Measurement and treatment in cerebral palsy: an argument for a new approach.

Physical therapy approaches for children with cerebral palsy (CP) have not succeeded to the extent envisioned with the development of neurophysiological approaches to treatment. I will argue that this is in part a function of the approaches and their underlying assumptions, the method by which we have attempted to document change, and, perhaps most importantly, the expectation regarding change in motor outcome. In order to improve the motor competence of patients with CP, we need to develop and test new assumptions, focus on the attainment of functional motor outcomes, and utilize creative methods for measuring these functional outcomes. The new assumptions and methods must be formulated from current research and thinking in motor control and motor learning as well as from psychology and the other behavioral sciences. Once formulated, these assumptions and treatment methods must be tested before they are accepted as therapeutic.

Approaches to TreatmentPrevious Assumptions and New Thoughts

The neurological approaches to treatment such as neurodevelopmental treatment (NDT) are founded on assumptions that were not adequately tested at the time the approaches were developed.(1) Subsequent testing of some of these assumptions suggests that we need to develop alternative foundations and modify the approaches themselves. My colleagues and 12- have argued elsewhere that the presumed importance for movement of muscle tone and primitive reflexes needs to be reexamined, as do assumptions relating to motor development such as the relationship of proximal and distal motor abilities. Harris(2) points out that not only are functional outcomes often missing in efficacy research, but the work that has been done has been by our colleagues in other disciplines, not physical therapists. I will highlight some of what is necessary to reexamine.

Increasing evidence exists that abnormalities such as spasticity may not be the sole or even the primary cause of movement dysfunction.(5,6) Spasticity might be conceptualized as multiple problems rather than a single entity classically defined as a velocity dependent resistance to passive stretch. In addition, primitive reflexes may actually be conceptualized as functional movement patterns as opposed to the culprits that may prevent more mature patterns of movement from emerging.(7,8) The premise that control of proximal musculature such as the trunk musculature develops before distal limb muscles for activities such as grasping has guided therapeutic intervention. Empirical evidence suggests that proximal and distal muscle control develops simultaneously rather than in a dependency relationship.(3,9) Simultaneously working toward proximal and distal muscle group control may more accurately reflect the normal developmental sequence, if indeed this sequence should be followed in therapy. The focus of this article will be on the importance of functional movement outcomes for individuals with movement problems such as CP. If functional movement is stressed, however, then measures of functional movement must be utilized for the assessment of the efficacy of therapy.

Documentation of Change

Motor Milestones

Gross and fine motor milestones have been used frequently for the assessment of change following therapeutic intervention.(10,11) Developmental tests such as the Peabody Developmental Motor Scales(l2) and the Bayley Scales of Infant Development(13) have been used as outcome measures in group comparison studies. When improvement in the experimental groups has been demonstrated, it has been modest." A major problem with these types of scales is their failure to measure functional movement or to capture the qualitative aspects of movement. It is these qualitative aspects of movement that may be the most amenable to change through intervention. In addition, the commonly used scales afford no opportunity to assess important soft tissue variables such as functional range of motion and strength, variables that may change significantly after treatment.

Change Versus Functional Change

Therapeutic interventions that stress the remediation of muscle tone for movement and for the development of normal movement patterns often include goals that do not lead to functional improvement.(15) The disparity between movement patterns that constitute typical therapeutic goals and movement patterns that are necessary for improved function win continue to exist until ecologically valid movement goals are developed. These goals should guide the treatment of motor dysfunction and the research in motor control used to develop treatment strategies. Ecologically valid stimuli and treatment programs take into account the movements that are necessary and useful to humans as they move about in their environment.(2,4) Treatment programs are often developed to improve patterns of movement without reference to the functional contexts of the movements. For instance, working on isolated dorsiflexion in a sitting or supine position has little to do with generating the complex action necessary to produce the dorsiflexion required to clear the foot during the swing-through phase of gait. In the same way, the study of motor control using single-joint paradigms will not necessarily lead to understanding the nature of human movement problems that involve multiple degrees of freedom, often in a constantly changing environment.(16) In order to evaluate change in functional movement, fundamental assumptions must be developed regarding the definition of functional movement as well as what constitutes a useful change. Definitions of functional movement may be specific to a population (such as individuals with CP) or even to a person (such as a particular individual with CP), although classes of functional movement may be defined for humans in general.(17) These general classes of movement, or species-specific movement, may include postural control of the head and trunk in sitting and standing positions or movement of the limbs against this postural control in activities such as walking, reaching, and grasping.

My colleagues and I (L Fetters, M Shelley, J Kluzik; manuscript in preparation)(18-20) have examined the organization of movement during reaching and grasping while subjects performed functional movements that are part of their everyday experiences. Traditional experiments in motor control have used movement, such as pointing to pin lights in the dark, that are not commonly experienced in the environment.(21) Models of motor control have been developed from these often two-dimensional, single-degree-of-freedom paradigms. These models have not always succeeded in attempts to understand three dimensional, multiple-degree-of freedom actions.(22) Stabilizing the upper extremity and allowing only flexion and extension movements at the elbow, for example, will yield data and models regarding that type of relatively simple movement. These findings will not necessarily explain the movement control parameters for a freely moving arm reaching to pick up a cup.

In order to understand complex human movement and plan motor intervention strategies, the problems encountered by the mover in functional contexts must be studied. The variables chosen for study must be guided by current behavioral science and neuroscience. The analysis of motion using kinematics provides multiple variables appropriate for study.

Motion Analysis for the Study of Cerebral Palsy

Kinematic analysis is not a new method for the study of movement in CP. Gait analyses using kinematic data are evident in the literature.(23,24) Kinematic data are obtained by identifying particular points on the body and tracking these points in space over some period of time. Each researcher decides on the points of interest. For the study of reach and grasp movements, we record from the wrist, elbow, shoulder, head, and trunk. Many motion analysis systems are available for this type of tracking. I am currently using the WATSMARTIM (Waterloo Spatial Motion Analysis Recording Technique)* coupled with videotape analysis. The particular system chosen should yield reliable measurements and answer the particular questions under study.(25) Some systems require the use of reflective tape over the points of interest, whereas others require later identification and digitizing of points from video or cinematographic film.

The WATSMART's requires attaching light-emitting diodes to the skin over the points of interest. These points may be digitized in real time or identified and digitized after data are collected. The points are recorded at a particular frequency (eg, 100 Hz) and stored in a computer for later reduction and analysis. After points have been identified and stored, the data can be transformed to yield a variety of variables.

Kinematic variables include the displacement of the limb (eg, how far the limb has traveled and in what direction) and derivatives of the displacement, such as the velocity and acceleration of the limb (eg, the speed of the limb and the constancy of the speed). The shape of the displacement (eg, the complexity of the path in relation to the task) and the number of direction changes (eg, the amount of curvature or straightness in the movement) can be studied. In addition, joint angles can be computed with software joining the points together in an elaborate human dot-to-dot portrait. The displacement, velocity, and acceleration of joint angles can also be computed for study. Phase-plane plots, including the plot of one joint against another or the plot of the velocity of a point against its displacement, have also been used to describe movement. Plotting the displacement of the elbow angle in flexion/extension against the displacement of the wrist angle in flexion/ extension during a reach-and-grasp task, for example, may yield a systematic pattern of the relationship of the displacements over time for healthy subjects. This systematic plot would most likely be altered in a subject with CP. The nature and regularity of the plots could be used to understand the atypical movement and also to plan an intervention strategy.

Controversy continues in the research literature as to what constitutes the important variables for control of movement.(16,26) Kinematics comprises one class of variables that have been used for study. The forces generated during movements, which can be estimated from kinematics (reverse dynamics problem) or electromyographic and force-plate data, constitute additional classes of variables.

Visuomotor Control in Cerebral Palsy

We have provided a unique use of kinematics in the study of visuomotor control of children with CP.(19) We have developed outcome measures that appear to be sensitive to change following therapy and are studying change following a variety of treatment strategies, including intense NDT as well as use of motor learning strategies. Although experimentation in a laboratory setting may not be viewed as ecologically valid, it does provide an opportunity for experimental control. We have coupled this desire for control with ecologically valid tasks such as reaching to touch a key pad or communication board, playing a video game, and reaching for food and drink.

The variables we have chosen to study include reaction time, movement time, movement units and how these units change, curvature, associated reactions, and the control of the head and trunk in relationship to the changes in the limb variables. A movement unit is defined as the portion of a reach between one acceleration and one deceleration, describing the stop-start or jerkiness of a Our initial study with preadolescent children with CP suggested that movement time and movement units were affected by treatment.(19) Movement time and the number of movement units(jerkiness) decreased following a single treatment using NDT-oriented therapy. An additional analysis, however, suggests that the limb movement may have improved at the "cost" of trunk control (L Fetters, M Shelley, J Kluzik; manuscript in preparation). That is, the speed and jerkiness of the movement were improved, whereas the overall control of the head and trunk was diminishing for some subjects. This finding suggests that functional outcomes may have a trade-off. Under these circumstances, improvement in limb control may be the most important functional goal for one patient, although it may not be worth the functional cost for another.

An additional, perhaps more important, consideration is that choices of functional outcomes may be necessary. Although for some subjects trunk control decreased during the post treatment measurements, the movements were faster and less jerky. Improved limb movement at the cost of trunk control may be functionally important to a child who is using a communication board. This type of cost benefit relationship has been demonstrated in studies of efficiency in ambulation.(24,29) Ambulation may be accomplished by children with CP at a cost, in terms of increased physiologic effort as indicated by measures such as heart rate and oxygen consumption. The child may gain independence in walking, but it will be at a heightened physiologic expense. As we emphasize the functional aspects of movement, these types of choices may need to be made increasingly. For the child with CP, the goal of normal movement patterns, as advocated by the neurophysiologic approaches, may be less desirable than the achievements of functional outcomes as defined by the child and his or her family.

Functional Movement Variables as Classification Variables

Historically, CP has been defined and categorized by the parts of the body involved and the nature of the muscle tone.(30) Diagnoses such as spastic quadriplegia, diplegia, or hemiplegia, however, ignore the fundamental aspects of movement that may be disordered.(2) Research in CP typically has been conducted by grouping subjects according to topographical description. One subject with spastic diplegia, however, may have a very different set of movement problems than another subject with the same diagnosis. Grouping subjects for study, as well as for selecting appropriate treatment by the use of more fundamental movement variables, leads to a clearer understanding of the effects of intervention on particular aspects of movement. Variables might include the ability to generate adequate speed of movement, the generation and maintenance of adequate force production and timing the sequencing of force production, and the ability to generate adequate frequencies of movements at constant speeds. Only the combined research efforts of clinical and research physical therapists will define the clinically relevant variables.

Summary

We are at an important time in the development of treatment strategies for patients with movement disorders. Traditional neurophysiologic approaches to treatment need to be examined in terms of the basic assumptions and treatment ideas. It is time for a shift in our paradigms for both research and clinical practice toward the study and treatment of species specific human movement. Current knowledge from motor control, motor learning, psychology, and many of the basic and behavioral sciences needs to be blended into our patient practices. It is critical to develop conceptual bases for treatment that are based on empirical evidence. Sound bases for treatment of movement disorders will evolve only as scientific bases are empirically developed. References 1 Bobath K. A Neurophysiological Basis for the Treatment of Cerebral Palsy. Philadelphia, Pa: JB Lippincott Co; 1980. 2 Harris SR. Efficacy of physical therapy in promoting family functioning and functional independence for children with cerebral palsy. Pediatric Physical Therapy 1990;2:160-164. 3 Fetters L, Fernandez B, Cermak S. The relationship of proximal and distal components in the development of reaching. Journal of human Movement Studies. 1989;12:832. 4 Fetters L. Developmental Concepts for Therapeutic Intetention. (In Touch Series. Topics in Pediatrics: Lesson 7,) Alexandria, Va: American Physical Therapy Association; 1990. 5 Sahrmann SA, Norton Bj. The relationship of voluntary movement to spasticity in the upper motor neuron syndrome. Ann Neurol. 1977;2:460-465. 6 Landau WM. Spasticity: the fable of a neurological demon and the emperor's new therapy, Arch Neurol. 1974;31:217 219. 7 Oppenheim RW. Ontogenetic adaptations and retrogressive processes in the development of the nervous system and behaviour: a neurological perspective. In: Connolly Kj, Prechtl HFR, eds. Maturation and Development: Biological and Psychological Perspectives. Philadelphia, Pa: JB Lippincott Co; 1981:73-109. 8 Easton TA. On the normal use of reflexes. American Scientist. 1972;60:591-599. 9 Loria C. Relationship of proximal and distal function in motor development. Phys Ther. 1980;60:167-172. 10 Carlsen PN. Comparison of two occupational therapy approaches for treating the young cerebral palsied child. Am j Occup Ther. 1975;29:267-272, 11 Palmer FB, Shapiro BK, Wachtel RC, et al. The effects of physical therapy on cerebral palsy: a controlled trial in infants with spastic diplegia. N Engl J Med 1988;318:803-808. 12 Folio MR, Fewell RR. Peabody Developmental Motor Scales and Activity Cards. Allen, Tex: DLM Teaching Resources; 1983. 13 Bayley N. Bayley Scales of infant Development. New York, NY: Psychological Corp; 1969. 14 Ottenbacher Kj, Biocca Z, DeCremer G, et al. Quantitative analysis of the effectiveness of pediatric therapy: emphasis on the neurodevelopmental treatment approach. Phys Ther. 1986;66:1095-1101. 15 Gordon J. Assumptions underlying physical therapy intervention: theoretical and historical perspectives. In: Carr CH, Shepard RB, eds. Movement Science: Foundations for Physical Tberapy, in Rehabilitation. Rockville, Md: Aspen Publishers Inc; 1989. 16 Gottlieb GL, Corcos DM, Agarwal GC. Organizing principles for single-joint movements, 1: a speed-insensitive strategy. j Neurophysiol 1989;62:342-357. 17 Fetters L, Holt K. Efficiency of movement: biomechanical and metabolic aspects. Pediatric Physical Therapy. 1990;2:55-59. 18 Fetters L, Todd J. Quantitative assessment of infant reaching movements. Journal of Motor Behavior. 1987; 19:147-166, 19 Kluzik J, Fetters L, Coryell J. Quantification of control: a preliminary study of effects of neurodevelopmental treatment on reaching in children with spastic cerebral palsy. Phys Ther. 1990;70:65-78. 20 Daleiden S, Fetters L. Kinematic analysis of reaching by active elders: a preliminary study, Presented at the 1988 joint Congress of the American Physical Therapy Association and the Canadian Physiotherapy Association; june 12-16, 1988; Las Vegas, Nev, 21 Atkeson CG, Hollerbach JM. Kinematic features of unrestrained vertical arm movements. J Neurosci. 1985;5:2318-2330. 22 Cordo PJ, Horak FB, Moore SP. On to real-life movements. (Commentary to Gottlieb GL, Corcos DM, Agarwal GC. Strategies for the control of voluntary movements with one mechanical degree of freedom.) Behavioral and Brain Sciences. 1989;12:189-250. 23 Olney Sj, Costigan PA, Hedden DM. Mechanical energy patterns in gait of cerebral palsied children with hemiplegia. Phys Ther. 1987;67:1348-1354. 24 Rose J, Medeiros JM, Parker R. Energy cost during assisted ambulation. Dev Med Child Neurol 1985;27:485-489. 25 Scholz P. Reliability and validity of the WATSMART three-dimensional optoelectric motion analysis system. Phys Ther 1989;69:679-689. 26 Stein RB, What muscle variable(s) does the nervous system control in limb movements? Behavioral and Brain Sciences. 1982;5:535-541. 27 von Hofsten C. Eye-hand coordination in newborns. Dev Psychol 1982;18:450-461. 28 Brooks VB. Some examples of programmed limb movements, Brain Res, 1974;71:299-308, 29 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;26:607612. 30 Bax MCO. Terminology and classification of cerebral palsy. Dev Med Child Neurol 1964;6:295-297.