Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Coordinated Treatment in Cerebral Palsy--Where Are We Today?

James P. Rogers, C.P.O.
Sharon H. Vanderbilt, P.T.

Introduction

The world of lower extremity orthotics evolves almost continuously due to advances in the technology of materials, biomechanical analysis, and our understanding of orthopedic and neurologic conditions. The area of upper-motor neuron disorders, specifically cerebral palsy, has been the subject of considerable research regarding various treatment modalities in physical therapy, orthotic approaches to control of the lower extremities, and corrective surgical procedures. Years after the work and notable contributions of Bleck, Bobath and Bobath, Duncan and Mott, Cusick, Sussman, and countless other dedicated professionals, those who care for patients with cerebral palsy find themselves atop a plateau of useful information. But has this vast pool of knowledge been used to provide a coordinated and focused approach to the management of cerebral palsy?

It is the authors' hope to chronicle high points in the development of modern approaches to management of cerebral palsy, to discuss the theories and modalities of experienced professionals, and to emphasize the need to explore and understand in depth concepts that traditionally have not been familiar to orthotists. These include but are not limited to: developmental biomechanics, neurodevelopmental technique, tone, reflexes, serial casting, splinting, and a variety of orthoses. To successfully treat these patients, it is imperative to become familiar with these terms and maintain a close collaborative relationship with a knowledgeable therapist, and ideally, one or more progressive physicians, orthopedic surgeons or neurosurgeons.

As early as the 1960s, Bobath and Bobath discussed the neurodevelopmental approach (NDT) to cerebral palsy management, along with detailed assessments of the development of gait patterns in patients with cerebral palsy.^1,2,3,9 By the late 1970s, Cusick and Sussman had acquired considerable experience with the use of plaster short leg casts and were defining their role as an adjunct to therapeutic treatment of cerebral palsy pa- tients.^4,5,6 An analysis of these early publications captures the essence of the modern approach to cerebral palsy management: no longer do we focus only on one segment of the affected individual (i.e., orthotic control of tone-induced plantarflexion). We must now view the entire individual and his postural integrity in the context of development or lack thereof.

Following a referral from a physician, therapists assess these patients on an ongoing basis in terms of their developmental and functional capacities. "Orthopedic neurological examination" as described by Bleck⁷ begins with postural assessment. Ideally this assessment occurs before the patient initiates ambulation and, therefore, orthotists are not generally involved at this stage of treatment. This fact, however, does not diminish the significance of the findings, and we recommend that orthotists who wish to treat cerebral palsy patients successfully become familiar with assessment techniques. (An excellent explanation is offered by Bleck in the suggested readings.)

Since cerebral palsy is somewhat inclusive as a description, we should explore a few particulars. A patient with cerebral palsy may be described as "hemiplegic," having unilateral flexion of the upper extremity (elbow, wrist and fingers) and marked equinus of the foot and ankle. A "spastic-diplegic" displays bilateral involvement usually concentrated grossly in the lower extremities, manifesting hip flexion, adduction, and internal rotation along with flexed knees and equinus at the foot and ankle. (Fine motor deficiencies are common in the upper limb.) A "quadriplegic" presents with whole body involvement, either spastic and/or athetoid, often dominated by extensor patterns (Figure 1) .

Orthotists sometimes use "spastic" to describe motor disorders in a general sense. It should be understood, however, that spasticity is an increase in the stretch reflex muscles, i.e., hypertonicity. Bleck describes tone as "the state of partial contraction in which muscles maintain their posture without fully relaxing . . . "⁷ The spastic response, also called myostatic reflex, often can be inhibited in all but the most severe cases if one is careful not to encourage the reflex. The "athetoid" patient, however, may display muscle tension similar to spasticity, but his muscles will elongate responding to repetitive stretches in what is called "tension type athetosis." The 'rigid" athetoid will display firm resistance to passive joint motion in a continuous or discontinuous pattern and is often labeled 'severely spastic," when in reality normal joint motion and neutral positioning are not obtainable. These are just a few examples of the subtle differences not often identified by orthotists during a relatively short duration of patient contact; but successful orthotic care depends upon the depth of understanding possessed with respect to the subtleties of the disorder.

Keeping in mind a "total body approach," the next step would be to understand developmental features as they relate to normal children and to assess patients comparatively. Fortunately for the orthotist, this is not part of his role. Let us explore for a moment the interdisciplinary nature and complexity involved. The orthopedic neurologic assessment of a child includes several tests for a series of reflexes that are known to be abnormal or absent after certain ages.⁷ The asymmetrical tonic neck reflex, parachute reaction, foot placement reaction and extensor thrust test, all taken together, give the physician a prognosis for walking potential. (Notice that only two of these reflexes deal directly with the lower extremities.) Motor assessment of passive range of motion, muscle tests for relative motor development, primarily the responsibility of the therapist, are used to assess overall function. But since normal children do not all develop on the same exact timetable, functional assessments are used for a more accurate picture of where the child is developmentally. Developmental milestones (Figure 2) are essential tools for assessment. A clearer picture of the reasons why an 18 month old child cannot walk is obtained when one documents the absence of sitting posture at 15 months. Similarly, a 10 month old who crawls reciprocally may not benefit from AFOs which block foot/ankle motion and reduce tactile sensation.

What about biomechanical development? Beverly D. Cusick in her discussion of Harold Frost, a notable contributor to the field of developmental biomechanics, notes an interesting fact about this field which closely parallels the fields of orthotics and prosthetics. The study of developmental biomechanics, which began in the 1940s, is a relatively new field. Similar to orthotics and prosthetics, new developments are dispelling long held beliefs and producing changes in technique. Just as orthotics and prosthetics looks anxiously to the future, developmental biomechanics, barely 50 years old, hold promise in areas including the structural design of the skeleton, the principles of the growing skeleton, and the properties of materials both inside and outside the body (i.e., bone, plates, implants, orthoses and casts).

For orthotists the essence of developmental biomechanics should concern the changes present normally as a child ages, in the foot and ankle; and primarily the intimate relationship between the hip, knee and ankle-foot complex. Cusick charts the develop-mental changes (Figure 3) that occur from birth through childhood to adulthood. ¹⁸ The challenge is to coordinate orthotic treatment using knowledge of where the child is developmentally to provide MAXIMUM FUNCTION while still allowing and enhancing chances for further development. It is clear that this management is not static.

It is also clear that orthotists treating patients with cerebral palsy must possess a working knowledge of the lower extremity and its biomechanics. Fvaluation of hip range of motion is helpful in determining the forces acting on the knee in the frontal plane (Figure 4) . Hip abduction/adduction ROM is especially important with spastic diplegics; a lack of abduction may dictate medial collapse of the hindfoot and midfoot creating a classic orthotic management problem (Figure 5) . Hamstring assessment is important due to the negative effect the flexed knee has on stride length. ¹⁶ Accurate and documented ROM of the gastroenemius and soleus will be invaluable in determining the long-term effectiveness of orthotic intervention. The lack of dorsiflexion not only prevents foot flat, but influences the entire gait cycle negatively and certainly will contribute to tone related problems during weightbearing.^11,12,13,16Figure 6 emphasizes the difference in testing for these muscles.

Understanding the influence of proximal structures on the foot and ankle complex is invaluable in attempts to analyze the distal lower extremity and support and facilitate its function. One of the easiest ways to evaluate calcaneal stance position is illustrated by McCrea (Figure 7) . The position of the calcaneus and its active ROM determine the positioning of the talus and subsequent midfoot and forefoot positioning. 15

The position of the talus, arch integrity and forefoot pronation or supination must be noted. The ability to provide orthoses which contribute to whole body function in addition to foot and ankle stability requires a clear picture as to the "positioning potential" of the foot.

Understanding the orthopedic and neurologic assessment, the functional and biomechanical developmental status of the patient and possessing a clear biomechanical "picture" of his lower extremity only bring us to the juncture of intervention. It can be argued that this stage is where professionals differ most on what direction to take. Clearly. Though, there are an increasing array of treatments available that represent a comprehensive and interdisciplinary cross-section

"Casting" the lower extremity as an adjunct to physical therapy has been described by several noted practitioners.^56,20,21,22 This intervention certainly has proved to increase the ability of therapists to successfully treat the trunk and upper extremities through a reduction in tone brought about by "inhibiting" extensor reflexes.

Long-term use of inhibitive casts has been shown to facilitate the development of motor skills such as standing, cruising and wilking.^19,23 (In fact, the authors have observed children in inhibitive casts weighing over 3 lbs. walking with a gait improved over that obtained in conventional AFOs and tennis shoes.) Many therapists now hi-valve these casts and use them at various times during therapeutic activities. Bi-valved inhibitive casts can also be used as night positioning splints with positive results.

The drawback to inhibitive casts is a lack of usefulness other than therapeutic or nocturnal. Because of this, a great many splints and orthoses have been used to provide the benefits of inhibitive casting without the obvious inconveniences. But what criteria should determine the appropriate splint or orthosis'? Again, the potential for controversy looms, but a few parameters do remain fairly clear. Whenever a patient exhibits a fixed deformity of the foot or ankle, splinting and/or orthotic treatment should be reconsidered. (Remember: because children grow and are not static, the goal is maximum function, not just support and stabilization.) Severe contractures of the knee which shorten stride length and prevent normal patterns in the foot during gait also beg for alternative treatment, or what may be termed as "pretreatment." Many professionals feel that advances in splints and orthoses have reduced the role of the plaster cast to just one: contracture reduction. When facing fixed deformities of the lower extremity serial casting is the treatment of choice.

Protocol for serial casting is rather straightforward. In the authors' experience casts applied in series have yielded excellent results in reestablishing mobility in connective tissue. The first east is carefully molded to the "comfortable" end point of the subtalar and mid-tarsal joints ranges of motion as they are brought toward neutral alignment. (Supination of the forefoot may be built-in to facilitate mid-tarsal and hindfoot correction.) After three to seven days the cast is removed and a new cast is applied using the end points of the additional range gained. It is imperative that accurate data on available range be maintained as the casting regimen will continue as long as significant increases in range are noted. Cusiek considers 50 or more as significant.²⁴ Length of the series can be three to six weeks. Forefoot supination used to facilitate correction at the subtalar and mid-tarsal joints is gradually reduced as a neutral attitude is achieved. The casts are extended above the knee once a neutral subtalar is achieved and dorsiflexion of at least 5° (and preferably 100) is reached. Cusick offers a comprehensive criteria as well as de tailed application procedures for this intervention.^18,24,25

Once the regimen is completed and an acceptable range of motion achieved at the affected joints, the necessity to maintain the corrected posture dictates the need for a bivalved east (use the final serial cast) for night use and the appropriate splint or orthosis required to maximize function while maintaining ROM.

The endless array of splints and orthoses underscore the need for accurate biomechanical assessment. Once again Cusick offers a concise and focused look at the roles of splints vs. orthotic devises in the December 1988 issue of Physical Therapy. 28 The authors concur with Cusick and feel that direct, formed, flexible splints are often indicated for children three years of age and younger who have flexible deformities. Frequent developmental changes as well as linear growth make direct molded splints practical from a physiologic as well as an economic standpoint. When deformities offer significant enough resistance, however, direct mold techniques fail to achieve adequate correction. Orthoses custom-made from plaster molds offer superior fit, correction and a wider range of material selection to suit the varying requirements of rigidity. This decision process leaves no room for professionals concerned with "protecting the turf." The focus on the functional goals of the patient dictates that a professional, collaborative atmosphere prevail.

It is helpful to break the orthotic choices down into basic groups and describe which patients would be best served by each. Note that each orthotic device can be duplicated in the direct-formed type of splint design. (Indeed some orthoses trace their beginnings to direct-molded splints.) Age, expected duration of use, economics, frequency of change, and desired rigidity are the parameters guiding the choice between splints and orthoses.

The solid ankle-foot orthosis is probably the most prevalent design used today. When close attention is paid to the contours off the hindfoot this orthosis not only effectively controls equinus but subtalar and mid-tarsal deviations as well. This design is not dynamic, however, and tends to serve only static functions well.

Cerebral palsy patients are constantly subjected to abnormal forces about the foot and ankle which lead to pronounced soft tissue deformity. Cusick appropriately refers to this phenomenon as "ereep"-a deformity related to soft tissue adaptation to the chronic application of abnormal stresses. Solid ankle orthoses work well to prevent "creep." Generally, deformities about the foot and ankle of a functional (correctable) nature respond well to solid ankle designs. Conditions well suited to solid ankle designs include:

1. functional equinus affecting knee alignment;
2. functional frontal plane deviations (subtalar and midtarsal);
3. standing frame use;
4. night splinting; and
5. post operative immobilization.

Articulated ankle-foot designs have become increasingly popular. Although these designs are often called "inhibitive," the inhibitive properties will be discussed later on a collective basis. The single most important criteria for an articulated design is passive range of motion at the ankle to at least 5° of dorsiflexion but again preferably 100 without compromise of neutral subtalar or midtarsal position.²⁶ Attempts to introduce motion to joints incapable of neutral attitudes over the entire range have disastrous implications.

Articulated AFOs can be fabricated with adjustable stops to control genu recurvatum and encourage rollover in stance phase (Figure 8) . They can also be fabricated to prevent excessive dorsiflexion by incorporating check straps, anterior panels or adjustable stops placed anterior to the ankle joints. Biomechanically, the articulated AFOs allow a stretch moment on the calf muscles during stance as the weight-line crosses anterior to the ankle joint in direct proportion to the amount of dorsiflexion built in. This motion is exactly opposite the common extensor response seen in many spastic cerebral palsy patients as the weight line moves anteriorly and pressure increases over the metatarsals. The authors feel this reversal may be a more important design characteristic than many "inhibitive" modifications. The introduction of this design without further modification to articulated AFOs has resulted in reduced spasticity, longer stride lengths, improved balance and increases in range of motion on a large percentage of patients receiving articulated AFOs (Figure 9) . Conditions well suited for treatment with articulated AFOs include:

1. function equinus/genu recurvatum;
2. classic hemiplegia;
3. equino-varus and valgus deformity; and
4. toe drag during swing phase.

A relatively new design - as far as orthotists are concerned - is the supramalleolar orthosis which remotely resembles a UCBL orthosis with trimlines extended proximally. This is where the comparison ends, however, because the variable trimlines of the supramalleolar orthosis give it increased and selective application over the UCBL orthosis. Used primarily in mild crouch deformities the supra-malleolar orthosis is well suited to facilitate control of calcaneus deformity, but the presence of strong or fixed equinus is a contraindication for use.²⁷

By extending the trimlines proximal to the malleolus and providing for "flairs" posteriorly the practitioner can selectively allow plantarflexion while controlling anterior motion of the tibia over the foot with the use of an anterior shell.²⁷ Additional or full plantarflexion can be allowed by trimming the posterior trimline increasingly distal. The predominate force at the foot is pronation and the higher medial trimline provides added support over a greater area to check this tendency. To date, trials with the supra-malleolar's different designs have yielded promising results in controlling mild crouch deformities and facilitating functional control during therapeutic exercise.

No discussion of orthotic control as it relates to cerebral palsy would be complete without mentioning inhibitive designs and their rationale. This topic was purposely presented last because any one of the preceding splints or orthotic devices can be modified to include any or all of these modifications. The goal of inhibitive design is most often a reduction in reflex induced foot deformities and/or the reduction or elimination of extensor thrust. The reality is that not all cerebral palsy patients require inhibitive modifications or orthotic devices to achieve acceptable functional goals. The most commonly used modifications include:

1. A well defined force along the medial aspect of the calcaneus and under the talus. This effectively relieves the navicular and promotes dorsiflexion during weightbearing as the weightline moves distally along the foot.
2. A force exerted on the lateral aspect of the plantar surface (sometimes called a cuboid modification) which promotes peroneal activity as well as a stabilizing effect on the gluteus medius.
3. A metatarsal pad or bar often accompanied by reliefs for the metatarsal heads which serves to inhibit the toe grasp reflex.
4. Extension of the digits along with metatarsal reliefs or padding to inhibit the toe grasp reflex. ^19,21,22,29

It is recommended that the practitioner utilize accurate assessment techniques and test manually these inhibitive forces where possible to determine their possible benefit. No modifications replace sound moulding principles or careful casting and modification-the foundations for all orthotic control. When used appropriately, inhibitive modifications perfect and enhance effective orthotic control.

An overview of the predominant theories on the management of cerebral palsy has been presented and the authors acknowledge the ongoing nature of these efforts. It is strongly recommended that practitioners facing the challenges of managing cerebral palsy, regardless of discipline, enhance their theoretical backgrounds through investigation and reading. Finally, as we strive to coordinate our efforts for the maximal benefit of our patients, our goals for orthotic intervention might be:

1. prevention of deformity;
2. correction of soft tissue deformity;
3. control of undesirable motions of the affected; supporting segments white permitting motion where it occurs normally;
4. protection of weak stabilizing muscles, either following surgery or secondary to hypotonia;
5. control of deviations associated with tonus deformity; and
6. enhancement of experience.²⁸

Suggested Reading List

1. McCrea, John D., Pediatric Orthopedics of the Lower Extremity, An Instructional Hand- book, Futura Publishing Company, Inc., New York, 1985.
2. Mann, R.A., "Biomechanics of the Foot," W.H. Bunch, R. Keagy, eds. American Academy of Orthopedic Surgeons, Atlas of Orthotics, 2nd Edit., C.V. Mosby Co., St. Louis, 1985, pp. 112- 125.
3. Wilson, Janet M., "Cerebral Palsy," Pediatric Neurological Physical Therapy, S. Campbell, Ed., Churchhill Livingstone, New York, 1984, pp.389-407.
4. Root, M.L., W.P. Orien, J.H. Weed, "Normal and Abnormal Function of the Foot," Clinical Biomechanics, 2, Clinical Biomechanics Corp., P.O. Box 35185, Los Angeles, CA 90035.
5. Root, M.L., W.P. Orien, J.H. Weed, R.J. Hughes, "Biomechanical Examination of the Foot," Clinical Biomechanics Corp., P.O. Box 35185, Los Angeles, CA 90035.
6. Carlson, S.J., A Neurophysiological Analysis of Inhibitive Casting," PT & OT in Ped., 4(4), Winter 1984, pp.31-42.
7. Duncan,, W.R., D.H. Mott, "Foot Reflexes and the Use of the Inhibitive Cast,"' Foot and Ankle, 4(2), 1983, pp.145-148.
8. Otis, J.C., L. Root, J.R. Pamilla, M.A. Kroll, "Biomechanical Measurement of Spastic Plantarflexors," Dev. Med. Child Neurol., 25, 1983, pp.804-805.
9. Zachazewski, J.E., E.D. Eberle, M. Jefferies, "Effect of Tone-Inhibiting Casts and Orthoses on Gait," Phys. Ther., 62(4), 1982, pp.453- 455.
10. Bleck, E.E., "Orthopedic Management of Cerebral Palsy," Saunders Monographs in Clinical Orthopedics, 2, W.B. Saunders Co., Philadelphia, 1979.
11. Yates, H.L., D.H. Mott, "Inhibitive Castmg," Proceedings: First William C. Duncan Seminar on Cerebral Palsy, University of Washington, Seattle, 1977.
12. Bunch, W.H., V.M. Dvonch, "Cerebral Palsy," American Academy of Orthopedic Surgeons, Atlas of Orthotics, 2nd Edit., C.V. Mosby Co., St. Louis, 1985, pp.259-269.
13. Kaplan, N., "Effect of Splinting on Reflex Inhibition and Sensorimotor Stimulation in Treatment of Spasticity," Arch. Phys. Med., 43, 1962, pp.565-569.
14. Duncan, W.R., "Tonic Reflexes of the Foot: Their Orthopedic Significance in Normal Children and in Children with Cerebral Palsy." J. Bone & Jt. Surg., 42-A(s), 1960, pp.859-868.

James P. Rogers, C.P.O., is with Southeastern Orthotics and Prosthetics, Inc., P.O. Box 3357, Chattanooga, TN 37404

Sharon H. Vanderbilt, M.A., P.T., is a practicmg physical therapist affiliated with T.C. Thompson Children's Hospital and The Signal Centers. She also assists in NDT instruction, and may be reached at 444 N. Crest, Chattanooga, TN 37404.

References:

1. Bobath, K., "An Analysis of the Development of Standing and Walking Patterns in Patients with Cerebral Palsy," Physiotherapy, 48, June 10, 1962, pp.144-153.
2. Bobath, K., "The Neurodevelopment Treatment of Cerebral Palsy," Physical Therapy, 47, November 1967, pp.1039-1041.
3. Bobath, K., B. Bobath, "The Facilitation of Normal Postural Reactions and Movements in the Treatment of Cerebral Palsy," Physiotherapy, 30, 1964, pp.246-262.
4. Cusick, B., "Developmental Program for Children in Below-Knee Casts," Proceedings.: Orthopedic Aspects of Developmental Disabilities, University of North Carolina Press, Chapel Hill, 1978, pp.60-67.
5. Sussman, M.D., "The Use of Casts as an Adjunet to Physical Therapy Management of Cerebral Palsy Patients," Proceedings: OrthopedicAs- pects of Developmental Disabilities, University of North Carolina Press, Chapel Hill, 1978, pp.47- 59.
6. Sussman, M.D., B. Cusick, "The Role of Short Leg, Tone Reducing Casts as an Adjunct to Physical Therapy of Patients with Cerebral Palsy," John Hopkins Medical Journal, 145,1979, pp.112-114.
7. Bleck, E., Orthopedic Management in Cerebral Palsy, MacKeith Press, London, 1987.
8. Bleck, E., "Locomotor Prognosis in Cerebral Palsy," Developmental Medicine and Child Neurology, 13, 1971, pp.188-195.
9. Bobath, B., K. Bobath, "The mental Treatment," Clinics in Developmental Medicine, 90, J.B. Lippincott, Philadelphia, 1984.
10. Coley, I., Pediatric Assessment of Self-Care Activities, C.F. Mosby Co., St. Louis, 1978.
11. Mann, R.A., "Biomechanics of the Foot," American Academy of Orthopedic Surgeons Atlas of Orthotics, 2nd Edit., C.F. Mosby Co., St. Louis, 1985, pp.112-125.
12. Oatis, C.A., "Biomechanics of the Foot and Ankle Under Static Conditions," Physical Therapy, 68(12), December 1988, pp.1815-1821.
13. Rodgers, M.M., "Dynamic Biomechanics of the Normal Foot and Ankle During Walking and Running," Physical Therapy, 68(12), December 1988, pp.1822-1830.
14. Berhnhardt, D. "Prenatal and Postnatal Growth and Development of the Foot and Ankle," Physical Therapy, 68(12), December 1988, 1831-1839.
15. Tibeno, D., "Pathomechanics of Structural Foot Deformities," Physical Therapy, 68(12), December 1988, pp.1840-1949
16. Perry, J., "Normal and Pathalogic Gait," American Academy of Orthopedic Surgeons Atlas of Orthotics, 2nd Edit., C.F. Mosby Co., St. Louis, 1985.
17. McRea, J. Pediatric Orthopedics of the Lower Extremity, An Instructional Handbook. The Futura Publishing Co., New York, 1985.
18. Cusick, B., Serial Casts: Their Use in the Management of Spasticity Induced Foot Deformity, Words at Work, Lexington, Kentucky, 1987.
19. Hinderer, K., "Effects of 'Tone Reducing' vs. Standard Plaster-Casts on Gait Improvement of Children with Cerebral Palsy," Developmental Medicine and Child Neurology, 30, 1988, pp.370- 377.
20. Watt, Joe, "A Prospective Study of Inhibitive Casting as an Adjunct to Physiotherapy for Cerebral Palsied Children," Developmental Medicine and Child Neurology, 28, 1986, pp.480-488. 21 Yates, H.L., D.H. Mott, "Inhibitive Casting," Proceedings: First William C. Duncan Seminar on Cerebral Palsy, University of Washington, Seattle, 1977.
21. Duncan, W.R., "Foot Reflexes and the Use of the Inhibitive Cast; Foot and Ankle," 4(2), 1983, pp.145-148.
22. Sussman, M.D., "Casting as an Adjunct to Neurodevelopmental Therapy for Cerebral Palsy," Developmental Medicine and Child Neurology, 25, pp.801-805.
23. Cusick, B., "Indications for Solid Below-Knee Casts," 1988.
24. Cusick, B., "Protocol: Serial Long-Leg Cast Application Procedure," 1989.
25. Cusick, B., "Indications for HAFS," 1988.
26. Cusick, B., "Indications for Crouch-Control Designs in Ankle-Foot and Supra Maleolar Splints," 1988.
27. Cusick, B., "Splints and Casts: Managing Foot Deformity in Children with Neuromoto Disorders," Physical Therapy, 68(12), December 1988, pp.1903-1912.
28. Ford, C., "The Neurophysiological Ankle-Foot Orthosis," Clinical Prosthetics and Orthotics, 10(1), Winter 1986, pp.15-23.