Orthotic devices for ambulation in children with cerebral palsy and myelomeningocele.

LM Knutson, PhD, PT PCS, is Lecturer, Physical Therapy Graduate Program, and Senior Physical Therapist, Division of Developmental Disabilities, The University of Iowa, 2600 Steindler Bldg, Iowa City, IA 52242 (USA). Address correspondence to Dr Knutson.

DE Clark, BA, CPO, is Prosthetist/Orthotist and Owner of Dale Clark Prosthetics in Waterloo, Coralville, and Dubuque, Iowa. He is also on the Board of Directors of the American Board for Certification in Orthotics and Prosthetics.

The prevalence of cerebral palsy (CP) at school entry is 2.0 per 1,000 live births.(1-3) The school-entry prevalence for myelomeningocele (MMC) may be slightly less than the MMC birth incidence of 1.0 per 1,000.4 Collectively, approximately 3.0 children per 1,000 have one of these disabilities. Projecting from the statistics that 4,021,000 children were born in the United States in 1989,(5) over 12,000 children in a given year can be expected to be affected by one of these neurological disabilities. A majority, of these children will have orthotic needs for which the attention of a physical therapist will he required. In a public health survey on payment for services, 54% of the 380 children with mixed diagnoses studied used braces.(6) Because so many children will use orthoses, physical therapists should be knowledgeable about orthotic programming.

The purpose of this article is to review contemporary practice in orthotic management, focusing on devices used for ambulation in children with CP and children with MMC. These two patient populations were chosen for four reasons. First, these children commonly need orthoses to walk. Second, their needs extend through and change during their lifetime. Third, their orthotic and gait training programs frequently fall under the direction of a physical therapist. Fourth, the orthotic management of these children can often be generalized to the care of children with other disabilities. This article is divided into sections organized to review the literature on orthotic devices for walking, present principles of lower-extremity orthoses, discuss orthotic design, and consider criteria for selecting orthotic devices according to patient needs.

Review of Literature

Although numerous references to and discussions of orthotic devices can he found in the literature, there are fewer reports of an investigative nature. Only recently have reports of studies been published on the effects of orthotic devices for children with disabilities. Table I summarizes key studies of pediatric orthotic use.

Cerebral Palsy

The use of orthoses in the management of children with CP has varied from the discussion by Little of "mechanical apparatus" in the mid-1800s and advocacy by Phelps and Deaver between 1930 and 1950 to avoidance of orthotic devices, particularly full-control orthoses, during the peak of the neurodevelopmental treatment (NDT) era.(7,8) Today, a new enthusiasm for the use of orthoses has surfaced, specifically for below-knee devices used by children who are ambulatory.(9)

A 1985 report on a group of English children with CP born between 1970 and 1974 states that 50% to 75% of the children were ambulatory.(2) Because a majority of children with CP will use orthoses to assist walking, research that complements orthotic selection or adds to an understanding of orthoses is helpful. in one such recent study, Mossberg and colleagues(10) evaluated walking velocity and heart rate in 18 children with spastic diplegia walking with and without plastic ankle-foot orthoses (AFOs). Use of orthoses increased walking velocity and reduced the heart rate compared with walking without AFOs; however, the differences were not statistically significant.

Fifteen years ago, steel or aluminum braces with adjustable ankle joints were common. Although today metal braces are rarely used for children with CP, the purported advantages of polypropylene orthotic designs have not been well studied. Brodke and co-workers(11) evaluated five nondisabled children walking barefoot and using both traditional metal AFOs and fixed polypropylene AFOs. Walking speed and cadence were observed to decrease with either AFO compared with the barefoot condition; however, the decrease was more pronounced with the metal AFOs. The duration of quadriceps femoris muscle electromyographic (EMG) activity was greater with molded plastic AFOs than with metal AFOs and was greater for both braces than for barefoot walking. Changes in lowel-extremity (ie, hip and knee), motion were not significantly different, but maximum knee extension during stance tended to be greater with polypropylene AFOs than with metal AFOs or barefoot walking. Both types of orthoses effectively limited ankle motion. The researchers concluded that metal AFOs impaired normal gait more than did polypropylene AFOs.

Polypropylene orthoses were first introduced in the 1960s.(12) Their use in children with CP did not become established until the mid- to late-1970s when clinical experience and satisfaction gradually dispelled views that plastic orthoses would increase spasticity (excessive contraction of the muscle and stiffness that prevents or opposes normal muscle lengthening) or would not be sufficiently strong to support the spastic (affected by spasticity) limb. The first designs of polypropylene orthoses, as described by Hoffer et al(13) in 1974, were strictly of the fixed design.

By the early 1980s, fixed polypropylene AFOs were being molded with features designed to hold the subtalar joint in neutral and inhibit spasticity.(14) The "subtalar-neutral" AFO was intended to oppose foot valgus or pronation during weight bearing, a problem that occurs frequently in children with spastic diplegia or quadriplegia.(15,16) The concept of subtalar-neutral alignment and "tone reduction," or reduction of the state of muscle contraction, was first introduced in inhibitive casts(17,18) and was later applied to orthoses.(14,19) Descriptive reports have been supplemented by case reports(20,21) and single-subject design research.(22-24) Hinderer and co-workers(22) found tone-reducing casts yielded a longer stride length than standard casts in two children under 6 years of age with CP.(22) Harris and Riffle(23) found inhibitive AFOs (right supramalleolar orthosis and left AFO) increased standing balance time in a 4 1/2-year-old child with quadriplegia. Embrey and colleagues(24) demonstrated knee flexion could be reduced during gait in a 2 1/2-year-old child who had hypotonia by the use of NDT and "ankle-height orthoses."

Early contoured AFO designs did not allow ankle motion. Recently, however, hinged polypropylene AFOs have been introduced and have received positive responses by physical therapy clinicians. Middleton and colleagues(21) reported the first comparison of fixed AFOs with hinged AFOs by studying a 4 1/2-year-old child with spastic diplegia. The child's pattern of ankle dorsiflexion was more normal and the knee muscle moments were lower during the stance phase of gait with the hinged AFOs than with the fixed AFOs. The investigators concluded that hinged AFOs appear to be more effective than fixed AFOs for treating children who have spastic CP.

Without replication, the results of the case reports and single-subject studies cannot be generalized to other children. Some investigators, however, have studied groups of patients using inhibitive short leg casts, (25,26) splints,(27) and AFOs.(28) In studying two groups of children before and after 10 weeks of NDT, Bertoti(25) found the group of eight children wearing short leg casts had a 27% increase in stride compared with a 13% decrease in stride for the group of eight children who were uncasted. Watt and colleagues examined the long-term effects of a 3-week treatment course involving inhibitive casting and NDT. Whereas improvements in passive ankle dorsiflexion and foot-floor contact when walking were seen 2 weeks after treatment, these changes were less evident after 5 months. Specifically, right ankle dorsiflexion showed an increase of 8.8 degrees 2 weeks after cast removal but only 3.6 degrees after 5 months. Right foot-floor contact was seen in 64% of the patients 2 weeks after casting and in 39% of the patients 5 months after casting. studied the EMG and range-of-motion (ROM) effects of inhibitive splints on eight adolescent and young adult subjects who had neurological insults. Compared with the presplint condition, no effects on integrated EMG activity during splint use were noted, but the ROM into extension increased. In a study by Lough,(28) the effects of no orthoses, fixed AFOs, and hinged AFOs were contrasted in 15 children with CP. Mean walking velocity increased by 5.6 cm/s, and ankle dorsiflexion increased during mid-stance of gait by 4.9 degrees for hinged orthoses compared with no orthoses. The EMG amplitudes were examined at 5% intervals across an ensemble average created for each of four muscles. Of 80 possible intervals of significant difference, only 3 were different-2 for the vastus lateralis muscle and 1 for the tibialis anterior muscle. No differences were found among the treatments for the medial hamstring or gastrocnemius muscles. The reduction in foot pronation, as determined from standing dorsoplantar radiographic measurements of the talonavicular angle, was significant for fixed AFOs (4.5[deg] less) and approached significance for hinged AFOs (3.6[deg] less) compared with no orthoses.

Myelomeningocele

The prognosis for ambulation by children with MMC is related to level of motor function. The effect of the spinal cord dysfunction level on energy expenditure was well documented in adults with paraplegia by Clinkingbeard and associates(29) and is in accord with reports of function for children with MMC.(30-38) Children with higher levels of spinal defect (eg, at and above L-2) require more extensive orthotic support and can he expected to expend more energy than children with defects below L-2. Table 2 summarizes reports on the status of and prognosis for ambulation in children with MMC. Consensus generally exists that the higher the level of defect, the smaller will be the percentage of the group who are ambulatory. Other factors influencing ambulation include mental retardation, muscle power within the level of defect, orthopedic deformities, hydrocephalus, and walking status at age 7 years.(30-39)

Whether children with high lumhar-and thoracie-level defects are provided early instruction for walking appears to he influenced by practice patterns of the geographic area. Although reports agree the majority of children with thoracic MMC who walk will later become nonambulatory, the study by Mazur and colleagues(37) Supports the value of bracing and walking. Subjects who walked as children had fewer fractures and pressure sores, were more independent, and were better able to transfer than were patients who used a wheelchair from early in life. Sankarank-utty et al(41) stressed the importance of effective bracing and suggested that modern orthoses can prolong ambulation in patients who otherwise may discard their devices. This notion was reinforced by Flandry et al(41) in their study, which demonstrated that the energy cost values for walking with the reciprocating gait orthosis (RGO) approximated those for wheelchair use.

Limited research has addressed the effects of orthotic designs on ambulation in children with MMC. In most practices, clinical experiences of the prescriber serve as guideposts in orthotic selection. A recent quantitative study by Thomas et al(42) revealed that children with MMC who wore AFOs showed reduced hip, knee, and ankle flexion; decreased muscle activation time; and decreased muscle coactivation during gait. Krebs and colleagues(43) found specific differences for children with MMC using metal versus polypropylene knee-ankle-foot orthoses (KAFOS); children preferred the polypropylene KAFOS and appeared to have better control of hip, knee, and ankle motion.

Studies in the 1970s and early 1980s at the Orthotic Research and Locomotor Assessment Unit (ORLAU) in Oswestry, England, measured velocity, heart rate, and ambulatory, status with different orthoses used primarily for children with high-level spinal defects.(44-46) Descriptive results suggested improvements in all variables when children switched from previous bracing (ie, conveiltional hipknee-ankle-foot orthosis [HKAFO] or swivel walker) to a hip guidance orthosis. In 1986, Lough and Nielsen(47) reported that children with MMC walked faster, but had higher energy expenditure and were less efficient, with their use of parapodiums than with their use of parapodiums adapted with a swivel-walker base.

Comparing the RGO used with different hip joint and cable configurations (normal reciprocal action, locked, or free motion), Yngve and colleagues(48) and McCall and Schmidt(49) favored the normal reciprocal mode based on patients achieving a faster walking velocity. Patients in both studies also reported preferring the reciprocal to the swing-through pattern of gait.

Principles of Lower-extremity Orthoses

Several texthooks are available that provide a more comprehensive coverage of orthoses; each of these textbooks presents a list of principles, purposes, and goals of bracing.(50-53) Following are guiding principles that should be applied with the pediatric client.

1. Prevent deformity. Orthoses may be prescribed to prevent a deformity that is anticipated to occur or that might increase without treatment. Orthoses can also be used following surgery or serial casting to maintain correction of a deformity. Occasionally, orthoses are used to correct a deformity.

2. Support normal joint alignment a mechanics. Molded orthoses often have a greater chance of achieving this goal than do conventional metal devices because they provide total contact. Options in molded AFOs solid, variable range, or free motion) can all support joint alignment. Even foot orthoses (FOs) can facilitate this goal as long as alignment includes the ankle joint, the subtalar joint, and other midfoot and forefoot structures. This principle is inherent in what some clinicians refer to as "tone reduction." Good alignment may be more important than the addition of extra pressure at selected sites in reducing tone.

3. Provide variable range of motion when appropriate. For ambulatory patients, restriction of some ROM is often necessary, to assist weak muscles, oppose spastic muscles, enhance balance, protect certain soft tissues postsurgically, or improve the appearance of gait for the patient who lacks selective control to achieve a normal pattern. At the ankle, this may require restricting plantar flexion while allowing free dorsiflexion or allowing plantar flexion while restricting dorsiflexion. The plantat-flexion stop, for example, might be used for a child with CP who has an equinus deformity, whereas the dorsiflexion stop might be used to eliminate a crouched gait by stopping anterior motion of the tibia in the presence of an overiengthened heelcord. The dorsiflexion stop could also be used for the child with an MMC at L-5 who heel walks or has excessive ankle dorsiflexion. Various combinations of motion and restriction are possible.

4. Facilitate function. The orthotic device should not restrict the child, but should encourage function. The physical therapist is particularly important for assessing the fit and function of an orthosis. Changes in the orthosis may be needed as the patient's ability or muscle function changes, thus ensuring the device facilitates and does not restrain the patient's progress.

Orthotic Design

Foot Orthoses

With FOs (Fig. 1), attention must be given to achieving a close-contact fit that does not impinge on the medial or lateral malleoli. The distal trimline should extend just proximal to the metatarsal heads. The high medial and lateral walls should extend to the dorsum of the foot. Studies on FOs are reported in Table

Supramalleolar Orthoses

Supramalleolar orthoses (SMOs) (Fig. 2) have proximal trimlines that extend 30 to 50 mm above the malleoli. Although the malleoli are covered by the orthotic walls, the orthoses are trimmed anteriorly and posteriorly to allow plantar flexion and dorsiflexion.

Ankle-foot Orthoses

Ankle-foot orthoses (Fig. 3) should be designed to extend 10 to 15 mm distal to the head of the fibula on the pediatric patient. All below-knee supports, whether AFOs fixed or hinged AFOs), SMOs, or FOs, should incorporate supportive features to align the foot in a subtalar-neutral position. An anterior floor-reaction AFO is a variation on the design of an AFO used for children with excessive flexible knee flexion posture. This design has particularly been recommended for children with L-4 myelomeningocele and for children with CP who stand with knee flexion attributable to overlengthened heelcords. The polypropylene mold includes an anterior rather than a posterior shell only. An illustration of the floor-reaction AFO is shown in the article by Rose and colleagues in this issue. If the FO is cut low or supports only beneath the foot, it will not contain all key components of ankle and foot support.

Five Key Components of Ankle and Foot Support

Control of the subtalar joint. The AFO, SMO, and some FOs should grasp the calcaneus firmly (Fig. 1) to prevent medial and lateral motion with subsequent collapse of the heel into valgus and the subtalar joint into pronation. A medial heel post with slight wedging of the plantar surface will also oppose pronation and encourage normal alignment (Fig. 1).

Control of the midtarsal joint. The AFO, SMO, and FO must resist forefoot abduction or adduction. This can be accomplished by ensuring well-molded medial and lateral walls and borders of the orthosis (Fig. 2). At times, the medial or lateral trimlines may extend distal to the metatarsal heads (eg, in a child with MMC following treatment of clubfoot deformity). In this case, consideration should be given to the effect this extension will have on restricting toe extension late in the stance phase.

High (flexible) medial and lateral walls. if the walls of the AFO, SMO, or FO wrap over the dorsum of the foot, the force exerted by the walls will help restrain the heel for maximum control (Fig. 2). Additional benefits of the high walls are (1) control of forefoot abduction or adduction, (2) distribution of pressures over the foot for reducing unwanted stimulation, and, (3) assurance of constant contact with the orthosis during the entire gait cycle. A potential disadvantage is the lack of ease with which the child can don the orthosis. Some parents have reported their children have difficult,,, keeping the orthosis walls spread while sliding the foot into the device.

Toe elevation. Toe elevations may be intrinsic (ie, built into the plastic) (Fig. 2) or extrinsic (ie, removable and repositionable) (Fig. 3). Intrinsic toe elevations may shorten the life of the orthosis because the child's growth cannot be readily accommodated. By contrast, extrinsic elevations can be repositioned as the child grows. Additionally, the height of the elevations can be more easily increased or decreased.

Transmetatarsal arch. As with toe elevations, the transmetatarsal arch can be intrinsic or extrinsic. Caution must be taken to ensure appropriate placement of this feature so the child does not feel inappropriately positioned pressure. This feature has particularly been used with toe elevation for tone reduction.

Variable-Motion Ankle Joints

Until the past 5 years, the solid, molded, fixed orthosis was standard. Recently, however, hinged ankle joints have been introduced, in part, to encourage a more normal gait pattern. Hinged or variable-motion ankle joints can be used in KAFOS and AFOs. The following variable-motion ankle joints represent commonly used designs at the present time.

Overlap articulation. This joint is constructed from polypropylene or copolymer (Fig. 4). The overlap articulation design is the most widely used hinged ankle joint. When used alone, the joint allows free dorsiflexion or plantar flexion. When used in combination with other features of the brace, several possibilities for controlling motion emerge. Typically, the orthosis will be molded with a plantar-flexion stop at neutral. Modifications to the polypropylene, however, can place the stop in dorsiflexion or in plantar flexion. A plantarflexion stop set in dorsiflexion is commonly used to restrict knee hyperextension.

If the goal is to stop dorsiflexion, a posterior strap can be added. A further adaptation involves the addition of screws to restrict motion. A screw placed horizontally through the posterior overlap can stop all motion until the therapist is ready to allow dorsiflexion. A posteriorly

placed vertical screw will allow variable ROM. Turning the screw adjusts the plantar flexion.

Gillette joint. The Gillette joint* is made of rubber vacuformed into polypropylene (Fig. 5). This hinge is normally used to provide a plantar-flexion stop and free dorsiflexion. By grinding the plastic posteriorly, some increased plantar-flexion ROM is possible.

Gaffney joint. This joint' is made of stainless steel vacuformed into the polypropylene (Fig. 6). Seven sizes of hinges are available. This design of hinged ankle joint is normally used to provide a plantar-flexion stop and unrestricted dorsiflexion.

Select joint. These joints* are made of color-coded aluminum disks vacuformed into plastic (Fig. 7). Two sizes of disks are available. Within each size, five variations in disks allow the therapist to insert different disks and change selectively the ROM allowed at the ankle. Because each color-coded disk has its ROM stamped on it, the changes are easy to document. One drawback is the joint's lack of durability. Both the inserts and the plastic forms housing them are subject to wear. The anterior stops seem particularly susceptible to the forces imposed on them during active, forceful walking and may yield to increased ROM. A second drawback is that the metal hinges must be mechanically aligned and thus are not anatomically aligned. Generally, anatomic alignment is used for joint placement, because cosmesis and function are assumed to be better when the brace's joint overlies the true (ie, human or anatomic) joint. A third disadvantage is the bulk of the brace. The bulk of the brace makes it difficult to place a shoe over the hinge and reduces the cosmesis of the device. Some difficulty may also be encountered by the child clicking his or her ankles together.

Oklahoma ankle joint. This joint' is made from polypropylene vacuformed in plastic (Fig. 8). Five sizes, pediatric to adult, are scheduled to be available commercially in 1991. This hinge joint allows all the flexibility of the plastic overlap joint.

Knee-ankle-foot Orthoses

The proximal medial aspect of the thigh portion of a KAFO (Fig. 9) should extend to within 45 mm of the perineum on the medial side and to within 75 mm of the greater trochanter on the lateral side. Various knee joints can be used, and the reader is referred to other references for coverage of the options available.(50-53) knee joint must be aligned in all three planes--sagittal, coronal, and transverse. In the sagittal plane, the joint should be one half the distance between the posterior aspect of the patella and the posterior aspect of the leg. From the coronal plane, the joint should be at the midpoint of the patella. In the transverse plane, the joints must be aligned with the line of gait progression.

Hip-knee-Ankle-foot Orthosis

Conventional metal HKAFOs are rarely used today by children with CP or children with MMC. if these devices are prescribed, however, their fit must be examined in the sitting and standing positions. If the patient is being fitted bilaterally, the critical fitting is done in a sitting position. ideally, the hip joint should be positioned 10 mm superior to the greater trochanter.

Special designs of HKAFOs (parapodium, swivel walker, and RGO) are described in the next section. Only the Rochester HKAFO, which is not used for walking, is mentioned here. This orthosis is mainly used by the very young child with MMC as a nighttime sleeping brace. The orthosis can also be helpful to encourage early weight bearing in the child under 15 months of age who has a high lumbar- or thoracie-level MMC. Parents are instructed to support and supervise their child in standing with this device. A belt across the chest, and thigh and calf straps, help restrain body position for standing.

Parapodium

The Toronto parapodium (Fig. 10) has one lock for both hip and knee joints, whereas the Rochester parapodium has separate hip and knee joint locks.(59,60) With the child standing, the parapodium should fit such that the center of the chest pad is at the xiphoid, the knee pad is across the patellae, and the metal portion of the trunk support clears the axillae by one to two fingers. If the metal expends too proximally, the brace may restrict arm use. If too low, the child may outgrow the appliance quickly. From the side, alignment of the trunk, hips, knees, and ankle joints should be checked. Cutouts on the knee pad should neither be so minimal they cause knee hyperextension nor so extreme they allow excessive knee fiexion. The foot mold should align the malleoli slightly posterior to the knee. The final step of a parapodium fit check involves determining whether the child's pelvis rotates excessively to one side of the pelvic support when the child moves. This check can be performed after the child has been rolling, pushing to a standing position, or walking in the brace. If excessive rotation occurs, increasing the anterior-posterior depth and the tightness of the pelvic support across the greater trochanters may help prevent the shifting.

Swivel Walker

The first swivel walker (Fig. 11) was designed at ORLAU in the 1970s as a modification of an earlier prosthetic device used for children affected by the thalidomide crisis in the 1960s.(61) The fit considerations of the swivel walker parallel those of the parapodium, except for the unique point of spacing for the swivel-walker foot-plates. Based on a case report, recommendation has been made for the spacing to be one-fifth of body height.(62) The crutchless mode of walking is characteristic of this brace. A modification merging the parapodium with the base and footplate assembly of the swivel walker was used and 1960s to an NDT philosophy and a departure from the previous orthopedic approach described by Deaver and Phelps in which extensive bracing was common.(7,8) Because new orthotic materials have been introduced, reconsideration of the present philosophy may be warranted.

The use of AFOs has remained a standard treatment because of their usefulness in preventing equinus contractures. Some clinicians have recommended a progression from casts before the child walks to AFOS once walking begins. Although debate continues over the relative advantages of hinged versus fixed AFOs, studies by Middleton et al(21) and Lough(28) have favored hinged AFOs. Because specific guidelines for selecting the style of hinged ankle joint do not exist, the choice should be made jointly by members of the treatment team consisting of physician, physical therapist, and orthotist.

Issues related to orthotic prescription for children with hypotonia or athetoid (dyskinetic) CP have not been well addressed in the literature. Many clinicians believe children with hypotonia are more stable with molded FOs, SMOs, or AFOs. By contrast, children with dyskinesia frequently have difficulty with molded orthoses because of friction irritation created by writhing movement. Partial or full lining of the orthosis may reduce the irritation.

The selection of an AFO for a child with CP will change depending on age, nature of the disability, associated findings, and surgical history. The following recommendations are based on reports in the literature, as well as personal clinical and research experience.

1. The preambulatory child under 30 months of age who appears to be at risk for plantat-flexion contracture should begin with a fixed AFO or an inhibitive cast." Hinged AFOs are more complex and should be postponed until the second fitting. Ankle straps or high medial-lateral walls will help secure the heel.

2. The child who walks in equinus will likely require restriction of plantar flexion to neutral. This can be accomplished with a hinged AFO or a fixed AFO. The hinged AFO will facilitate a more normal gait pattern.

3. When genu recurvatum is present, an AFO that blocks plantar flexion at 5 degrees of dorsiflexion should create a knee flexion moment at the onset of stance phase that will counter the hyperextension.

4. If the child with CP responds well to a hinged AFO that restricts plantar flexion, gradual reintroduction of a small range of plantar flexion may be possible. When this is anticipated, the Select joint, which allows several joint settings, may be the best choice.

5. The child who is heavy or large, is extremely active, or does not reside near an orthotist may not be a good candidate for the Select joint because of the potential for earlier wear with this type of hinge.

6. For a child with very mild hypertonia and medial-lateral instability, but not equinus, a free-motion hinged AFO or SMO may suffice. Caution is needed to identify, increasing plantar flexor tightness, which would indicate the need for an AFO that restricts plantar flexion.

7. A child with weak plantar flexors because of surgical lengthening or selective posterior rhizotomy should have an AFO that restricts dorsiflexion.

Myelomeningocele

The type of orthosis prescribed for a child with MMC is influenced by the child's level of motor function. The physical therapist should assume a key role in clarifying muscle function before an orthosis is prescribed. if a child demonstrates muscle activity in the lower extremities that is not willful or strong, or is reflective, a judgment about "functional motor level" should be made. We believe that the functional motor level reflects the child's ability to use muscles for skilled activities rather than the muscles innervated. When a discrepancy exists between the functional motor level and the motor level represented by the muscle activity present, the functional motor level should guide selection of the orthosis.

An early consideration in selecting an orthosis for a child with MMC is the timing for the first device. Whereas Drennan(64) recommended bracing at the neurodevelopmental age when children normally learn to stand and walk, Carroll(65) advocated beginning standing when the child begins to demonstrate interest in being upright. Based on Drennan's recommendation, an orthosis would be prescribed when a child is 9 to 12 months old. Carroll's recommendation appears preferable, because it encourages adjustment for individual differences in social or cognitive developmental readiness. If the developmental readiness guideline is used, however, an exception becomes necessary for orthopedic benefits when the child reaches 20 to 24 months of age but remains below the 9-month level developmentally. In such cases, an orthosis should he ordered and the parapodium may be the best choice for a first device, even if the child has a low lumbar MMC.

A further consideration related to orthotic selection is that the types of orthoses used by children with MMC typically change with age. Changes are particularly common during the first 6 years of life, and families should be assisted in planning for these changes. Before changing from the current orthosis to a more sophisticated design, a child should he able to walk with the current device. Table 3 presents guidelines for responding to factors that may impede walking progress in young children learning to use their first orthosis. Minor changes may be indicated within 6 months of initiating orthotic use. Such changes could include addition or removal of a swivel base to the bottom of a parapodium or addition of an above-knee component of AFOs.

Based on clinical reports, personal experience, and research, the following guidelines for initial to later orthotic devices are provided.

1. A child with an S-2 motor level may begin standing and walking without an orthosis but may later benefit from FOs.

2. A child with an L4-5 motor level may ambulate throughout life with AFOs or SMOs. Rarely would a parapodium he needed as a first device.

3. A child with an L3-4 motor level will benefit from the postural support and stability, provided by a parapodium used as a first standing and walking orthosis. Although some reports(48,49) have cited the use of an RGO at this motor level, the child should eventually ambulate with AFOs, AFOs attached to twister cables or a single lateral upright and pelvic hand, or standard KAFOs.

4. A child with a motor level above L-2 will likely require an HKAFO as a first and later device. The child may begin with a parapodium on a swivel walker, then switch to an RGO. Transition from a parapodium to an RGO is not recommended before 30 to 36 months' developmental age. The likelihood of a child with high spinal-level MMC using only a wheelchair after age 15 years is great; however, factors the orthosis used, the philosophy of the program, and personal characteristics of the child and family can influence outcome.

5. A child with discrepant motor and "functional motor" levels, should be fitted with an orthosis according to the functional motor level. For example, a child having some muscle activity through the L-5 level, who belly crawls rather than creeps, does not pull to stand or support body weight in vertical, and uses mainly hip flexors, should be considered to have a functional motor level of L-2. A parapodium or standing frame would be more appropriate than AFOS for a first device. Changes can be made as indicated with age.

Summary

This article has provided a summary of research on orthoses used for children with CP and children with MMC. Principles of orthotic application and guidelines for orthotic selection have been presented. Each child needs to be assessed individually, with consideration given to functional ability, management goals, and orthotic options. Although progress has been made in the development of orthoses and in the understanding of how the orthoses affect gait variables, further research will be needed as advances in material properties and designs are made available. Additionally, physical therapists should give further consideration to such matters as critical timing for introducing or progressing gait and methods for gait and other functional skill training with orthoses. The issue of using a total-mobility, approach, which represents a combined program of seated mobility and ambulation versus either mobility mode used exclusively, also warrants attention from clinical researchers.

References

1 Parieth N, Kiely. N. The frequency, of cerebral palsy; a review of population studies in industrialized nations since 1950. In Stanley F, Alberman E, eds. The Epidemiology of the Cerebral Palsies. Philadelphia, Pa: JB Lippincott Co; 1984;46-56.

2 Evans P, Elliott M, Alberman E, Evans S, Prevalence and disabilities in 4 to 8 year olds with cerebral palsy. Arch Dis Child. 1985; 60:940-945.

3 Kudrjavcev T, Schoenberg BS, Kurland KT, Groover RV. Cerebral palsy: survival rates, associated handicaps, and distribution by clinical subtype Rochester, MN, 1950-1976) Neurology. 1985;35:900-903

4 Shurtleff DB, Lemir RJ, Warkany J. Embryology, etiology, and epidemiology. In: Shurtleff DB, ed. Myeldysplasias and Extrophies: Significance, Prevention, and Treatment. New York, NY: Grune & Stratton Inc; 1986:39-64.

5 Wegman ME. Special article: annual summary vital statistics-1989. Pediatrics. 1990; 86:835-847.

6 Walker DK, Palfreys, Butler JA, Singer J. Use and sources of payment for health and community services for children with impaired mobility. Public Health Rep. 1988;103:411-415.

7 Wolf JM, ed. The Results of Treatment in Cerebral Palsy. Springfield, Ill: Charles C Thomas, Publisher; 1969:18-26.

8 Bleck EE. Orthopaedic Management in Cerebral Palsy. London, England: MacKeith Press; 1987:142-183

9 Cusick BD. Progressive Casting and Splinting for Lower Extremity Deformities in Children with Neuromotor Dysfunction. Tucson, Ariz: Therapy Skill Builders; 1990.

10 Mossberg KA, Linton KA, Friske K. Ankle-foot orthoses: effect on energy expendiof gait in spastic diplegic children, Arch Phys Med Rehabil. 1990;71:490-494.

11 Brodke DS, Skinner SR, Lamoreux LW, et al. Effects of ankle-foot orthoses on the gait of children. J. Pediatr Orthop. 1989;9:702-708.

12 Yates G. Molded plastics in bracing. Clin Orthop. 1974;102:46-57

13 Hoffer MM, Garrett A, Koffman M, et al. New concepts in orthotics for cerebral palsy. Clin Orthop. 1974;102:100-107.

14 Rosenthal RK. The use of orthotics in foot and ankle problems in cerebral palsy. Foot Ankle,. 1984;4:195-200.

15 Banks HH. The management of spastic deformities of the foot and ankle. Clin Orthop. 1977;122:70-76.

16 Bennet GC, Rang M, Jones D. Varus and valgus deformities of the foot ill cerebral palsy. Dev Med Child Neurol. 1982;24:499-503.

17 Cusick BD, Sussman MD. Short leg casts: their role in the management of cerebral palsy. Physical and Occupational Therapy in Pediatrics. Binghamton, NY: The Haworth Press Inc; 1982;2:93-110.

18 Duncan WR, Mott DH. Foot reflexes and the use of the "inhibitive cast. " Foot Ankle. 1983;4:145-148.

19 Jordan RP. Therapeutic considerations of the feet and lower extremities in the cerebral palsied child. Clin Podiatr Med Surg. 1984;1:547-561.

20 Selby L. Remediation of toe-walking behavior with neutral-position, serial-inhibitory casts: a case report. Phys Ther. 1988;68: 1921-1923.

21 Middleton EA, Hurley BRB, Mcllwain JS. The role of rigid and hinged polypropylene ankle-foot-orthoses in the management of cerebral palsy: a case study. Prosthet Orthot Int. 1988;12:129-135.

22 Hinderer KA, Harris SR, Purdy AH, et al. Effects of "tone-reducing" vs standard plaster-casts on gait improvement of children with cerebral palsy. Dev Med Child Neurol. 1988;30:370-377.

23 Harris SR, Riffle K. Effects of inhibitive ankle-foot orthoses on standing balance in a child with cerebral palsy: a single-subject design. Phys Ther. 1986;66:663-667.

24 Embrey DG, Yates L, Mott DH. Effects of neuro-developmental treatment and orthoses on knee flexion during gait: a single-subject design. Phys Ther. 1990;70:626-636.

25 Bertoti DB. Effect of short leg casting on ambulation in children with cerebral palsy. Phys Ther. 1986;66:1522-1529,

26 Watt J, Sims D, Harckham F, et al. A prospective study of inhibitive casting as an adjunct to physiotherapy for cerebral-palsied children. Dev, Med Child Neurol 1986;28: 480-488.

27 Mills VM. Electromyographic results of inhibitory splinting. Phys Ther. 1984;64:190-193.

28 Lough LK (Knutson LM). The Effects of Fixed and Hinged Ankle-foot Orthoses on Gait Myoelectric Activity and Standing joint Alignment in Children with Cerebral Palsy. Iowa City, Iowa: The University of Iowa; 1990. Doctoral dissertation.

29 Clinkingbeard JR, Gersten Mw, Hoehn D. Energy cost of ambulation in the traumatic paraplegic. Am j Phys Med. 1964;43:157-165.

30 Richings JC, Eckstein HB, Locomotor and educational achievements of children with myelomeningocele. Ann Phys Med. 1970;10: 291-298.

31 Lorber J. Results of treatment of myelomeningocele: an analysis of 524 unselected cases, with special reference to possible selection for treatment. Dev Med Child Neurol 1971;13:279-303.

32 Hoffer MM, Feiwell E, Perry R, et al. Functional ambulation in patients with myelomeningocele. J Bone Joint Surg [Am]. 1973;55: 137-148.

33 Desouza LJ, Carroll N Ambulation of the braced myelomeningocele patient. j Bone Joint Surg [Am] 1976;58:1112-1118.

34 Barden GA, Meyer LC, Stelling FH. Myelodysplastics: fate of those followed for twenty years or more. J Bone joint Surg [Am]. 1975; 57:643-647.

35 Schopler SA, Menelaus MB. Significance of the strength of the quadriceps muscles in children with myelomeningocele. J Pediatr Orthop. 1987;7:507-512.

36 Findley TW, Agre JC, Habeck RV, et al. Ambulation in the adolescent with myelomeningocele, L early childhood predictors. Arch Phys Med Rehabil. i987;68:518-522.

37 Mazur JM, Shurtleff DB, Menelaus MB, Colliver J. Orthopaedic management of high-level spina bifida: early walking compared with early use of a wheelchair. J Bone Joint Surg A-1. 1989;71:56-61.

38 Taylor A, McNamara A. Ambulation status of adults with myelomeningocele. Z Kinderlchir. 990;45:32-33.

39 Samuelsson L, Skoog M. Ambulation in patients with myelomeningocele: a multivariate statistical analysis. J Pediatr Orthop. 1988; 8:569-575.

40 Sankarankutly M, Rose GK, Stallard J. The effect of orthotic treatment on spina bifida patients. Spina Bifida Therapy. 1979;1:187-196.

41 Flandry F, Burke S, Roberts JM, et al. Functional ambulation in myelodysplasia: the effect of orthotic selection on physical and physiologic performance, J Pediatr Orthop. 1986; 6:661-665.

42 Thomas SE, Mazur JM, Child ME, Supan TJ. Quantitative evaluation of AFO use with myelomeningocele children. Z Kinderchir. 1989;44:38-40,

43 Krebs DE, Edelstein JE, Fishman S. Comparison of plastic/metal and leather/metal knee-ankle-foot orthoses. Am J Phys Med. 1988;67:175-185

44 Davies JB. Use of heart rate in assessment of orthoses. Physiotherapy. 1977;63:112-115, 45 Stallard J, Rose GK, Farmer IR. The ORLAU swivel walker. Prosthet Orthot Int. 1978;2: 35-42.

46 Rose GK, Stallard J, Sankarankutty M, Clinical evaluation of spina bifida patients using hip guidance orthosis. Dev Med Child Neurol 1981;23:30-40.

47 Lough LK (Knutson LM), Nielsen DH. Ambulation in myelomeningocele: parapodium versus parapidium with ORLAU swivel modification. Det, Med Child Neurol 1986;28: 489-497.

48 Yngve DA, Douglas R, Roberts JM. The reciprocating gait orthosis in myelomeningocele. J Pediatr Orthop 1984;4:304-310.

49 McCall RE, Schmidt WT. Clinical experience with the reciprocal gait orthosis in myelodysplasia. J Pediatr Orthop. 1986;6:157-161.

50 Bunch WH, Keagy RD. Principles of Orthotic Treatment. St Louis, Mo: CV Mosby Co; 1976.

51 Rose GK. Orthotics: Principles and Practice. London, England: William Heinemann Medical Books Ltd; 1986.

52 Redford JB, ed. Orthotics Etcetera. 3rd ed. Baltimore, Md: Williams & Wilkins; 1986.

53 Shurr D, Cook T. Prosthetics and Orthotics. East Norwalk, Conn: Appleton & Lange; 1990.

54 Wenger DR, Mauldin D, Speck G, et al. Corrective shoes and inserts as treatment for flexible flatfoot in infants and children. J Bone joint Surg [Am]. 1989;71:800-810.

55 Penneau K, Lutter LD, Winter RD. Pes planus: radiographic changes with foot orthoses and shoes. Foot Ankle. 1982;2:299-303.

56 Bordelon RL. Hypermobile flatfoot in children: comprehension, evaluation, and treatment. Clin Orthop. 1983;181:7-14.

57 Bleck EE, Berzins UJ. Conservative management of pes valgus with plantar flexed talus, flexible. Clin Orthop. 1977;122:85-94.

58 Mereday C, Dolan CME, Lusskin R. Evaluation of the University of California biomechanics laboratory shoe insert in flexible" pes planus. Clin Orthop, 1972;82:45-58

59 Motloch WM. The parapodium: an orthotic device for neuromuscular disorders. Artificial Limb,,;. 1971;15:36-47,

60 Kinnen E, Gram M, Jackman KV, et al. Rochester parapodium. Clin Prosthet Orthot. 1984;8(4):24-25.

61 Rose GK, Henshaw JT. Swivel walkers for paraplegics: considerations and problems in their design and application. Bull Prosthet Res. 1973;10-20:62-74.

62 Stallard J, Rose GK, Tait JH, Davies JB. Assessment of orthoses by means of speed and heart rate. J Med Eng Technol. 1978;2:22-24.

63 Solomonow M, Baratta R, Hirokawa S, et al. The RGO Generation 11: muscle stimulation powered orthosis as a practical walking system for thoracic paraplegics. Orthopedics. 1989; 12:1309-1315.

64 Drennan JC. Orthotic management of the myelomeningocele spine. Det, Med Child Neurot 1976;18(suppl 37):97-103.

65 Carroll N. The orthotic management of the spina bifida child. Clin Orthop. 1974;102: 108-114.