Using Gait Analysis to Compare teh Sagittal Plane Biomechanical Effects of Three Designs of Ankle Foot Orthoses
Brock Sande, CPO
Todd Dewees, CPO
Dr. Michael Aiona, MD
Ankle foot orthoses (AFOs) are a common treatment for patients with a wide range of lower extremity disability. Through the phases of the gait cycle the AFO resists detrimental motions and promotes beneficial forces to improve the patient's gait. After clinical evaluation and gait observation, the physician prescribes and the orthotist designs an AFO to specifically suit each patient and their gait. There are a few general brace designs to choose from such as solid ankle, articulated, and posterior leaf spring. Each design affects the biomechanics of the leg differently, but the goal of every AFO is to promote the most functional motions and forces of gait.
Gait is a complex system of simultaneous motions and corresponding forces acting through several joints in the body. Through years of studying gait, the medical community has widely adopted a system of defining the phases, motions, and forces associated with gait.1 This study will focus on the biomechanics in the sagittal plane because those characteristics are the most pertinent to forward progress. The weight bearing events at the ankle are divided into three stages called first, second, and third rockers.2 The first rocker occurs during the loading response phase of gait and consists of the dorsiflexor muscles eccentrically contracting to allow 10 degrees of plan-tarflexion.3 As the tibia is pulled forward by the first rocker the knee consequently undergoes a knee flexion moment which is resisted by the activation of the quadriceps muscles in order to maintain stance stability. The second rocker occurs during the midstance phase of gait and consists of the plantarflexor muscles eccentrically contracting to allow the tibia to advance to 10 degrees of dor-siflexion.2 Lastly, the third rocker occurs in late stance and consists of the plantarflexors contracting to decelerate the advancement of the tibia and create a distal lever arm for propulsion.2 The primary propelling force in reciprocal gait is created by gravity acting on the body weight through the limb.3 The propelling momentum is primarily absorbed and maintained by rockers in the ankle-foot complex and moments at the knee. In normal gait, the transfer of body weight onto the extended limb has been documented to be 60% of the person's body weight in 0.02 seconds.3 In pathological gait, the forces are often even greater and more abrupt.
Patients who are treated with AFOs typically have musculoskeletal involvement which prevents them from utilizing the anatomical rockers.4 Solid ankle AFOs are used to substitute for the rockers and promote tibial advancement by limiting the plan-tarflexion motion at the ankle. However, by limiting the plantarflexion motion the solid ankle AFO transfers a greater flexion moment to the knee and consequently requires a greater knee extension contraction from the quadriceps.5,4 To prevent knee instability in early stance, it is recommended that patients have at least fair (3/5 MMT) knee extension strength to be suitable candidates for treatment with a solid ankle AFO.4 Although posterior leaf spring or articulated AFO designs can reduce knee instability, the designs can also have drawbacks.6 For example, the posterior leaf spring (PLS) design allows relative plantarflexion at loading response and thus reduces knee flexion moment, but the design also allows excessive dorsiflexion in late stance due to an inadequate distal lever arm. The PLS design is also unable to control medial lateral motion at the ankle if the patient requires ML stability. Although articulated AFOs can be aligned to allow some relative plantarflexion and consequently reduce the excessive knee flexion moment, this alignment would also allow some plantarflexion during swing phase of gait and can cause toe drag. Articulated AFOs also require more weight, bulk, and maintenance. There are patients that would benefit from an AFO design that allows relative plantarflexion and a reduction in knee flexion moments while still maintaining appropriate alignment and distal lever arm for propulsion.
After observing patients with excessive knee flexion moments, orthotists at Shriner's Hospital's for Chil-dren-Portland designed a solid ankle orthosis which allows the heel to compress at loading response. A literature review found no other published research that studied a comparable AFO design. The purpose of this research is to compare the biomechanical effects in the sagittal plane of an AFO designed with a compression heel to the standard solid ankle design.
- The solid ankle design orthoses promote peak knee flexion moments greater than those of normal gait.
- When compared to standard solid ankle design, the orthoses designed to allow the heel to compress leads to an increase in the relative plantar-flexion at initial contact and a reduction of peak knee flexion moments.
- When compared to the plastic compression heel orthoses, the carbon laminate compression heel orthoses will deflect less and provide an increased peak extension moment in midstance.
The subjects were recruited from the staff of Shriner's Hospitals for Children - Portland by a sample of convenience. The inclusion criteria were that the subjects be healthy without significant gait deviations and be willing to walk in three different types of AFOs. The sample consisted of three adult males ages 24-39 years old and all of the subjects consented to the casting, fitting, and collecting of gait analysis data while walking with each of the three AFO variables.
At the first appointment, the subjects were molded for an AFO using fiberglass wrap. The subjects were molded in a semi-weight-bearing position with their foot flat on the ground and their lower leg perpendicular to the floor. The semi-weight-bearing position was chosen to compress the fat pad of the calcaneus and appropriately contour the hindfoot of the mold for a compression heel design. The molds were filled with plaster and modifications made using standard techniques. All the casting, modifying, and fabricating was done by the investigating orthotists while consistently implementing the same techniques.
At the second appointment, the subject was fit with the solid ankle AFO. Once the subject had an appropriate fit, the exact same model was used to fabricate both the compression heel AFO made of plastic and the compression heel AFO made of carbon laminate. All of the plastic AFOs were made of 3/16" polypropylene. All of the laminated AFOs were made with a consistent layup of materials: one layer of 6" braided carbon, two layers of 4" nylon, and a final layer of 6" braided carbon. Acrylic epoxy and catalyst were weighed and mixed to the manufacturer's specifications for each lamination.
With the exception of the compression heel cut out, all AFO designs had consistent trimlines. Applying the standard solid ankle design, the anterior trimlines were anterior to the malleoli, and 1" inferior to the head of the fibula. The distal medial and distal lateral trimlines extended to the metatarsal heads and the foot plate was full length to retain full distal leverage. All of the AFOs had a 2" strap across the proximal tibia and a 1" strap across the instep.
The design and dimensions for the compression heel cut out were determined by patient anatomy. For the subjects comfort, the material was trimmed superior to the calcaneal tuberosity. The cut out area extended anteriorly to inferior to the distal tip of the lateral malleolus. This location was chosen because it lies inferior to the axis of rotation for the ankle joint in the sagittal plane as well as provides a sufficient lever of material to deflect in weight bearing. Where the two trimlines intersect, (laterally the calcaneal sulcus and medially the sustentaculum tali) an even radius was trimmed to diminish undue concentrations of force. Overall, the cut out is shaped like a dome.
On the third appointment, a physical examination was conducted to ensure that the subjects were within normal limits of range of motion and strength. Then, gait analysis was accomplished using an eight camera motion analysis system with two force plates. Markers were applied to the patient's skin to denote anatomical landmarks that the infrared cameras collect. After walking barefoot to establish the control, the subjects walked wearing each of the three AFO variables on a thirty foot carpet strip. The order that the subjects wore the AFO variables was determined randomly. The force sensors were imbedded in the floor and allow the researcher to collect data as the subject walks normally. In order for the force plate to properly collect and record each step, the researchers select only the steps that the entire foot landed on the force plate. Four steps of each brace were selected and collected to ensure a reasonable average of the steps.
The data from the motion cameras and force plates was graphed to visually assess trends. All of the data was plotted against the percentage of Gait Cycle (GC) to show the average collected values during initial contact through mid-stance. The kinematic data was analyzed to compare motion at the ankle and the graphs indicate the values in degrees. The initial starting orientation of the each orthosis varies slightly. The data reported in the tables has accounted for the differences in initial orientation by measuring only the relative motion from that initial orientation. The kinetic data was analyzed to compare the peak moments at the knee and the graphs indicate that the moments measured in Newton Meter per Kilogram (NM/Kg). Because of the small number of subjects, data can be best compared between AFO variables on the same subject.
Control Comparison of Knee Flexion Moment
The first hypothesis addressed the understanding that a solid ankle design AFO promotes a peak knee flexion moment greater than that of normal gait. The data supports that hypothesis in all three subjects. Subject #1 showed an increase from a barefoot peak knee flexion moment of -0.330 NM / Kg to -0.336 NM / Kg while wearing the solid ankle AFO. Likewise, Subject #2 had a peak knee flexion moment increase from -0.142 NM / Kg barefoot to -0.393 NM / Kg in the solid ankle AFO. Lastly, Subject #3 also increased from -0.143 NM / Kg barefoot to -0.216 NM / Kg in the solid ankle AFO. Although wearing the solid ankle AFO consistently lead to an increase in peak knee flexion moments, the magnitude of increase varied. Comparing the barefoot control to the solid ankle AFO condition, Subject #1 measured a peak knee flexion moment increase ofjust 10%. In the same comparison, Subject #2 and Subject #3 measured substantial increases of 177% and 51% respectively. The considerable range in increased torque measured at the knee could be attributed to placing the markers on the skin while collecting the barefoot data and on the shoes while collecting the solid ankle AFO data. The markers were carefully applied to specific landmarks on the skin in the barefoot control and on the shoe in the AFO conditions. However, when walking, the materials of the shoe compress and flex separately from the skin of the subject. Due to the extreme sensitivity of the infrared cameras, any minor variance in marker placement or movement would affect the motion data and in turn the force data. To minimize this source of variance, tennis shoes could be worn while collecting the control and all AFO conditions. Essentially this would standardize shoe wear and allow a more direct comparison of the affects of the AFO versus the control. As the study was performed, although the results varied in magnitude, the data for all three subjects consistently indicated an increase in peak knee flexion moments while wearing the solid ankle AFO.
The second hypothesis addressed whether the orthoses designed to allow the heel to compress lead to an increase in the relative plantarflexion when compared to a standard solid ankle design. In all three subjects, the degree of plantarflexion at loading was consistently greater in the plastic compression heel orthosis than the solid ankle orthosis. The relative plantarflexion of Subject #1 increased from 1.71 degrees in the solid ankle AFO to 2.34 degrees in the plastic compression heel AFO. Subject #2 measured an increase from 2.71 degrees in the solid ankle AFO to 3.24 degrees in the compression heel AFO. Lastly, Subject #3 also measured an increase from 2.80 degrees in the solid ankle AFO to 3.37 degrees in the compression heel AFO. Although the compression heel AFO condition measured only a slight increase in motion, those increases were 37%, 20% and 20% respectively for each subject. This study did not attempt to decipher between increased motion within the ankle joint and increased motion of the materials of the AFOs. What was measured was from on the outside the AFOs and so the data is reported as relative plantarflexion. In normal gait without an AFO, the human foot usually achieves approximately 10 degrees of plantarflexion during first rocker. 3 The solid ankle AFO reduced relative plantarflexion to less than 3 degrees in each of the three subjects. Although contrary to the name, the solid ankle AFOs did not completely limit relative ankle motion, but rather it was found that even sturdy polypropylene defected under the significant forces of human gait. When applied to the subjects while wearing the compression heel AFO made of plastic, those forces consistently yielded increased relative plantarflexion at loading response.
On the other hand, the findings for the compression heel brace made of carbon laminate were not consistent. Subject #1 measured 1.71 degrees of relative plantarflexion in the solid ankle AFO and 1.76 degrees while wearing the compression heel AFO made of laminate. Subject #2 measured 2.71 degrees of relative plantarflexion in the solid ankle AFO and 2.03 degrees while wearing the laminate compression heel AFO. Lastly, Subject #3 measured 2.80 degrees of relative plantarflexion in the solid ankle AFO and 2.86 degrees while wearing the laminate compression heel AFO. Subject #1 and Subject #2 only measured minor increases in relative plantarflexion while Subject #2 actually measured a 26% reduction in relative plantarflexion. The variance in results of the laminated compression heel AFO cannot be easily explained. During fabrication, researchers were careful to keep the design of the cut out consistent. Even though the dimensions were based on anatomical reference marks, the dimensions were retrospectively measured to be proportional to the length of the footplate in each orthosis. The design and materials used for the laminated orthosis were consistent, but the subject's height and weight varied. A possible explanation for inconsistent findings was that the subjects did not generate enough force to deflect the compression heel made of laminate. Yet even the study's heaviest subject (#2) achieved far less relative plantarflexion with the laminate AFO than in the other two designs. The amount of force applied to the orthosis is unlikely to be the source of variance. If the source of variance is not due to design or the forces imposed by the subject, perhaps the variance stems from minor differences in the subjects' gait. More subjects are required to better determine the cause of variance in the laminate AFO design.
The second portion of the second hypothesis considered whether the compression heel AFOs reduced the knee flexion moments compared to those of solid ankle AFOs. The data supports the hypothesis for the plastic compression heel AFO in all three subjects. Subject #1 was measured to have a peak knee flexion moment of -0.364 NM / Kg in the solid ankle AFO and a moment of -0.188 NM / Kg while wearing the plastic compression heel AFO. Subject #2 was measured to have a peak knee flexion moment of -0.393 NM / Kg in the solid ankle AFO and a moment of -0.353 NM / Kg while wearing the plastic compression heel AFO. Lastly, Subject #3 was measured to have an average peak knee flexion moment of -0.216 NM / Kg in the solid ankle AFO and a moment of -0.180 NM / Kg while wearing the plastic compression heel AFO. While wearing the compression heel AFO made of plastic, the peak knee flexion moments were reduced by 48%, 10%, and 17% respectively. Although the data supported the reduction of knee flexion moments for the plastic compression heel AFO in all three subjects, the data was inconsistent regarding the laminate compression heel AFO. Subject #1 was measured to have a peak knee flexion moment of -0.364 NM / Kg in the solid ankle AFO and a moment of -0.193 NM / Kg while wearing the laminate compression heel AFO. Subject #2 was measured to have a peak knee flexion moment of -0.393 NM / Kg in the solid ankle AFO and a moment of -0.256 NM / Kg while wearing the laminate compression heel AFO. Lastly, Subject #3 was measured to have a peak knee flexion moment of -0.216 NM / Kg in the solid ankle AFO and a moment of -0.283 NM / Kg while wearing the laminate compression heel AFO. As hypothesized, Subject #2 and Subject #3 measured decreases of 47% and 35% in peak knee flexion moments while wearing the laminate compression heel AFO compared to the solid ankle AFO. Contrary to the hypothesis and the results from the other subjects, Subject #3 measured to have an increase of 31% in peak knee flexion moment while wearing the laminate compression heel AFO. The inconsistent results from the laminated brace require further investigation. The results from the plastic compression heel AFO consistently supported the hypothesis that the design did reduce peak knee flexion moments at initial contact.
The third and final hypothesis suggested that the carbon laminate compression heel AFO would have an increased peak knee extension moment in mid-stance compared to the plastic compression heel AFO. The data suggested the opposite of the hypothesis and is consistent in all three subjects.
While wearing the plastic compression heel AFO, the subjects measured 0.989, 0.812, and 0.468 NM / Kg respectively. While wearing the laminate compression heel AFO, the subjects measured 0.921, 0.725, and 0.381 NM / Kg respectively. Thus, contrary to the hypothesis, while wearing the laminate compression heel AFO the subjects were recorded to have reductions in peak knee extension moments of 7%, 11%, and 19% respectively. The hypothesis was based on the properties of the materials. The stiffness of the plastic orthosis is partially determined by cross sectional area of plastic that is able to resist the strain and compression forces applied during gait. The researchers predicted that removing material from the posterior of the plastic compression heel AFO would allow the orthosis to deflect into more plantarflexion at initial contact and into more dorsiflexion in midstance. This prediction was supported by the data, as the plastic compression heel AFO allowed the most relative ankle motion in all three subjects. The researchers chose the laminate material as a comparison because carbon laminates require greater forces to deflect. It was predicted that if an AFO that allowed less dorsiflexion, the knee extension moment would increase. The predicted inverse relationship between deflection into dorsiflexion and knee extension moment was not supported by the data. Indeed, for all three subjects, the laminate material was measured to deflect into the least dorsiflexion of any AFO tested. While wearing the laminate compression heel AFO, the subjects were measured to achieved 31%, 10%, and 59% less dorsiflexion than when the subjects wore the plastic compression heel AFO. Yet the knee extension moments were not found to increase, but rather decrease by 7%, 11%, and 19% respectively. From the study's data at midstance, it would appear that measurements of motion and moment did not have an inverse relationship but rather a direct correlation. When dorsiflexion motion was limited, knee extension moment limited as well. This result could be due to many unknown factors and it is important to keep in mind the small sample size and specific design of brace.
Further analysis and studies should be performed before applying the compression heel design to patients. There are several limitations of the study. Primarily, the small number of healthy adult male subjects (n=3) does not provide the depth or power to make strong statistical assertions. Instead of statistical trends, the results on the AFO variables were limited to intersubject comparisons. Moreover, clinical decisions can not be made from data on a population of only healthy adults. Secondly, only the specific data of sagittal plane motion at the ankle and moment at the knee were analyzed. It is possible that further data regarding other planes of motion or other characteristics in the sagittal plane would clarify results.
The objective of the study was to explore the sagittal plane affects of a compression heel AFO design. After this first and basic study, the findings could be useful. First, although it was expected that solid ankle AFOs increase knee flexion moments to greater than normal walking, that finding should not be regarded as inconsequential. The data reaffirms the understanding that when ankle motion is reduced or limited (as with a solid ankle AFO) the peak flexion moment at the knee increases. The increase in peak flexion torque consequently requires an increase in peak extension torque provided by the extensor muscles. This results in the patient either expending more energy with each step or altering their gait to compensate. Compared to the data for solid ankle AFO design, the plastic compression heel design AFO consistently resulted in increased relative plantarflexion and reduced peak knee flexion moments at loading response. Albeit subjective information, all three of the subjects reported that the compression orthoses allowed a smoother transition from swing to stance and required less effort from their knee extensor muscles. Considering that 60% of the person's body weight is transferred through the limb in a matter of 0.02 seconds 3 , the substantial forces generated can cause instability and discomfort, especially for patients with motor coordination and strength deficits. Researching how to properly balance beneficial motions while reducing excessive forces could lead to increased energy efficiently, comfort, and symmetry of gait. It is the best interest of patients for orthotists and physicians to continue investigating AFO designs to improve patient's gait.
- Observational Gait Analysis. Downey, California U.S.A.: Los Amigos Research and Educational Institute, Inc.; 2001.
- Perry J. Total limb functions. In: Perry J. Gait Analysis Normal and Pathological Function. Thorofare,NJ U.S.A.: SLACK Incorporated; 1992:149-168.
- Perry J.Basic functions. In: Perry J. Gait Analysis Normal and Pathological Function. Thorofare, NJ U.S.A.: SLACK Incorporated; 1992:19-48.
- Lin RS. Ankle-foot orthoses. In: Lusardi MM, Nielsen CC. Orthotics and Prosthetics in Rehabilitation. St. Louis, Missouri U.S.A.: Saunders Elsevier; 2007:219-236.
- Perry J. Ankle and foot gait deviations. In: Perry J. Gait Analysis Normal and Pathological Function. Thorofare, NJ U.S.A.: SLACK Incorporated; 1992:185-222.
- Michael JW. Lower limb orthoses. In: Goldberg B, Hsu JD. Atlas of Orthosis and Assistive Devices. St. Louis, Missouri U.S.A.: Mosby; 1997:209-224.