PHYS THER
Vol. 84, No. 4, April 2004, pp. 355-356
Invited Commentary
James C Galloway
Infant Motor Behavior Laboratory
Department of Physical Therapy
Biomechanics and Movement Sciences Program
301 McKinly Lab
University of Delaware
Newark, DE 19716
 |
Introduction
|
|---|
Let me first congratulate Fonseca and colleagues on an article that advances our understanding of important features of gait in children with cerebral palsy (CP). Moreover, this article highlights the question of how these features emerge during development, which is, of course, a central issue to pediatric neurorehabilitation.
In summary, the authors found that children with hemiplegic CP produced greater vertical stiffness, greater potential energy (K/P energy, also called "PE" in the article), and a decreased landing angle during stance on their more affected extremity compared with the less affected extremity as well as with age-matched, typically developing children. The authors suggest that their data and their previous work1 are compatible with a novel view of the underlying process by which children with CP develop independent locomotion, namely that these "impairments" are, in part, adaptations by the child to the mechanical and energetic requirements of walking.
I have focused my comments first on the authors' "selection theory," as I call it, followed by a few points on the method and results, several of which were touched on in the article's "Discussion" section.
Selection Theory
Atypical movement patterns in people with pathology of the nervous system, such as those of children with CP, are often viewed as "impairments" imposed primarily by the initial neurological insult. Fonseca et al propose an interesting and important alternative view, namely that these features are adaptive and develop, at least in part, as a result of the child optimizing his or her own system in order to walk. Here the child is viewed as an active participant in a process of exploration and selection of features to match the task requirements of gait such as moving the center of gravity forward during stance.
The authors' proposal is in line with the exploration and selection principles of modern theories of infant development2–5 and is a much-needed application of these principles to pediatric neurorehabilitation. I would submit, however, that these particular data are largely ambiguous on whether children adaptively select these features or whether these features are neural and mechanical constraints as traditionally viewed. One assumption underlying the selection theory is whether children with CP have the flexibility within these features from which to explore? For example, earlier in their walking development, did they have a sufficiently wide range of stiffness from which to explore? If so, do they, in fact, select a narrower, task-specific range of stiffness? These simple questions demand complex experiments tracking multiple behaviors in young children with CP before they become independent walkers. This is not so much a criticism as a reflection of my own impatience with not having definitive support for such an attractive theory.
Vertical Stiffness Calculation
Although mass-spring models continue to provide useful information, there are limitations to these calculations (of which these authors are well aware). First, the authors calculated vertical stiffness based on a spring model that assumes the line of force application and line of displacement are both perpendicular to the ground. During the stance phase of a typical gait pattern, however, this is not always the case. This may be particularly relevant for the atypical gait patterns of children with CP. It would be interesting to compare these findings with other, less restrictive models of stiffness such as that proposed by McMahon and Cheng.6 Second, stiffness was calculated as the slope of the linear regression between vertical acceleration and displacement of the center of mass. As a result, there is an assumption of a relatively linear relationship, reflecting a simple spring. Given that the gait of children with CP is atypical, the acceleration-displacement relationship may not be linear. Nonlinearity could affect the meaning of group differences in stiffness, especially if the control group's stiffness is relatively linear. Lastly, an obvious follow-up question of clinical interest is to what degree do hip, knee, and ankle stiffness contribute to global stiffness? It would be particularly interesting to see the distribution of joint stiffness across both the slower and faster speeds, given that only the former showed group differences.
Changes With Speed
The authors normalized stiffness, K/P energy, and landing angle to walking speed. Given that certain features of the atypical gait of children with CP may well show nonlinear changes over speeds, it would have been interesting to see the non-normalized data. This is relevant given that differences in stiffness were seen only at the slowest speeds and important group differences in K/P energy were seen only at the fastest speed. Moreover, the implications of these group differences are not completely clear, given that children with CP typically walk at their preferred walking speeds where no differences in stiffness or K/P energy were noted for the more affected lower extremity.
Role of Plantar-Flexed Ankle
Children with CP differ from typically developing children in many features of their neuromusculoskeletal system, including passive and active range of motion (ROM). It is not clear what trunk, hip, knee, and ankle ROM was present in the children with CP in this study. As briefly mentioned by the authors, the concern here is that a lack of passive ROM, such as at the ankle, could account for the changes in global stiffness, K/P energy, and landing angle. With the small numbers of subjects, a few children lacking even a small amount of ankle dorsiflexion ROM may have skewed the group results. This may be important because very different interventions could be envisioned depending on whether the differences in stiffness, K/P energy, and landing angle were associated with a significant limitation in lower-extremity passive or active ROM or produced with full ROM.
The authors state that limitations of passive ROM were probably not a primary factor in group differences in their variables because the non–speed-normalized values scaled up with speed. The relationship between features, such as stiffness and ROM, during functional movements may not be so straightforward. Stiffness, K/P energy, and landing angle are the result of the interplay of multiple joints. Thus, landing angle, for example, could theoretically change with speed even with little ankle movement due to changes at the trunk, hip, and knee. I again encourage these authors to include additional joint components, such as hip and knee stiffness, within their future work on the development of atypical gait patterns as well as the effects of intervention.
The authors propose that both typical and atypical movements emerge from the adaptive interplay of the nervous system, the body's properties, and the environment. This is the type of rich theoretical grounding necessary to advance modern pediatric neurorehabilitation. I look forward to the authors' follow-up studies and wish them continued success.
|
References
|
|---|
- Fonseca ST, Holt KG, Saltzman E, Fetters L. A dynamical model of locomotion in spastic hemiplegic cerebral palsy: influence of walking speed.
Clin Biomech (Bristol, Avon).2001; 16:793–805.
- Edelman GM.
Neural Darwinism: The Theory of Neuronal Group Selection. Oxford, United Kingdom: Oxford University Press;1987
.
- Gibson EJ. Exploratory behavior in the development of perceiving, acting and the acquiring of knowledge.
Annu Rev Psychol.1988; 39:1–41.[Web of Science]
- Thelen E, Smith L.
A Dynamic Systems Approach to the Development of Cognition and Action. Cambridge, Mass: MIT Press;1994
.
- Turvey MT, Fitzpatrick P. Commentary: development of perception-action systems and general principles of pattern formation.
Child Dev.1993; 64:1175–1190.[Web of Science][Medline]
- McMahon TA, Cheng GC. The mechanics of running: how does stiffness couple with speed?
J Biomech.1990; 23:65–78.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Copyright © 2004 by the American Physical Therapy Association.
Top
References
