A Geometric Model to Predict the Effect of Ankle Joint Misalignment on Calf Band Movement in Ankle Foot Orthoses

Introduction: Accurate alignment of anatomical and mechanical joint axes is one of the major biomechanical principles pertaining to articulated orthoses, yet knowledge of the potential effects of axis misalignment is limited. Accurate joint alignment is a challenge since the location and orientation of joint axes have long been debated. The ankle joint has been described by some investigators as approximating an ideal fixed hinge joint [1-3] and by others as a polycentric joint [4, 5]. Congruency between anatomical and mechanical joint axes is considered important because misalignment results in undesirable forces (both shear and compressive) and moments generated as the joints move through their range [6, 7]. Inappropriate alignment of joint axes has consequences for the soft tissue at the interface and the integrity of the joints (mechanical and anatomical).

Current orthotic technology requires the identification of a best-fit hinge axis position for the alignment of most mechanical ankle joints, especially single axis joints. Based on anthropometric studies of cadaver legs, Isman and Inman [1] determined that the talocrural axis could be established by points 3-5 mm distal to the distal tips of the lateral and medial malleoli. Orthotic texts base their alignment of orthotic ankle joints on this description of the anatomical ankle joint location, recommending that for the purposes of alignment the axis be approximated as passing through the centers of the medial and lateral malleoli at the level of the distal tip of the medial malleolus [8, 9]. Joint axis misalignment consists of two components: linear and angular misalignments [10].

Misalignment of orthotic ankle joints is believed to result in vertical calf band travel or pistoning [11]. Lamoreux [10] observed that anterior-posterior misalignment of a prototype dual-axis mechanical ankle joint resulted in vertical travel of the calf band. Based on clinical observation of pistoning, Rubin and Danisi [12] proposed a mobile calf band to absorb shear forces occurring at the interface with the skin due to pistoning. More recently, Sumiya et al. [13] used mechanical models to determine the degree of pistoning of the calf band that occurred with alterations in trim-line location in a non-articulated, shoe-horn type, polymer ankle foot orthosis (AFO). They reported that the instant center of rotation was located posterior to the anatomical ankle joint axis and dispersed over the junction between the calf shell and the shoe insert, with dorsiflexion movement tending to produce instant centers of rotation that were inferior to the anatomical ankle joint axis and plantarflexion movement tending to produce instant centers of rotation that were at the level of the anatomical ankle joint axis. These combined misalignments between the anatomical and mechanical axes of rotation led to pistoning of the calf band: upward sliding in plantarflexion and downward sliding in dorsiflexion. Pistoning increased with ankle angle for both dorsiflexion and plantarflexion.

The purpose of this project was to model the effects of systematic linear (anteriorposterior and proximal-distal) misalignments of single axis mechanical ankle joints in an AFO in order to determine the degree and direction of calf band travel that would occur over a functional range of ankle joint motion.

Figure 1 (A) AFO with mechanical joints that are not aligned with the ankle joint. (B) Simple planar model of the AFO involving three links. A grounded link is used to represent the foot and the lower part of the orthosis (a shoe in this diagram). Two other links are used to represent the leg and the upright of the AFO.

Methods: Sagittal plane misalignments of the ankle joint centers of an AFO were simulated using a simple two-dimensional model (Figure 1). The model consists of three links. Calculations are made using one link as a reference point (grounded link). This link represents the combination of the foot and the lower portion of an articulated orthosis. The model assumes that these parts do not move relative to each other. The two other links have pivot joints with the grounded link. The distance between the pivot joints represents the amount of misalignment of the mechanical joint axis of the orthosis with the anatomical ankle joint axis. The two links that pivot with the grounded link represent the anatomical shank (tibia-fibula and surrounding tissues – “leg link”) and the upper part of the articulated AFO (“upright link”). The end of the upright link joins the leg link at a sliding joint that represents the calf band of the AFO. In a real AFO there can be slight movements at the calf band interface and also between the foot and the lower portion of the articulated AFO. However, this model allows movements only at the calf band. Results of simulations must be analyzed with this in mind.

The relative calf band movements were determined using an able-bodied adult’s ankle dorsiflexion curve for self-selected normal walking speed as input and for the following misalignments: (1) mechanical joint placed 5% of the length of the shank anterior to the anatomical joint, (2) mechanical joint placed 5% of the length of the shank posterior to the anatomical joint, (3) mechanical joint placed 5% of the length of the shank proximal to the anatomical joint, and (4) mechanical joint placed 5% of the length of the shank distal to the anatomical joint.

Figure 2 Proximal calf band travel (normalized to shank length) as a function of the gait cycle, using an average ablebodied ankle dorsiflexion curve as input.

Results: In early stance when the ankle is plantarflexing, posterior and distal misalignments would both cause proximal movements of the calf band with respect to its neutral position, whereas anterior and proximal misalignments would both cause distal movement of the calf band with respect to its neutral position (Figure 2). During the midstance portion of the gait cycle, the ankle is dorsiflexing from its neutral position. Therefore, anterior and distal misalignments would both cause proximal calf band movement, while posterior and proximal misalignments would both cause distal calf band movement. The effects of misalignments with similar magnitudes in the anterior-posterior direction have a much larger effect on calf band travel than misalignments in the proximal-distal direction. See Table 1 for a summary of the simulation results.

Table 1 Direction of calf band travel predicted by model.

Discussion: Clinical observation [12, 14], and mechanical [10, 13] and computer models all support the idea that the calf band of an AFO does travel or piston when the ankle joint is misaligned and moved through a range of motion. The results of this simulation agree with ‘hypothetical’ diagrams presented in the NYU manual [11]. Our results indicate that anterior misalignments will cause the calf dorsiflexion and distally with plantarflexion, and will have a converse effect for posterior misalignments. Proximal misalignments would produce only distal movements of the calf band while distal misalignments would cause only proximal movements. Ankle joint axis misalignments within an AFO may be undesirable in instances where calf band travel would result in shear forces that compromise the integrity of the skin.

Lamoreux [10] attributed calf band travel predominantly to anterior-posterior misalignment of their prototype dual-axis ankle joint. Similarly, Sumiya et al. [13] found that the instantaneous centers of rotation of posterior-type polymer AFOs were always posterior to the anatomical ankle axis and concluded that the pistoning movements they measured between the calf band and the calf were due to this posterior positioning. The model presented in this paper for single axis mechanical joints indicates that calf band movement is more sensitive to misalignments in the anterior-posterior direction than in the proximal-distal direction. The effects of the small misalignments noted in the proximal-distal direction on calf band movement in the Sumiya et al. [13] study were likely overshadowed by the effects of the anterior-posterior misalignments.

Ankle joint axis misalignments may lead to increased resistance to ankle joint motion. Bottlang et al. [7] studied similar issues of axis misalignment using cadavers to identify a best-fit hinge position for the application of articulated external ankle fixation that minimized ankle motion resistance. The authors reported that the energy required to rotate the unconstrained ankle was negligible but was found to increase when external fixation was applied along a bestfit axis determined by them. Distal misalignments of the mechanical joint axis required less energy than did proximal misalignments. Bottlang et al. [7] suggested that the consequences of ankle axis misalignment included increased compressive forces at the articular surfaces and/or tension in the ligaments that constrain the ankle joint. They indicated that this would be likely to happen for articulated external fixation that imposes a horizontal hinge to the ankle joint when the best-fit axis was determined to be externally rotated and obliquely oriented. Since similar constraints apply to the alignment of single axis mechanical ankle joints used in orthoses (i.e., they must be aligned in order to function smoothly), there may be similar consequences for the anatomical ankle joint and ligaments with use of an AFO.

In attempting to understand the consequences of these findings, two additional factors should be considered: the degree of friction at the interface between the calf band and leg, and the phase of the gait cycle. Assuming friction is present at the interface between the calf band and leg, distal misalignments would push the leg out of the orthosis throughout the entire gait cycle while proximal misalignments would effectively pull the leg into the orthosis throughout the gait cycle. However, if friction is absent and the phase of the gait cycle is considered, the opposite might occur. The ‘clinical rule of thumb’, which suggests that a slightly shorter stirrup be used in the event a correct length stirrup is unavailable, appears to be reasonable if minimal friction is assumed. During loading response body weight drives the heel into the shoe, while in terminal stance heel-off occurs. At heel-off the ankle is plantarflexing (from a position of peak dorsiflexion) and a proximal misalignment should cause the calf band to move distally relative to the leg. The combined effect of heel-off and distal calf band travel would tend to push the heel further out of the shoe than if the axes were aligned correctly or if there was a distal misalignment. Our model assumed that there was no movement between the foot and the footplate of the AFO, making it difficult to comment on the effect of calf band travel on the heel. Many factors can affect movement of the heel within the AFO, including the fit of the orthosis, presence and location of an ankle strap, degree of tension on ankle and calf straps, the type of interface material, the type of joint (single versus multiaxial), and the type of shoe. However, in general, clinicians should probably be more concerned with anterior-posterior misalignments as they would tend to produce greater magnitudes of calf band travel. This finding has implications for the accuracy of the tibial torsion measurement since it is important to the determination of the anterior-posterior location of the ankle joints [8].

The results of the two-dimensional model presented here are limited to linear misalignments of single axis mechanical joints in the sagittal plane and their effect on calf band travel. The model does not predict the effect of angular axis misalignments on calf band travel or any other outcome variable, such as forces or moments applied to the leg by the calf band. Further work is required to develop a three-dimensional model that can more completely analyze the many combinations of ankle axis misalignments possible in an AFO and their effect not only on calf band travel, but also on other variables such as ankle joint motion resistance and compressive forces at the articular surface. Further work is also required to develop predictive models of the effect of misalignment of joints at the knee and hip and multi-joint models that would allow us to analyze the impact misalignment of the ankle joint axis may have at the knee and hip, and vice versa.

References: [1] Isman, R. & V. Inman, Bull Prosthet Res, 1969. 10(11): 97-129. [2] Singh, A.K., et al., Foot Ankle, 1992. 13(8): 439-46. [3] van den Bogert, A.J., et al., Biomech, 1994. 27(12): 1477-88. [4] Sammarco, G.J., et al., Orthop Clin North Am, 1973. 4(1): 75-96. [5] Lundberg, A., et al., J Bone Joint Surg Br, 1989. 71(1): 94-9. [6] Lehneis, H.R., Orthot Prosthet Appl J, 1964. 18: 110-14. [7] Bottlang, M., et al., J Biomech, 1999. 32(1): 63-70. [8] Lehneis, H., Principles of Orthotic Fit and Alignment, in New York University Lower-Limb Orthotics: Including Orthotists' Supplement - Prosthetics and Orthotics. 1974, New York University Post-Graduate Medical School. p. 181-89. [9] Mann, R.A., Biomechanics of the Foot, in Atlas of Orthotics. 1985, CV Mosby Co.: St Louis. p. 112- 125. [10] Lamoreux, L., Bull Prosthet Res, 1969. 10(11): 146-83. [11] New York University, Lower-Limb Orthotics: Including Orthotists' Supplement - Prosthetics and Orthotics. 1974: New York University Post-Graduate Medical School. [12] Rubin, G. & M. Danisi, Othot Prosthet, 1974. 28(1): 43-4. [13] Sumiya, T., et al., J Rehabil Res Devel, 1997. 34(3): 279-85. [14] Rubin, G. & M. Dixon, Bull Prosthet Res, 1973. 10(19): 20-41.

Acknowledgements: This work was funded by the National Institute on Disability and Rehabilitation Research (NIDRR) of the U.S. Department of Education under Grant No. H133E030030. The opinions contained in this publication are those of the grantee and do not necessarily reflect those of the Department of Education. The authors wish to acknowledge Michael Brncick, MEd, CPO, for his insight and clinical guidance on this project.