriate in patients with brain lesion? A special communication. Phys Ther. 1989; 69: 225–236.
14. Kramer JF, MacPhail HEA. Relationships among measures of walking efficiency, gross motor ability, and isokinetic strength in adolescents with cerebral palsy. Pediatr Phys Ther. 1994; 6: 3–8.
15. Rose J, Gamble JG, Lee J, et al. The energy expenditure index: a method to quantitate and compare walking energy expenditure for adolescents and adolescents. J Pediatr Orthop. 1991; 11: 571–578.
16. Darrah J, Wessel J, Nearingburg P, et al. Evaluation of a community fitness program for adolescents with cerebral palsy. Pediatr Phys Ther. 1999; 11: 18–23.
17. Ott RL. An Introduction to Statistical Methods and Data Analysis. Pacific Grove, CA: Duxbury Press; 1993: 297–302.
18. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. Norwalk, CT: Appleton and Lange; 1993.
19. Olney SJ, Costigan PA, Hedden DM. Mechanical energy patterns in gait of cerebral palsied adolescents with hemiplegia. Phys Ther. 1987; 67: 1348–1354.
20. Crenna P. Spasticity and ‘spastic’ gait in adolescents with cerebral palsy. Neurosci Behav Rev. 1998; 22: 571–578.
21. Tecklin JS. Pediatric Physical Therapy. Philadelphia: JB Lippincott; 1998: 68–105.
22. Bohannon RW. Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil. 1986; 67: 390–392.
23. Detrembleur C, Willems P, Plaghki L. Does walking speed influence the time pattern of muscle activation in normal adolescents? Dev Med Child Neurol. 1997; 39: 803–807.
physical therapy/methods; cerebral palsy; adolescent; treatment outcome; gait/physiology; weight lifting/physiology; exercise therapy
© 2004 Lippincott Williams & Wilkins, Inc.