Ross, Sandy A. MHS, PT, PCS; Engsberg, Jack R. PhD; Olree, Kenneth S. MS; Park, T. S. MD
Selective dorsal rhizotomy (SDR) is a surgical procedure performed to reduce lower extremity spasticity in children with cerebral palsy (CP). After SDR, children undergo an intensive rehabilitation program. SDR has already been shown to effectively reduce lower extremity spasticity. 1–6 However, the effect of this surgery and rehabilitation on muscle strength remains unclear. Muscle weakness in potential candidates has been reported as a contraindication to performing SDR. 7,8 Thus, it is speculated that after surgery and the subsequent reduc tion in spasticity, there will also be a reduction in strength. Postsurgical muscle weakness, both temporary and permanent, has also been reported as a limitation of this surgery. 2,7,9 Muscle strengthening is a main focus of physical therapy (PT) in children with cerebral palsy after a selective dorsal rhizotomy. 10 Despite the apparent importance of muscle strength in relation to this surgery, it is not routinely or objectively examined during the patient selection process or post rhizotomy. The number of SDR investigations that have been reported on strength is quite limited. Quantification of impairments, like muscle weakness, is needed to help determine the outcome of therapeutic procedures.
Functional outcomes after SDR are controversial. Several investigators have shown functional gains in gross motor function and gait after SDR. 4,11,12 Others have reported that the gains after SDR are because of the intensive PT and not necessarily the surgery. 13 A clearer understanding of outcomes will occur with quantification of impairments, functional limitations, disability, and societal limitations. 14 The focus of the current investigation is the quantification of strength at the impairment level.
Only three groups of investigators have reported strength changes as a function of SDR. 2–4,9,15 Peacock et al 2 and Arens et al 9 measured “power” based on subjective estimates reported from a survey of treating physical therapists. They reported an immediate postoperative reduction in strength, which improved with physical therapy. The report indicated that in the long-term, some muscle groups improved in power, although others remained weak. The method used to determine power, which muscles were tested, and the specific results were not included in the report. Thus, the significance of these subjective reports must be viewed in terms of these limitations.
Steinbok et al 3,4 quantified isometric quadriceps strength using a myometer (a hand-held force transducer) pre- and post-SDR and noted conflicting results. They documented a 30% increase in quadriceps strength one-year post-SDR in an initial study of children with spastic quadriplegia and diplegia, 3 but subsequently reported no significant change in strength nine months post-SDR in a randomized clinical trial of children with spastic diplegia. 4 They did not report the joint angle at which isometric quadriceps strength was measured. One limitation of the myometer is that it only records the maximum force at one joint angle. Force over a broad range of motion might be more revealing than a single maximum value.
Several investigators have quantified strength in children with cerebral palsy using isokinetic dynamometers. 15–19 Strength in these studies and in the current investigation is defined as the maximum torque-producing capability of a muscle group on an isokinetic dynamometer. McCubbin and Shasby 16 have demonstrated a 59% increase in strength of the triceps muscle group with an isokinetic dynamometer after six weeks of isokinetic resistance exercises in children with cerebral palsy. MacPhail and Kramer 18 and Engsberg et al 19 used a Kincom (an isokinetic dynamometer) (Chattanooga Group, Inc., Hixson, Tex) to quantify maximum knee torques in children with CP.
They found children with CP were significantly weaker than their peers without disabilities. Isokinetic devices can measure strength defined here as maximum torque and work (maximum torque throughout a range of motion). Additional benefits of using an isokinetic dynamometer include broad availability for clinical use and application to many joints.
Only one group of investigators, Engsberg et al, 15 objectively quantified changes in knee strength, using an isokinetic dynamometer, in a group of subjects with spastic diplegia CP as a function of SDR. They documented that post-SDR work values in the hamstrings were significantly improved over preoperative values. Although they noted a gain in strength of the hamstrings, the investigators only reported work values. They did not present maximum hamstring strength or results for other muscles (eg, quadriceps).
The primary purpose of this investigation was to quantify changes in strength of the quadriceps and hamstrings in children with cerebral palsy as a function of SDR surgery and rehabilitation. It was hypothesized that a significant increase in strength would occur as a result of the surgery and rehabilitation. A second purpose of this investigation was to compare the strength of the quadriceps and hamstrings of children pre- and post-SDR to the strength of children without disabilities. It was hypothesized that children without disabilities would be significantly stronger than children with CP, both pre- and post-SDR.
Twenty children without disabilities (WD group; mean age nine years, SD 3.2, range four–16 years; 10 boys, 10 girls; mean mass 36.9 kg, SD 14.9, range 19–64 kg) and 19 children with spastic diplegia (CP group; mean age nine years, SD 4.2, range four–16 years; nine boys, 10 girls; mean mass 32.7 kg, SD 14.6, range 15–58 kg) were recruited for this investigation. The children without disabilities were recruited from parents within the hospital community or were siblings of the children with CP tested in Human Performance Laboratory. Children with CP were diagnosed with spastic diplegia and referred by a neurosurgeon before SDR. Ambulatory status of the 19 children with CP both pre- and post-SDR is listed in Table 1. Fifteen of these children wore ankle-foot orthotics pre-SDR (seven were independent ambulators, seven used assistive devices, and one child was nonambulatory). All children and parents asked to participate signed an informed consent. Children with CP were tested the day before SDR surgery and approximately eight months postsurgery. Children without disabilities were tested on a single occasion.
Selective dorsal rhizotomy surgery has been previously described 20 but will be briefly reviewed. The goal of this surgery is to reduce spasticity in the lower extremities of children with CP. The children were placed prone on the surgical table with fine-wire electrodes inserted bilaterally into six lower extremity muscles (adductor longus, vastus lateralis, biceps femoris, peroneus longus, tibialis anterior, and medial gastrocnemius). The ventral roots were separated from the dorsal roots. Beginning with the L1 level each dorsal root was separated into four to seven individual rootlets. Each rootlet was electrically stimulated at a rate of 50 Hz while suspended between two probes. The response to this stimulation was recorded in six ipsilateral and three contralateral muscles of the lower extremity. When the response to dorsal rootlet stimulation was excessive compared with normal, it was severed. This procedure was performed for each rootlet at each root level. Approximately 50% to 80% of the rootlets were severed at each level.
Physical Therapy Management
After surgery, the children received inpatient physical therapy twice-daily beginning at bedside on post-op day three. On days four to seven, the children received PT twice daily in the gym. Physical therapy followed a protocol developed by the rhizotomy team. The purpose of this intervention was to improve range of motion, strength, and functional abilities. Emphasis was placed on sitting balance, creeping if appropriate, tall kneeling, half kneeling, sit to stand from floor and chair, standing balance, and ambulation with an assistive device if needed. Developmental exercises were used during strengthening with a focus on isolating quadriceps, hamstrings, gluteus maximus/medius, and ankle dorsiflexors. Upon discharge, the children continued intensive PT in their community four to five times per week (45- to 60-minute sessions) for eight months and three to four times per week thereafter. All children and their families were instructed in a thorough home exercise program to be performed daily.
The methods used in this investigation to determine strength have been reported elsewhere 15,20 and will be briefly summarized. Attempts were made to test both legs of the children with CP on the Kincom. However, if the child was too tired or became uncooperative, only one leg was tested. When both legs could be tested, a single leg from each child was selected randomly for analysis. The randomization was done to avoid bias in selecting the leg to be analyzed. The same legs were tested both pre- and post-SDR. A single, randomly chosen leg was tested in the group of children without disabilities. The child was secured on the seat of the Kincom with a strap at the pelvis and the thigh and the arms free (Fig. 1). The knee joint axis was aligned with the axis of the Kincom lever arm.
The Kincom lever arm, without the limb attached, was brought to a horizontal position and assigned zero degrees to establish a fixed coordinate system. The lever arm was not anatomically referenced to the knee. There was generally a 10- to 15-degree offset between lever arm angle and the actual angle of the knee joint. The lower leg was strapped into the lever arm attachment. With the lower leg secured, the stop angle for knee extension was set based on the length of the hamstring muscles in a seated position or end-range knee extension. The lever arm stop angle ranged from 15 to 23 degrees from the horizontal or approximately zero to 38 degrees knee flexion depending on hamstring tightness and joint contracture. The start angle was set based on the amount of knee flexion possible without pushing the posterior calf against the fixed seat bottom of the dynamometer. This was set at approximately 70 degrees below the horizontal or 80 to 85 degrees of knee flexion.
The speed was set at 10 degrees per second in the passive mode because some children with CP did not have enough strength to initiate movement of the lever arm but could contribute force to a slowly moving lever arm. Passive mode indicates that the lever arm would continue to rotate toward end-range, even when the child was unable to contribute a positive force in that direction. With the lever arm at the start angle (knee flexion), the child was asked to push the leg straight, performing a maximum concentric contraction of the quadriceps. Next, with the knee extended, the child was instructed to bend the knee, performing a maximum concentric contraction of the hamstrings. To encourage maximum effort, the child repeated the test until the current effort could no longer exceed the previous trial, which generally occurred within three to five trials. A standard number of repetitions were not used because it took some children longer to feel comfortable and produce maximal effort, whereas others produced maximal effort on the first trial. The Kincom has an overlay feature that gave the children immediate and continuous visual feedback during their attempts to improve the previous trial. Only the best trial, based on visual comparison by the tester, was saved for analysis. All children understood the task after two trials.
The testing information was downloaded to a personal computer. The effects of the leg and foot weight, estimated at 5.8% body weight, were removed from the recorded torque-angle data. 21 This calculation was required because the standard Kincom gravity correction feature was not used. Gravity correction required weighing the passive lower leg in a horizontal position. This position was difficult for children with CP to attain because of hamstring tightness, hypertonicity, and knee flexion contractures and would have resulted in an overestimation of leg weight. The maximum knee extension and flexion torque values were recorded and normalized to body weight to permit intersubject comparisons (Fig. 2). In addition, the resultant area within the torque-angle curve was calculated producing a work value (ΣT × Δθ where T = torque and Δθ is a small change in angle) for knee extension and knee flexion. If the torque value crossed the zero line during end-range knee extension or flexion, the negative work was subtracted from the positive work to yield a resultant work value (Fig. 3).
An independent t test was used to determine if significant differences existed between groups (p < 0.05). A dependent t test was used for comparison between pre- and postsurgery values (p < 0.05). The pre-SDR strength of the children with CP compared with those of children without disabilities was previously reported, 20 but will be included in this investigation for completeness.
To clarify the results of this investigation, the data of a typical child without disabilities and a typical child with CP pre- and post-SDR will be presented first. The quadriceps strength results for a typical child without disabilities indicated that peak extensor torque occurred when the Kincom lever arm was approximately 50 degrees below horizontal, or when the knee was flexed at approximately 60 to 65 degrees (Fig. 2). The torque decreased gradually as the lever arm moved toward knee extension. Peak flexor torque occurred near the start position with the knee extended and gradually decreased as the knee moved toward greater flexion. As previously reported, all children without disabilities were able to sustain either an extensor or flexor torque throughout the range of movement. 19 When comparing a typical curve of a child with CP pre-SDR to that of the child without disabilities, extension maximum torque for the child with CP (1.28 Nm/kg) was less than the child without disabilities (1.68 Nm/kg), but occurred near the start of the knee extension test (Figs. 2 and 3). The maximum flexion torque for the child with CP (0.36 Nm/kg) was also less than the child without disabilities (0.98 Nm/kg) and occurred near the start of the knee flexion test. Extension and flexion work for the child with CP (0.52 and 0.27 Nm/kg) were both less than the child without disabilities (1.30 and 0.93 Nm/kg). It should be noted that the reduced total range of motion in the child with CP (59 degrees) compared with the child without disabilities (76 degrees) would also reduce the work values. The child with CP was unable to keep the torque values from crossing the zero line as the Kincom lever arm moved toward end-range knee extension. Thus, during the extension test, although this child was attempting to produce a maximum extensor torque throughout the range of motion, the child was unable to maintain an extensor torque and instead produced a resultant flexor torque toward the end-range (approximately 23-degrees flexion). The effect of this crossover is a reduced resultant knee extensor work value.
Comparing pre-SDR results with post-SDR results for the same child with CP, this child demonstrated gains in peak extensor torque (1.28 Nm/kg pre-SDR and 2.20 Nm/kg post-SDR) and extension work (0.52 Nm/kg pre-SDR and 1.02 Nm/kg post-SDR) (Figs. 3 and 4). For pre- and post-SDR knee flexion, the results were inconsistent. There was a gain in peak flexor torque (0.36 Nm/kg pre-SDR and 0.49 Nm/kg post-SDR). However, there was a reduction in flexion work post-SDR (0.27 Nm/kg pre-SDR and 0.19 Nm/kg post-SDR). The reduced work value was attributed to the inability to sustain a flexor torque throughout the range of motion.
Results demonstrated that the group of children with CP had significantly greater strength values (Table 2) post-SDR and rehabilitation compared with pre-SDR for both quadriceps and hamstring peak torque (Fig. 5) and work (Fig. 6). Thus, the hypothesis that a significant increase in strength would occur as a result of this surgery and rehabilitation was supported.
Post-SDR knee extension peak torque and work and flexion peak torque and work values for children with CP remained significantly less than children without disabilities (Table 2). Thus, the hypothesis that children without disabilities would be significantly stronger than children with CP pre- and post-SDR was supported.
The standard deviations associated with all measures for the group of children with CP were large (Table 2). As a result, individual results were considered. Although most children made gains in all strength measures, some of the children had a decline in some of the measures. Of the 19 children with CP in our investigation, the post-SDR percentages had a decline greater than one standard deviation from the mean; 11% in peak extension maximum, 16% in peak flexion maximum, 21% in extension work, and 16% in flexion work. However, no single child demonstrated a decline greater than one standard deviation from the mean in all four strength variables.
The inability to sustain an extensor torque throughout the available range of motion occurred in none of the children without disabilities and in 63% of the children with CP tested pre-SDR. In other words, many of the children with CP crossed the zero-torque line at end-range and were unable to continue to produce positive work. Post-SDR, the number of children with CP unable to sustain an extensor torque throughout the range remained at 63%. The inability to sustain a flexor torque occurred in none of the children without disabilities and 63% of the children with CP pre-SDR. Post-SDR, the inability to sustain a flexor torque throughout the range of motion was improved to 47% of the children with CP.
The purposes of this investigation were to quantify changes in strength of the quadriceps and hamstrings in children with CP as a function of SDR surgery and rehabilitation, and to compare the results for children without disabilities with children with CP pre- and post-SDR. There were several limitations associated with this investigation.
Electromyography (EMG) was not performed during this test. EMG may have helped to demonstrate the reason for the weakness at end-range knee extension. If the hamstrings were active during active knee extension in children with CP, it would effectively reduce the amount of extensor torque produced. Thus, as the knee approached end-range extension, the amount of active resistance from the hamstrings may have increased to a level where the quadriceps could no longer maintain an extensor torque. As a result, a flexor torque was produced causing the torque values to cross the zero line. If the hamstring EMG was absent during knee extension, passive tightness of the hamstrings, a knee flexion contracture or quadriceps weakness may have contributed to the reduced knee extensor torque. Regardless of the reason, until a clear relationship between EMG and torque exists, the reason for the lack of extensor torque produced at end-range cannot be specified to be because of hamstring spasticity, tightness, or lack of quadriceps strength at end-range. The addition of EMG to the test may have offered some insight into which muscle groups were active or passive throughout the range of motion.
Strength and Spasticity/Motor Control
The strength results reported for the children with CP included torque and work, which may have been influenced by spasticity. Spasticity, as noted above, can either limit a desired movement (hamstrings contracting at end-range knee extension) or assist the desired movement (extension synergy-assisting knee extension). Some clinicians believe that strength testing should not be performed where spasticity is present because the results do not represent the individual’s “true strength,” but some combination of strength and spasticity. 22,23 Sussman 22 recommended foregoing objective strength assessments when children with CP lacked motor control. Instead, he suggested that clinicians should make subjective comments about strength based on the ability to squat or kick a ball. The inability to isolate a muscle group out of synergy (poor motor control) in children with spasticity has contributed to a lack of objective strength values for this population.
In the current investigation, we chose to quantify torque production regardless of the influence of spasticity or motor control. Therefore, during this test, no attempt was made to work out of synergy, or to isolate the muscles for knee extension or flexion. As a result, during knee extension, the children with CP often extended the ipsilateral hip into the pelvic stabilizing strap, plantarflexed the ankle, and extended the contralateral knee. Bilateral synergistic movements at the hip and ankle joints also accompanied knee flexion. Although the results may not represent the isolated strength of a muscle group in children with CP, they do represent the torque-producing capabilities about a joint, which may have been influenced by spasticity and a lack of motor control. In addition, if spasticity was contributing to strength pre-SDR, one would expect to find reduced strength post-SDR, which was not supported by our findings. The reduction in spasticity may also improve motor control, and thereby improve the ability to isolate muscles and to produce greater force. Currently no method is available to determine the amount of torque produced as a “synergistic response” to overall leg extension and flexion, from the amount of torque produced by isolated quadriceps or hamstring strength. Despite these limitations, children with CP demonstrated an increase in strength after SDR, a contradiction to current theory.
Isokinetics vs Myometer Results
Children undergoing SDR also received intensive PT after surgery. The intent of the present investigation was to consider the effects of the entire treatment (ie, combined SDR and PT). Therefore, the reported gains in strength may have been due to the surgery, the intensive PT or a combination. In contrast, Steinbok et al 4 conducted a randomized clinical trial which separated a SDR and PT group (n = 5) from a group of children with CP who received only PT (n = 5). The groups were tested before intervention and nine months later, after intervention. The investigators reported no difference in the quadriceps strength results between the two groups at nine months. These investigators additionally reported no significant gains in quadriceps strength in either the SDR and PT group or the PT-only group at nine months. The lack of strength gains in either group is contrary to our results. A potential reason for the disparity may be the use of a myometer to measure isometric strength of the quadriceps at a single joint angle. Quantification of strength at only one joint angle may have prevented the realization of strength gains that may have occurred at other joint angles. For example, Damiano et al 25 reported larger gains in strength in children with CP after a quadriceps-strengthening program as the knee approached extension when compared with 90-degree flexion. Steinbok et al 4 did not report the single angle used during isometric quadriceps testing. If they had tested near 90 degrees of knee flexion, they may not have noted a significant change in strength. We chose to measure peak torque and work isokinetically, to allow an assessment of strength throughout the range of motion.
Variability of Results
A large standard deviation was associated with all strength results in the current investigation, indicating that some children did not demonstrate improved strength post-SDR. Although the majority of children with CP had improved strength post-SDR, it remains unclear if SDR is indicated for all children with CP. A potential reason for some declines in strength is that some of the children may have been weak with little spasticity before surgery and still underwent SDR. The use of objective measures of impairments and functional limitations should improve candidate selection criteria for SDR.
Study of Impairments
Objective functional measures were not assessed in this investigation. The importance of investigating the relationship of strength to function cannot be minimized. The classification scheme proposed by the National Center of Medical Rehabilitation Research (NCMRR) at the National Institutes of Health (NIH) states that to assess the efficacy of a given treatment it is necessary to have outcome measures that encompass many domains. 14 The domains include; pathophysiology, impairment, functional limitation, disability, and societal limitations. The current investigation was designed to determine the outcome of strength examination at the impairment level. Thus, although the results showed a gain in strength, they do not indicate whether the gains in strength resulted in improved function or reduced disability. Functional changes in the amount of assistance required to ambulate were recorded in this investigation.
Gait improvements occurred in five of the 19 children post-SDR (Table 1). Two children who previously required a single crutch/cane to ambulate were able to ambulate independently. Two children who previously used a walker progressed to canes or forearm crutches, and the child who was nonambulatory progressed to ambulating with a walker. Although the findings indicate improvements for some of the children, it remains unclear whether independent ambulators made functional gains in their gait patterns. Additional research is needed in the area of functional outcomes after SDR and rehabilitation.
Strength was tested pre-SDR and eight months post-SDR in the group with CP and only on one occasion in the group of children without disabilities. Without a longitudinal comparison of both groups, some of the strength gains made in the group with CP may have been attributed to maturation. Future work will include a longitudinal comparison.
Most authors have reported a reduction in strength as a result of SDR. 2,7,9 These reports however, were not based upon objective measures. Oppenheim 7 stated that the residual weakness observed in some patients post-SDR was “masked preoperatively by overlying spasticity.” Thus, if SDR reduces spasticity or apparent strength, the child would have underlying muscle weakness. The results of this investigation support the original Steinbok et al 3 study and the publication by Engsberg et al 15 indicating an improvement in strength in children with CP post-SDR. The results demonstrate that, after SDR and rehabilitation, children with CP have improved quadriceps and hamstring strength. The results also demonstrated that children with CP were significantly weaker, both pre- and post-SDR, when compared with children without disabilities. The demonstrated weakness was especially evident as the knee approached full extension and again when approaching full flexion.
IMPLICATIONS FOR FUTURE RESEARCH
To determine the effectiveness of various exercises and surgical procedures, impairments and functional limitations must be quantified. Weakness or a lack of strength has been consistently associated with CP, and is often identified as a cause for atypical and delayed function or a lack of functional gains after intervention. However, because of the complications of spasticity and motor control, strength has rarely been quantified in children with CP. The results of this study demonstrate that children with CP are weak both pre- and post-SDR when compared with children without disabilities. Strengthening both the extensors and flexors of the knee should be a focus of treatment for children with CP. Strengthening exercises should focus on end-range strength as the knee approaches full extension or flexion. For example, a PT might have the patient perform wall squats and focus on maintaining the knee in extension with active quadriceps for a certain time period. Then progress the patient to perform mini wall squats moving from full knee extension to 20 degrees of knee flexion and return to extension. This exercise is only a suggestion to apply the results of this study to clinical practice.
Quantifying strength in CP is the first step in a series of proposals to quantify impairments and learn how and if they are related to function. Future research will include an objective assessment of the strength-spasticity relationship in the lower extremities in subjects with CP, the relationship of these findings to function, and the determination of the effectiveness of the SDR surgery.
1. Fasano VA, Broggi G, Barolat-Romana G, et al. Surgical treatment of spasticity in cerebral palsy. Child Brain 1978; 4: 289–205.
2. Peacock WJ, Arens LJ, Berman B. Cerebral palsy spasticity: selective posterior rhizotomy. Pediatr Neurosci. 1987; 13: 61–66.
3. Steinbok P, Gustavsson B, Kestle JRW, et al. Relationship of intraoperative electrophysiological criteria to outcome after selective functional posterior rhizotomy. J Neurosurg. 1995; 83: 18–26.
4. Steinbok P, Reiner AM, Beauchamp R, et al. A randomized clinical trial to compare selective posterior rhizotomy plus physiotherapy with physiotherapy alone in children with spastic diplegia cerebral palsy. Dev Med Child Neurol. 1997; 39: 178–184.
5. McLaughlin JF, Bjornson KF, Astley SJ, et al. The role of selective dorsal rhizotomy in cerebral palsy: critical evaluation of a prospective clinical series. Dev Med Child Neurol. 1994; 36: 755–769.
6. Engsberg JR, Olree KS, Ross SA, et al. Quantitative clinical measure of spasticity in children with cerebral palsy. Arch Phys Med Rehabil. 1996; 77: 594–599.
7. Oppenheim W. Selective posterior rhizotomy for spastic cerebral palsy. Clin Orthop. 1990; 253: 20–29.
8. Peacock WJ, Staudt LA. Functional outcomes following selective posterior rhizotomy in children with cerebral palsy. J Neurosurg. 1991; 74: 380–385.
9. Arens LJ, Peacock WJ, Peter J. Selective posterior rhizotomy: a long-term follow-up study. Child Nerv Syst. 1989; 5: 148–152.
10. Stern L. The role of physical therapy in selective dorsal rhizotomy. BNI Q. 1995; 11: 19–24.
11. Sienko TS, Aiona MD, Pierce R, et al. Gait changes in children with spastic diplegia after selective dorsal rhizotomy. J Pediatr Orthop. 1996; 16: 747–752.
12. Wright FV, Sheil EMH, Drake JM, et al. Evaluation of selective dorsal rhizotomy for the reduction of spasticity in cerebral palsy: a randomized controlled trial. Dev Med Child Neurol. 1998; 40: 239–247.
13. McLaughlin JF, Bjornson KF, Astley SJ, et al. Selective dorsal rhizotomy: efficacy and safety in an investigator-masked randomized clinical trial. Dev Med Child Neurol. 1998; 40: 220–232.
14. National Institutes of Health. Research Plan for the National Center for Medical Rehabilitation Research
. Bethesda, Md: National Institute of Child Health and Human Development. NIH Publication No. 93–3509:1993.
15. Engsberg JR, Olree KS, Ross SA, et al. Spasticity and strength as a function of selective dorsal rhizotomy. J Neurosurg. 1998; 88: 77–83.
16. McCubbin JA, Shasby GB. Effects of isokinetic exercise on adolescents with cerebral palsy. Adapt Phys Act Q. 1985; 2: 56–64.
17. Kramer JF, McPhail 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.
18. MacPhail HEA, Kramer JF. Effect of isokinetic strength training on functional ability and walking efficiency in adolescents with cerebral palsy. Dev Med Child Neurol. 1995; 37: 763–775.
19. Engsberg JR, Olree KS, Ross SA, et al. Maximum effort resultant knee joint torques in children with cerebral palsy. J Appl Biomech. 1998; 14: 52–61.
20. Parks TS, Gaffney PE, Kaufman BA, et al. Selective lumbosacral dorsal rhizotomy immediately caudal to the conus medularis for cerebral palsy spasticity. Neurosurgery. 1993; 33: 929–934.
21. Clauser CE, McConville JT, Young JW. Weight, volume, and center of mass of segments of the human body. No. AMRL-TR-69–70. Wright-Patterson Air Force Base, Ohio: 1969.
22. Sussman MD. Deficits other than spasticity;consensus. In: Sussman MD. The Diplegic Child Evaluation and Management. Shriners Hospitals for Crippled Children Symposium. Rosemont, Ill: American Academy of Orthopaedic Surgeons; 1992: 111–113.
23. Giuliani CA. Dorsal rhizotomy for children with cerebral palsy: support for concepts of motor control. Phys Ther. 1991; 71: 248–259.
24. Damiano DL, Vaughan CL, Abel MF. Muscle response to heavy resistance exercise in children with spastic cerebral palsy. Dev Med Child Neurol. 1995; 37: 731–739.
© 2001 Lippincott Williams & Wilkins, Inc.