COMPARATIVE KINEMATIC ANALYSIS OF THERMOPLASTIC AFO DESIGNS

Clinical experience suggesting that ankle-foot orthoses significantly improve the gait of some patients following CVA provides the foundation for this preliminary study. Thermoplastic ankle-foot orthoses have evolved from molded solid ankle devices to include ankle joint designs exhibiting a wide range of inherent mechanical characteristics including varying degrees of plantar flexion and dorsiflexion and differences in brace stiffness. The use of devices with such a wide range of features raises numerous questions regarding the relationship between a specific design and its potential for improving function in a specific patient. The answer to these design questions requires that we understand the inherent structural properties of braces especially those properties that describe the mechanical behavior of the brace in isolation from the patient and the relations among these properties and gait when used by a patient. In particular, what are the inherent structural properties of a series of braces of different designs?

It is generally accepted that wearing an ankle-foot-orthosis can have a positive effect on the gait of some patients post-CVA (Ryerson, 1988; Edelstein, 1994; Huber, 1990). Prescriptions for orthoses have been generally based on these ideas. However, as implemented, prescriptions generally rely on qualitative assessment rather than quantitative data (Sarno and Lehneis, 1971; Sarno, 1973; Lehmann, 1979). More recently, prescription of ankle-foot orthoses has moved toward a more quantitative process (Yamamoto et al. 1993a; Yamamoto et al., 1993b; Sumiya et al., 1996a; Sumiya et al., 1996b). Quantitative attempts at a prescription for ankle-foot orthoses have clearly shown measured differences in stiffness among different orthoses and that the characteristics of the orthosis have measurable effect on gait. A particularly striking feature of a comparison of trimlines (that is, the amount of plastic removed from the medial and lateral aspect of the brace) to effectiveness of an orthosis in gait shows a relatively narrow range of acceptable trim for a particular individual (Sumiya et al., 1996b). This raises a question as to the effectiveness of a purely qualitative approach to prescribing orthoses. A qualitative prescription procedure may specify a type of orthosis (solid ankle, posterior leaf spring, etc.) that would be acceptable to a particular individual, but based on the results in Sumiya et al., it cannot lead to the best brace for an individual. The control of stiffness by trimming or any other method is subject specific. This preliminary biomechanical analysis is designed to improve the quantitative prospective prescription of ankle-foot orthoses through measured structural properties of orthoses.

In preparation for this preliminary study we developed a device to evaluate ankle-foot orthoses statically (Figure 1). This device is designed to evaluate inherent structural properties of the orthoses. When testing braces statically, we characterize orthoses by (1) the inherent stiffness (the slope of the moment-flexion angle curve) and (2) the rotation of the foot segment of the orthoses with respect to the shank portion and the location and orientation of the screw axis. The accuracy of both the moment and kinematic measurements were characterized prior to collecting the preliminary data. The OPTOTRAK^TM motion analysis system (Northern Digital Inc., Waterloo, Canada) was used to determine the kinematics of the orthoses. This system tracks the three- dimensional coordinates of markers (infrared emitting diodes) placed on the shank and fool sections of the orthoses. The markers were arranged into two arrays of four markers each, with one marker out of plane with respect to the other three. One of these arrays was fixed rigidly to both the shank and foot. Rotational accuracy measured with the OPTOTRAK^TM was determined to be ±0.05 degrees.

To describe the inherent structural properties of the orthoses statically we applied a moment in one of three orthogonal planes corresponding to either dorsiflexion/plantar flexion, or inversion/eversion, or adduction/abduction (Rasch, 1989). According to the sign convention we used dorsiflexion, inversion and adduction were positive. Orthoses were fit on a prosthetic foam mold of a shank. The mold was firmly attached to the testing device by means of a T-shaped metal frame cast into the shank segment. The surrogate shank was cast to the same shape as the shank segment of the mold used to form the orthoses. Each orthosis was attached to the surrogate shank with a proximal strap which is typically how these devices are used clinically. Moments were applied to the foot of the orthosis through a lever arm and were measured by three independent strain gage circuits, one for each direction of ankle movement. Orthoses were tested by manually applying a varying moment about one axis while measuring moments about all three axes.

Four orthoses were tested using this device: a solid ankle orthosis, a posterior leaf spring orthosis, a locked hinge orthosis1, and a flexible hinge orthosis2. Moment and angle data obtained from these orthoses were compared with those for a normal ankle in the weight acceptance and weight release phases of gait. These comparisons were made in dorsiflexion/plantar flexion using data from approximately 0.2 sec following heel strike and prior to toe off. (Winter, 1990). Measurements among three braces of substantially different design (solid ankle, locked hinge, flexible hinge) showed that the solid ankle brace was the stiffest in plantar flexion (Figure 2). In contrast, the stiffness of the solid ankle brace decreased with increasing dorsiflexion. This appeared to be due to buckling of the medial and lateral edges of the orthosis as they were compressed during dorsiflexion. This was the only brace that showed this behavior. Stiffness of the locked hinge brace increased with both increasing dorsiflexion and plantar flexion (Figure 2). The magnitude of the maximum moments were similar to those of the solid ankle brace but occurred at different amounts of dorsiflexion/plantar flexion. Despite having moments of similar magnitude in dorsiflexion, the stiffncss of the locked hinge and solid ankle brace were quite different. The stiffness of the locked hinge brace increased with increasing dorsiflexion while it decreased for the solid ankle brace. The form of the moment-angle curve for the flexible hinge brace was similar to that of the locked hinge brace however the magnitudes of the maximum moments were much lower for the flexible hinge brace and rotation in dorsiflexion was greater (Figure 2). Relative to the solid ankle and locked hinge braces, moments remained low over a large range of angular displacements suggesting this brace would give less support at the ankle.

Comparing the stiffness of each orthosis with the stiffness at the ankle in normal gait shows that the orthoses are generally stiffer. During weight acceptance, when the foot is plantar flexing, the normal ankle has an average stiffness of 3.8 Nm/degree. In weight release the ankle is again in plantar flexion, normal ankle moments are the highest, and ankle stiffness is approximately 3.8 Nm/degree. This is lower than the stiffness of the solid ankle brace but comparable to that of the locked hinge. In these preliminary studies we have characterized the kinematics of the braces by the rotation angles under load and by the location of the point at which the screw axis intersects a sagittal plane passing through the most medial aspect of the medial malleolus. For the solid ankle brace, the intercept moves essentially inferosuperiorly along a line approximately 35 mm posterior to the center of the malleolus of the surrogate (Figure 3). For the posterior leaf spring brace the intercept motion again was predominately in the inferosuperior direction but the path has shifted approximately 10 mm to 15 nun posteriorly (Figure 4). For the flexible hinge brace, the intercept again was posterior to the center of the malleolus. However, for the flexible hinge brace the intercept moved approximate 10 mm in the anteroposterior direction as well as in the inferosuperior direction. Again, this results in a center of rotation of the foot section that is posterior with respect to the malleolus.

These preliminary assessments illustrate the dependence of structural stiffness and kinematics on both gross brace design changes (solid ankle brace, flexible ankle brace, locked hinge brace) as well the effect of parametric changes in specific characteristics of a single basic design (trim line as applied to the solid ankle brace and posterior leaf spring brace). These results also indicate that orthoses do not necessarily produce the same response as a normal ankle.

References

1. Edelstein, J.E. (1994) Orthotic assessment and management. In Physical Rehabilitation: Assessment and Treatment, eds S.B. O'Sullivan and T.J. Schmitz. F.A. Davis Co., Philadelphia.

2. Huber, S.R. (1990) Therapeutic application of orthotics. In Neurologic Rehabilitation, ed D.A. Umphred, C.V. Mosby Co., St. Louis.

3. Lehmann, J.F. (1979) Biomechanics of ankle-foot orthoses: Prescription and design. Archives of Physical Medicine and Rehabilitation 60, 200-207.

4. Rasch, P.J. (1989) Kinesiology and Applied Anatomy. Lea and Febiger, Philadelphia.

5. Ryerson, D.S. (1988) The foot in hemiplegia. In Physical Therapy of the Foot and Ankle, ed. C.G. Hunt. Churchill Livingston, New York.

6. Sarno, J.E. (1973) Below knee orthoses: A system, for prescription. Archives of Physical Medicine and Rehabilitation 54, 548-552.

7. Sarno, J.E. and Lehneis. H.R. (1971) Prescription considerations for plastic below-knee orthoses. Archives of Physical Medicine and Rehabilitation 52, 503-510.

8. Sumiya, T., Suzuki, Y. and Kashahara, T. (1996a) Stiffness control in posterior-type plastic ankle-foot orthoses: effect of ankle trimline Part 1: a device for measuring ankle moment. Prosthetics and Orthotics International 20,129-131.

9. Sumiya, T, Suzuki, Y. and Kashahara, T. (1996b) Stiffness control in posterior-type plastic ankle-foot orthoses: effect of ankle trimline Part 2: orthosis characteristics and orthosis/patient matching. Prosthetics and Orthotics International 20,132-137.

10. Yamamoto, S., Ebina, M., Iwasaki, M., Kubo, S., Kawai, H. and Hayashi, T. (1993a) Comparative study of mechanical characteristics of plastic AFOs. Journal of Prosthetics and Orthotics 5, 59-64.

11. Yamamoto, S., Ebina, M., Iwasaki, M:., Kubo, S., Kawai, H. and Hayashi, T. (1993b) Quantification of the effect of dorsi-/plantar flexibility of ankle-foot orthoses on hemiplegic gait: A preliminary report. Journal of Prosthetics and Orthotics 5,42-48.