Lower extremity.

[Carmick J. Clinical use of neuromuscular electrical stimulation for children with cerebral palsy, part 1: lower extremity. Phys Ther. 1993; 73.-505-513.]

Key Words: Cerebral palsy, Lower extremity, Motor control, Neuromuscular electrical stimulation, Pediatrics.

Research showing that electrical stimulation can increase the strength[1] of normally innervated muscle has increased interest in neuromuscular electrical stimulation (NMES). Neuromuscular electrical stimulation has been used to improve range of motion (ROM), temporarily reduce excessive spasticity (hyperreflexia),[2] facilitate motor control and muscle reeducation,[3] and assist in gait training.[4] Most studies have been conducted on adult subjects; very little has been written about electrical stimulation with children.

Leyendecker[5] studied 20 children with cerebral palsy whose mean age was just over 10 years. There was no description of the NMES treatment method or measurements, including which muscles were stimulated or whether the children were required to be active. At the beginning of the study, the 10 children who received NMES with neurodevelopmental therapy (NDT)(*) progressed faster than the 10 children who received only NDT. At the end of the study, however, the two groups were equal. Gracanin et al[11] reported the results of a 6-year study of functional electrical stimulation (FES) given to 120 children with cerebral palsy. The peroneal nerve was stimulated during gait in children with varus deformities of the feet who exhibited hyperreflexia. The FES results had a positive effect in the improvement of walking and posture in mild cases in which the child had a varus forefoot, a hypertonic triceps surae muscle, and no hypotonia. Some of the children showed improvement in walking and posture that lasted after the removal of FES. The positive effects were mainly in the swing phase, including absence or decrease of "hikes" (abnormal elevations) and medial (internal) rotation of the hip joint, decrease of equinovarus, and inadequate dorsiflexion at the ankle joint. In the stance phase, the effects were manifested in the decrease of medial rotation and adduction of the hip joint and in hyperextension of the knee joint. The child was able to touch the floor with the heel, which was not possible before application of FES.

Kieklak and DeVahl[12] and Kieklak[13] discussed using NMES as an adjunct to traditional developmental pediatric therapy. Carryover was not mentioned, other than a decrease in muscle tone that was proportional to the frequency of NMES use.

Dubowitz et al[14] reported on the use of chronic electrical stimulation applied to two 3-year-old girls with hemiplegia secondary to cerebral palsy to determine whether motor performance could be improved. The tibialis anterior muscle of the affected leg received stimulation for 1 hour three times a day while the children were active. Motor performance and gait improved after a few weeks of stimulation. One child also showed signs of equilibrium reactions for the first time.

Atwater and colleagues[15] used electromyography (EMG)-driven NMES with 10 children with cerebral palsy, aged from 5.5 to 16 years. The children received EMG-driven NMES to either the wrist extensors or ankle dorsiflexors and a total-body NDT exercise program three times a week for 8 weeks. The children needed to actively contract and then fully relax their muscles. The youngest child, the only one with athetoid cerebral palsy, dropped out of the study because the child did not like the electrical stimulation. The children who remained in the study showed improvement in general motor function; those receiving wrist stimulation improved more than those receiving ankle dorsiflexor stimulation. Results suggested that children needed to be at least 6 years old in order to actively participate in the treatment procedure.

Nothing was found in the literature that described using NMES as an adjunct to the contemporary task-oriented or systems-based model of physical therapy. This report, part 1 of a two-part case report on the clinical use of neuromuscular electrical stimulation (NMES) for children with cerebral palsy, documents the functional changes that occurred with the application of NMES to the lower extremity of three children, 1.6, 6.7, and 10 years of age, all with hemiplegia due to cerebral palsy.

Neuromuscular Electrical Stimulation Used as an

Adjunct to a Task-oriented Physical Therapy Program

The subjects in the case studies presented in this report received NMES, along with a physical therapy program based on a task-oriented model of motor control.[16-18] It was not necessary for the children to be able to actively contract the stimulated muscle, as in the study by Atwater et al.[15] Thus, a 19-month-old child could participate as well as others who are not physically able to isolate muscle contraction. Various techniques were used, with an emphasis on the task-oriented motor model. The NDT techniques were also used where the principles did not conflict.[10] The NMES was used for muscle education, strengthening, increasing sensory awareness, and evaluation.

The principles of motor learning that were emphasized included having the child play in a task-specific activity or a self-directed, goal-driven activity while in a position in which the stimulated muscle would normally function. Pathological movement patterns were considered to be compensations for central nervous system dysfunction and/or biomechanical mechanisms. Neuromuscular electrical stimulation was used to facilitate muscles or movements that were weak, whether they were considered to be spastic or not. The superficial muscle that was felt not to be working at the proper moment was chosen. A remote-control switch was used to stimulate the muscle at the time it was needed.

The children did not have to be in a perfect position and were given freedom to make and learn from mistakes. They were allowed at times to move unsuccessfully and to independently try to correct the movement. They were often given concrete directions in order to assist in motor learning.[18] After success occurred, the stimulus was given intermittently in a weaning process to increase motor learning.

Neuromuscular Electrical Stimulation Equipment Used

The Respond II unit([dagger]) was used. This unit has two channels that allowed two different muscles to be stimulated at the same time or at alternating times. The portability of the unit allowed the child to move freely.

Reusable, self-adhering, pregelled, carbonized-rubber electrodes([dagger]) were used. The active electrode was placed on the motor point, and the inactive electrode was placed on the same muscle or muscle group a short distance away from the involved extremity. The active electrode was cut to size, never smaller than a nickel, so that only the desired muscle would be stimulated and thus eliminate or decrease overflow. The suggestions of Baker and Parker[19] and Delitto and Robinson[20] for electrode placements were often used.

The pulse rate was often initially set to about 5 to 7 pulses per second (pps), which gave a tapping sensation and sensory input but not a fused contraction. When this sensation was tolerated by the child, the setting was increased to between 30 and 35 pps to give a fused muscle contraction. The pulse duration was fixed at 300 microseconds.

The output-limit control was set for 50% of the stimulator's maximum amplitude. The amplitude was increased very slowly, to the child's tolerance, and only when the current was on. It was not necessary to have a maximal contraction to increase sensory awareness or the strength of weak muscle[2,21] (see part 2 of this case report in this issue). With a low pulse rate, a higher amplitude is usually comfortable.[3]

The ramp, or rise time, was initially set at the lowest setting (8 seconds) for the most comfort. When the child was accustomed to NMES, the ramp was set for approximately 2 seconds. When a quicker response was needed, as for gait activities, the fastest rise time (0.5 second) was used.

On-off times initially were set for 10 seconds on and 25 seconds off. Once the child was comfortable with electrical stimulation and not showing signs of fatigue, the cycle was set at 15 seconds on and 15 seconds off for a total of 15 to 20 minutes. When the remote-control switch was used, the on time was determined by activation of the switch, which was stimulated at the time the child was seen to need that muscle to function. For example, when the child in case 1 was walking and not flexing the knee, the remote-control switch was used to facilitate the NMES, which was stimulating the medial hamstring muscle. When stimulating the lower extremity and while the child remained standing, not moving out of position, the stimulus control was set to the alternating mode to prevent the muscle from being constantly stimulated or constantly silent.

Introducing Children to Nouromuscular Electrical Stimulation

Initially, NMES was demonstrated on the therapist's arm and then on the parent's arm. This was to give the parent experience with NMES to help eliminate any fear of electrical stimulation that might be transferred to the child.

The patients in cases 1 and 2 were first introduced to a hand-held vibrator, which was called a "tickler." The child in case 3 was 10 years of age, had experience with EMG, and did not need this elementary level of introduction. The NMES was described as being like the "tickler," and the children were told they would feel a tapping sensation. The electrode was then placed on the cleaned body part. The amplitude was increased very slowly until the child indicated that he felt the stimulation or until a trace contraction was felt or seen.

The child was given time to experience and accept the sensation. Letting the child in case 2 turn the unit on and off with the remote-control switch gave him a chance to control what was happening to his body. Once he had a turn at controlling the mechanics of the NMES unit, he was then willing to return to the actual therapy. This took little time as he quickly showed interest in returning to his therapeutic activity. With acceptance and experience, the amplitude was increased as needed and tolerated to improve the muscle function. Acceptance of NMES is expected if the child's needs are respected, the amplitude is kept within the child's tolerance, and the child is given some control over the situation.

Case Studies of Three Children With Cerebral Palsy, Hemiplegic Type

Case 1

The child in case 1 was a 1.6-year-old boy who was diagnosed with left hemiplegia secondary to cerebral palsy following a computerized tomography CT) scan. The lesion was believed to have occurred prenatally. He began receiving physical therapy at age 7 months. Physical therapy followed the task-oriented model of motor learning and included NDT techniques and various therapeutic techniques to increase sensory awareness. He walked at age 15 months. The child initially received NMES when he was only 19 months of age. He received NMES to both the lower and upper extremities.

Before neuromuscular electrical Stimulation. At age 21 months, the child toe-walked on the left foot with excessive lateral (external) rotation of the leg. He stood with a pronated foot. Passive range of motion (PROM) was complete throughout the left leg and left hip. No active dorsiflexion of the left foot was possible. Passive dorsiflexion was 15 degrees with knee extension.

Treatment of the tibialis anterior MUSCLE. Neuromuscular electrical stimulation was used on the motor points of the child's tibialis anterior muscle when he was 21 months of age. This treatment was initially done while the child was in a sitting position. in the second session, NMES was applied while the child was standing and walking. The remote-control switch was used to stimulate the tibialis anterior muscle during the swing phase of gait.

When the child was standing and walking, there was no real gait change. He intermittently used a flat foot, but only slightly more than without NMES. No active dorsiflexion occurred when requested. There was little or no carryover. This treatment was tried intermittently once weekly for 5 months without success. As the child became heavier and more active, his gait deteriorated. There was less weight shift to the left and more toe-walking, and the child developed genu recurvatum.

Treatment of the triceps surae muscle. At 26 months of age, after 5 months of stimulation of the tibialis anterior muscle, the triceps surae muscle was added to the child's NMES lower-extremity program. The muscles were alternately stimulated while the child was standing and walking with a remote-control switch to achieve the correct timing. It was discovered on the second session that when the NMES was facilitating the gastrocnemius-soleus muscle group, the child's foot was plantigrade. The child's gait had intermittent improvement of foot flat on initial contact, but the improvement did not last after NMES was removed. The program was done for 15 minutes at each of four weekly sessions conducted over a 5-week period. The child's gait improved intermittently, but the ROM of the foot was not improved.

At 29 months of age, when only the triceps surae muscle was stimulated, the child's foot immediately became plantigrade. The NMES to the triceps surae muscle was used for about 5 minutes while the child was standing. The tibialis anterior muscle was then added for alternate stimulation while the child walked and stepped in and out of hoops for a total of 15 minutes with the use of the remote-control switch. The child's gait was not changed, and his foot was plantigrade when NMES stimulated the gastrocnemius-soleus muscle group.

For the next 5 weekly sessions, the triceps surae muscle was stimulated for 15 minutes without stimulation of the tibialis anterior muscle. The child stood plantigrade more frequently, and after 3 weeks, he was walking with foot-flat gait and shifting his weight to the left. Although he was doing well without NMES stimulating the tibialis anterior muscle, that muscle was again stimulated for fear that the muscle was not functioning well enough. After that session, the child returned to toe walking for 2 days.

When only the triceps surae muscle was stimulated, the child's foot immediately became plantigrade during stance and he had a foot-flat gait. Fear of increasing reflex activity in the triceps surae muscle if that muscle was stimulated alone without intermittently stimulating the tibialis anterior muscle led to continued stimulation of the tibialis anterior muscle. A foot-flat gait, however, was not achieved with the alternating stimulation of the tibialis anterior and triceps surae muscles. The foot-flat gait occurred only when the triceps surae muscle was stimulated. The child's PROM improved throughout the foot and ankle. Active dorsiflexion was intermittently improving. There was no increase in reflex activity. The child's gait improved with triceps surae muscle NMES, and he used a flat foot and occasionally a heel-strike. With NMES, he would spontaneously shift weight to the left side and maintain that weight shift for several minutes while playing at a table. The NMES needed to be repeated each week, as his gait improvements lasted a few days and then decreased. After success with the triceps surae muscle, the gluteus maximus muscle was stimulated to increase strength. When this was done, the child would shift weight to the left side while standing and playing at a table. He did not need verbal direction.

Treatment of the medial hamstring muscle. When the child learned to run, it turned into a gallop because of the excessive retraction of the left side of the body; he led with the right extremities. The photograph shown in Figure 1 was taken while the child was running. His usual body position is shown when walking or running; the left extremities, hip, and shoulder were held and remained behind the right hip and shoulder.

During physical therapy, an attempt was made to correct the compensating trunk rotation and improve symmetry. Two adults each grasped one of the child's hands and rotated his body so that the left extremities would lead. He strongly refused by pulling down on the arms, rotating his body to the right and starting to cry. Continuing with this attempt was unwarranted; the sensation of leading with the left extremities was extremely upsetting to the child. It was decided to try NMES to correct the unwanted rotation.

Electrodes were placed on the child's medial hamstring muscle to balance the excessive use of the lateral hamstring muscle thought to cause the excessive lateral rotation of the left leg. Neuromuscular electrical stimulation was used as the child played while standing at a bench. He showed total acceptance of the stimulus. After a few minutes, he was again encouraged to gallop, with NMES still operating, leading with the left side as two adults each held one hand.

When NMES was applied, galloping and leading with the left extremities became immediately possible. The child enjoyed the activity and for a few times galloped about 4.6 m (15 ft) at a time. The manual assistance was removed, and he was able to gallop unassisted, leading with the left (involved) side.

The entire time of stimulation to the medial hamstring muscle was 15 minutes. The cycle was set for 15 seconds on and 15 seconds off while the child was galloping. When he walked, the remote-control switch was used to stimulate the medial hamstring muscle in the normal, age-appropriate timing in accordance with Sutherland and colleagues'[22] EMG charts.

Following this treatment, the child's gait improved; the hip moved symmetrically, and the left leg was no longer excessively laterally rotated. Carryover was maintained for 2 days, as reported by his mother. The NMES was given on the return visit in the same manner. The photograph shown in Figure 2 was taken after two sessions of NMES applied to the medial hamstring muscle. It shows improved rotation of the left lower extremity and shoulder while the child was walking. After only two 15-minute sessions, he was able to maintain the improved alignment without further stimulation. He was wearing a supramalleolar foot orthosis.

The extent of carryover is shown in the photograph in Figure 3, which was taken while the child was running, 4 weeks after the last time the medial hamstring muscle was stimulated. Figure 3 shows that he was still able to run while alternating his arms and legs; the left leg and right arm were leading at that moment in the gait cycle. It had not been necessary to repeat the treatment at that time because the child was able to maintain the skill.

Case 2

The child in case 2 was a 6.7-year-old boy when he began receiving NMES to both the upper and lower extremities. He was diagnosed as having cerebral palsy attributable to sporadic oxygen deprivation at birth. A CT scan showed damage in the left hemisphere. At 10 months of age, he began twice-weekly sessions of traditional NDT physical therapy at this clinic and occupational therapy at a local clinic. When he was 5.5 years of age, he moved to a new location where the frequency of his program was decreased to once weekly.

The child in case 2 was beyond the age that physical therapy is expected to be beneficial.[23] In order to document changes, the Physiological Cost Index (PCI)[24] was measured, and pedographs and gait videos were taken. The PCI is a measure of locomotor efficiency, and a reduction in PCI indicates improved efficiency.

At age 6.7 years, when the child returned for a trial of NMES, the right foot had full passive dorsiflexion while the foot was in subtalar neutral position. Active dorsiflexion was limited to 2 or 3 degrees from a plantarflexed position with knee flexion. When standing, the right foot was markedly pronated. The child walked with a right drop foot and tended to use toe touch at initial contact. He wore a right supramalleolar foot orthosis and a left University of California, Berkeley (UCB) orthosis. A Shutrak pedograph paper([double dagger]) evaluation showed that the left, stronger foot was rotated medially 13 degrees, often crossing in front of the right foot.

The child received NMES for both the right arm and leg. During summer vacation, he was treated for 6 weeks, for a total of 10 visits, with a 2-week break in the middle. The NMES was used initially to stimulate the child's tibialis anterior muscle. The triceps surae muscle was also stimulated after the second week. As the child walked, the tibialis anterior muscle was stimulated with the remote-control hand switch during swing. The triceps surae muscle was stimulated with the release of the switch at foot-strike. The normal immature triceps surae muscle gait pattern[23] was used to determine when to stimulate the leg muscles. Neuromuscular electrical stimulation was used during 15 minutes of walking outdoors, where the child chose the pace.

Following treatment with NMES, the child's PCI[24] measurements showed a factor of four reduction (Tab. 1), indicating a large improvement in walking efficiency. [TABULAR DATA 1 OMITTED]

The child demonstrated increased active range of motion (AROM) of the right foot after receiving NMES. He exhibited from 2 to 3 degrees of active dorsiflexion from a plantarflexed position with knee flexion and from 8 to 10 degrees of active dorsiflexion beyond 90 degrees with or without knee flexion.

Pedographs of the child's left foot showed a change of 13 degrees of medial rotation to 3 degrees of lateral rotation, a 16-degree total change (Tab. 2). Heel-strike occurred at foot contact following treatment with NMES, and the child was able to progress from a right supramalleolar foot orthosis to a UCB orthosis. [TABULAR DATA 2 OMITTED]

The child succeeded in riding a two-wheel bicycle without training wheels. He had tried unsuccessfully to ride a two-wheel bicycle without training wheels for many months before being treated with NMES. According to his parents, he could not balance well enough to stay on the bike to get it moving. After using NMES, he was strong enough to push the pedal with his right foot to get the bike moving. Once it was moving, he could ride and maintain his balance. Increased strength in plantar flexion gave the needed push.

Carryover remained for 1 year after stopping NMES and physical therapy. On return to the school clinic, following the 10 sessions of NMES, his clinic physician discontinued physical therapy. Occupational therapy was continued.

Case 3

The patient in case 3 was a 10-year-old boy at the time NMES was begun. He had left hemiplegia, secondary to cerebral palsy due to prematurity, which affected the lower extremity and trunk more than the upper extremity. His left hand was functional, and no therapy was given to the upper extremity.

The child participated in a brief program of physical therapy during the previous year and was, at the time, on a home program to improve weight shifting to the left side and general strength and coordination. He had also been receiving EMG biofeedback for the tibialis anterior muscle once weekly at another clinic for a year to try to improve active dorsiflexion.

The patient was 10 years of age at the start of the 2-month program with NMES and thus beyond the age that physical therapy is expected to be beneficial.[23] In order to document changes, the child's PCI[24] was measured, and pedographs and gait videos were taken.

Full PROM of the boy's involved foot was present at the beginning of the NMES sessions. Active dorsiflexion was limited to a few degrees from a plantar-flexed position with knee flexion. He was active in sports and attended a regular school program. He wore a left supramalleolar foot orthosis and walked with an obvious vault of the right foot. He had excessive bilateral knee flexion. On initial foot contact, there was approximately 15 degrees of flexion, which increased to 20 to 30 degrees on the loading response. Flexion was more excessive in the right uninvolved knee than in the left involved knee. The child could balance while standing on the right foot with hands on his hips, but he could not do this on the left foot. He was not able to stand on both feet and balance on a large platform swing without strongly holding on with both hands, and his weight shift to the left was inadequate.

On the initial visit, NMES was applied to the motor points of the tibialis anterior and triceps surae muscles reciprocally for a total of 15 minutes while the child was in a sitting position. He was encouraged to actively assist the contractions. When walking, the remote-control switch activated the tibialis anterior muscle during swing, and when released at heelstrike, the triceps surae muscle was stimulated. Although the child was well over 7 years of age, the immature EMG activity pattern[22] was chosen, as it was felt that he had not experienced this pattern. The time off and on was determined by the timing of the gait. After a return visit to evaluate the home program, the child came once monthly for physical therapy. His mother was instructed in the procedure, and NMES was used at home an average of three times a week for 15 minutes during ambulation. Active dorsiflexion and plantarflexion exercises were also performed over the 2-month treatment period.

Following the 2-month period of treatment, the child's PCI measurements showed a factor of two reduction (Tab. 1), indicating a significant improvement in walking efficiency. Active dorsiflexion improved from 3 degrees of movement in a plantarflexed foot position with a flexed knee to 5 degrees of dorsiflexion beyond 90 degrees with a straight knee in the long-leg sitting position.

The child obtained a smoother gait, with a more symmetrical posture. Vaulting on the right foot was no longer obvious, and the lateral trunk deviation was gone. His balance also improved. He was able to stand on a large mobile four-point swing and tilt it in every direction without holding on with his hands. He was able to put as much weight on the left leg when shifting to the left as he did when shifting to the right side.

The circumference of the boy's left (involved) calf increased 1.3 cm (0.5 in), and that of the right (uninvolved) calf increased 0.6 cm (0.25 in) in the 2-month treatment period. His mother reported a noticeable increase in endurance, as he was able to spend more time in sports. He had an apparent increase in strength in plantar flexion. His improved gait pattern is believed to be due to increased strength of the left leg with increased knee extension.

Physiological Cost Index measurements were taken while the child walked 100 m at a normal pace. The average step length was 58.6 cm/step before NMES and 60.2 cm/step after NMES. His walking speed also increased after NMES (Tab. 1). The cadence was 113 steps/min before NMES and 116 steps/min after NMES.

Pedographs (Tab. 2) show a decrease in stride and step length, perhaps because, as noted at the time, the child slowed down when he reached the starting edge of the paper. Step width narrowed, perhaps due to improved balance. The record also showed slightly improved lateral rotation of the right foot, which could be due to increased strength in the left plantar flexors. Perhaps the increased strength of the left leg allowed time to increase lateral rotation of the right leg.

Discussion

The child in case 1 showed immediate changes, with carryover in his gait, when NMES was used to stimulate the medial hamstring and triceps surae muscles, but NMES did not help when given to the tibialis anterior muscle. Stimulation of the tibialis anterior muscle gave a slight momentary correction to the gait, but this change was not as dramatic as it was when the triceps surae or medial hamstring muscle corrected the gait. After the tibialis anterior muscle was stimulated, the child walked more on the toe than when only the triceps surae muscle was stimulated. Dubowitz et al[14] and Atwater et al[15 ]also used the tibialis anterior muscle for stimulation without dramatic or immediate results.

Berger and colleagues' EMG studies[25] have shown that the tibialis anterior muscle is active throughout the gait cycle in the immature gait pattern and that the magnitude of the tibialis anterior muscle's EMG activity does not change with the maturation of gait. In nondisabled children, the gastrocnemius muscle's EMG magnitude begins to increase in strength at around 2 years of age and reaches maximal strength at about 4 or 5 years of age. The muscle's EMG magnitude is then about twice as high as that of the tibialis anterior muscle. These EMG studies have also shown that children with cerebral palsy walk with the immature gait pattern of nondisabled 1-year-old children. That is, they walk with a reduced magnitude of gastrocnemius muscle EMG activity, and the tibialis anterior muscle is active throughout the whole step cycle. It may be that lack of coordination and strength of the triceps surae muscle is a significant problem with the walking pattern of children with cerebral palsy and not the lack of tibialis anterior muscle contraction. The triceps surae muscles of the children in cases 1 through 3 were treated with NMES with no resulting increase in spasticity. Instead, increased ease of PROM of the foot occurred, and the children later learned to actively dorsiflex the foot. In my clinical experience, eight out of eight children with a variety of types of cerebral palsy immediately moved from a plantar-flexed to a plantigrade position with stimulation of the triceps surae muscle.

More time was needed to see improvements in the children in cases 2 and 3. Once the triceps surae muscle of the child in case 1 was stimulated, he immediately became plantigrade. The child in case 2 achieved full dorsiflexion AROM after seven sessions. The child in case 3 achieved full dorsiflexion AROM after 2 months of home sessions conducted three times per week. It appears that the early use of NMES before a pattern becomes too strong a habit makes success much more likely. The older children may have needed more time to build enough strength in the triceps surae muscle to lift their weight. It has been shown that the plantar flexors are normally responsible for two thirds of the total concentric work of gait, whereas the power generation of the ankle plantar flexors in children with cerebral palsy provides only one third.[26]

Another concept concerns stimulating just the triceps surae muscle and not the tibialis anterior muscle. The child in case 1 did not progress with heelstrike or a flat foot until the tibialis anterior muscle was not stimulated. Perhaps the older children would have progressed faster if the triceps surae muscle alone had been stimulated, and then later, if the tibialis anterior muscle was seen to be weak, it could have been stimulated. At the time, the tibialis anterior muscle appeared to be weak in the older children, as the leg appeared somewhat atrophied compared with the uninvolved leg.

It was not expected that the children in cases 2 and 3 would improve in gait due to their age and current skill level. Bleck[23] cites various physicians and researchers who have found that, for children with cerebral palsy, motor gains reached a plateau or ceased between the ages of 6 and 7 years. The child in case 2, at almost 7 years of age, and the child in case 3, at 10 years of age, however, showed significant improvement after a short time with NMES.

The child in case 3 had been treated with biofeedback EMG to the tibialis anterior muscle without significant change. It was only when alternating NMES was applied to the triceps surae and tibialis anterior muscles that complete active dorsiflexion occurred and strength and gait efficiency increased in plantar flexion and dorsiflexion.

The child in case 2 had received 4 years of NDT physical therapy and occupational therapy without NMES. The only change in his program that occurred during the 6-week period was the addition of NMES with the task-oriented model of motor control. I believe that the use of NMES was very important in improving the child's sensory awareness and strength, which resulted in increased coordination.

Summary and Conclusions

Following a program of NMES, three children demonstrated rapid functional changes. The children in cases 2 and 3 showed significant improvement in gait efficiency and balance in a short time even though they were past the age when this improvement was expected. The child in case 2 maintained the improved gait for 1 year after stopping physical therapy. The child in case 1 showed immediate and more rapid results, with greater carryover, than did the older children. It is possible that the abnormal movement strategies were not as habitual in the younger child and biomechanical changes had not yet occurred in the muscles, making change easier to obtain.

Preliminary findings suggest that NMES may be a useful physical therapy tool when used with task-oriented functional activities. Electrical stimulation may enhance muscle contraction and provide sensation so that a child can add a weak response with effective results and assist in improving motor control.

Acknowledgments

I thank Cathleen Pewthers for her initial suggestion to try NMES when biofeedback EMG was unsuccessful; the subjects and their parents for allowing the clinical results to be shared; and Medtronic Nortech Division for the loan of one of the NMES instruments, together with the necessary disposable electrodes, used in this research. (*) Traditional NDT[6,7] follows a hierarchial and reflex model[8,9] of motor control. Abnormal musc tone (spasticity) is considered a major problem for treatment and a cause of much of the motor deficit. Patients are advised not to exert excessive effort, which is believed to be a potential for increasing abnormal muscle tone and to result in loss of reciprocal movement activity. Development was expected to progress from proximal to distal, so the hip was expected to develop before the foot. Current NDT assumptions[10] are changing as new knowledge of motor control and motor learning is discovered.

References

[1] Delitto A, Snyder-Mackler L. Two theories of muscle strength augmentation using percutaneous electrical stimulation. Phys Ther. 1990; 70:158-164. [2] Alfieri V. Electrical treatment of spasticity. Scand J Rehabil Med. 1982;14:177-182. [3] Devahl J. Neuromuscular electrical stimulation in rehabilitation. In: Gersh MR, ed. Electrotherapy in Rehabilitation. Philadelphia, Pa: FA Davis Co; 1992:218-268. [4] Carnstam B, Larsson L, Prevec TS. Improvement of gait following functional electrical stimulation. Scand J Rehabil Med. 1977;9:7-13. [5] Leyendecker CH. Electrical stimulation therapy and its effect on the general activity of motor-impaired cerebral-palsied children: a comparative study of the Bobath physiotherapy and its combination with the Hufschmidt electrical stimulation therapy. Rehabilitation. 1975; 14:150-159. [6] Gorden J. Assumptions underlying physical therapy intervention: theoretical and historical perspectives. In: Carr JH, Shepherd RB, Gorden J, et al, eds. Movement Science Foundation for Physical Therapy in Rehabilitation. Rockville, Md: Aspen Publishers Inc; 1987:1-30. [7] Valvano J, Long T. Neurodevelopmental treatment: a review of the writings of the Bobaths. Pediatric Physical Therapy. 1991; 3:125-129. [8] Horak FB. Assumptions underlying motor control for neurologic rehabilitation. In: Contemporary Management of Motor Control Problems: Proceedings of the II Step Conference. Alexandria, Va: Foundation for Physical Therapy Inc; 1991:11-27. [9] Atwater SW. Should the normal motor developmental sequence be used as a theoretical model in pediatric physical therapy? in: Contemporary Management of Motor Control Problems: Proceedings of the II Step Conference. Alexandria, Va: Foundation for Physical Therapy Inc; 1991:89-93. [10] Bly L. A historical and current view basis of NDT. Pediatric Physical Therapy. 1991; 3:131-135. [11] Gracanin F, Vrabic M, Vrabic G. Six years of experiences with FES method applied to children. Europa Medicophysia. 1976; 12:61-68, [12] Kieklak H, Devahl J. Protocol for pediatric applications. In: Medtronic Bulletin on Use of the Respond II. San Diego, Calif: Medtronic Nortech Division; 1986. [13] Kieklak H. Use of neuromuscular electrical stimulation (NMES) for children with neuromuscular disorders. Tot Line (publication of the Americal Physical Therapy Association Section on Pediatrics). 1987;13:27. [14] Dubowitz L, Finnie N, Hyde SA, et al. Improvement of muscle performance by chronic electrical stimulation in children with cerebral palsy. Lancet. March 12, 1988:587-588. Letter. [15] Atwater SW, Tatarka ME, Kathrein JE, Shapiro S. Electromyography-triggered electrical muscle stimulation for children with cerebral palsy: a pilot study. Pediatric Physical Therapy. 1991;3:190-199. [16] Carr JH, Shepherd RB, Gorden J, et al. Movement Science Foundation for Physical Therapy in Rehabilitation. Rockville, Md. Aspen Publishers Inc; 1987. [17] Carr JH, Shepherd RB. A motor-learning model for rehabilitation of the movement disabled. Ada L, Canning C, eds. Key Issues in Neurological Physiotherapy. Oxford, England: Butterworth-Heinemann Ltd; 1990:1-24. [18] Van der Weel FR, Vandermeer ALH, Lee DN. Effect of task on movement control in cerebral palsy: implication for assessment and therapy. Dev Med Child Neurol 1991; 33:419-426. [19] Baker LL, Parker K. Neuromuscular electrical stimulation of the muscles surrounding the shoulder. Phys Ther. 1986;66:1930-1937. [20] Delitto A, Robinson AJ. Electrical stimulation of muscle: techniques and applications. In: Snyder-Mackler L, Robinson AJ, eds. Clinical Electrophysiology. Baltimore, Md: Williams & Wilkins; 1989:97-138. [21] Kahanovitz N, Nordin M, Verderame R, et al. Normal trunk muscle strength and endurance in women and the effect of exercises and electrical stimulation, part 2: comparative analysis of electrical stimulation and exercises to increase trunk muscle strength and endurance, Science. 1982;216:112-118. [22] Sutherland D, Olshen R, Biden E, Watt M. The Development of Mature Walking. Philadelphia, Pa; Mac Keith Press; 1988:158. [23] Bleck E. Orthopaedic Management in Cerebral Palsy. Philadelphia, Pa: Mac Keith Press; 1987. [24] Butler P, Engelbrecht M, Major RE, et al. Physiological cost index of walking for normal children and its use as an indicator of physical handicap. Dev Med Child Neurol. 1984; 26:607-612. [25] Berger W, Altenmueller E, Dietz V. Normal and impaired development of children's gait. Human Neurobiology. 1984;3:163-170, [26] Olney SJ. Work and power in hemiplegic cerebral palsy gait. Phys Ther. 1990;70:431-438.