Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Determination of Stress Distribution In Various Ankle-Foot Orthoses: Experimental Stress Analysis

Tai-Ming Chu, PhD
Rong Feng, MS

ABSTRACT

Experimental stress analysis on various ankle-foot orthoses (AFOs) was conducted. A complete experimental testing system has been developed successfully. Results revealed that the peak stress in the orthoses occurred in the neck region. This was consistent with earlier Finite Element Analysis (FEA) (1, 2) and clinical observation. Thus, failure in the neck region was common. Based on the present investigation, following suggestions has been made. AFO should be made asymmetrically and the lateral side of the neck region should be wider and/or thicker than the medial side.

Key Words: Stress Analysis; Strain Gage; Average Peak Stress (APS); Finite Element Analysis (FEA); Wheatstone Bridge.

Introduction

An ankle-foot orthosis (AFO) is a rehabilitation device designed to support, align and improve the functions of the ankle and the foot (3). In the orthotic industry, polypropylene is used as a major orthotic plastic due to its high-fatigue resistance, light weight, and excellent molding characteristics (3). However, earlier failure and improper design of an AFO bring inconvenient and discomfort to many patients. Therefore, the primary objective of this study was to understand the mechanical performance of AFOs under loading and the resulting deformation developed while in use.

Five different custom-made AFOs were tested (see Figure 1) . All were made by Kessler Institute of Rehabilitation Inc. in Livingstone, N.J. The material was Polypropylene, a Colyene? Co-polymer plus Fleshtones and Colors. Five AFOs were the Flex, the Standard, the Moderate, the Solid and the Varus AFOs. The Flex orthosis, usually used for minor injured patients, is very flexible but provides minimum support. The Standard orthosis, the most popular one, provides limited support. The Moderate orthosis has a wider width in the neck region than the Standard AFO, which limits dorsiflexion/planterflexion but provides more stability. The Solid orthosis has a wider neck than the Moderate orthosis. It not only prohibits dorsiflexion/planterflexion but also provides much more necking support. The Varus orthosis has the widest neck width and additional support part in the lateral upper-neck region to prohibit dorsiflexion/planterflexion and inversion/eversion. It provides the maximum support and is generally used for severe injured patients. The anthropometric data of the five orthoses are measured and listed in Table A .

Studies showed only few investigations on such topics have been conducted. A three-dimensional Finite Element model was developed by Chu, et al. (1-2) for a computational stress analysis on the AFO. Although the three-dimensional model provided useful information on static analysis, limited dynamic results were obtained. A comparative study of mechanical characteristics of plastic AFOs was made by S. Yamamoto, et al. (4) in 1993. Eleven AFOs were measured using a muscle-training machine. However, only dorsiflexion/plantarflexion and inversion/eversion were tested.

In the present study, it was intended that through the experimental analysis, the design parameters of AFOs can be improved. The entire experimental testing procedures involve 1) mounting the strain gages at desired location, 2) instrumentation development that simultaneously record strains at various locations, and 3) parameter analysis for the prediction of AFO failure.

Experimental Methodology

Method of measurement of stress was the strain gage technology. The strain gages convert a change in dimension (elongation or compression) into ohmic resistance, which can be accurately measured by a Wheatstone bridge circuit to determine the exact magnitude of mechanical strain. By applying the Hookie's law, the positive strain output represents the tensile stress and the negative strain output represents the compressive stress (see Figure 2) . The experimental study began with strain gages attachment to AFOs. The strain gage used is biaxial rosette pattern strain gage (EA-Series) manufactured by Measurements Group Inc. It can measure a vertical and a horizontal deformation simultaneously. The strain gages were installed using standard bounding procedures. However, a UV light was used as part of the surface preparation (5). The AFO was placed under UV light approximate one inch for around 30 minutes. However, it depends on the type of UV light and its intensity.The hardware system developed included Wheatstone bridge circuit, EXP-16 amplifier board, DAS-800 data acquisition board and the key component PC. The EXP-16 and DAS-800 boards were manufactured by Keithley Instruments, Inc. By using VTX software (product of Keithley Metrabyte) with Visual Basic, the data were displayed, analyzed, graphed and stored to spreadsheet in Microsoft Excel. For each AFO, a total of eight strain gages were installed to the specific locations along the lateral side, the medial side, and the middle of the lower neck of the orthoses (see Figure 3) . Each strain gage records the dynamic strain change during motions. The Average Peak Stress (APS) was then obtained using the Hookie's law. The APS distribution contour among the eight tested locations provides information for prediction of the failure (6).

In the present study, the neck of the AFO referred to the narrowest place of the leg portion of the orthosis, where gage 2 and gage 7 were located. Upper-neck and lower-neck referred to 2 to 3 cm above and below the neck. Gage 1 and gage 6 were located at the upper-neck. Gage 3 and gage 8 were located at the lower neck. Gage 4 and gage 5 were located at the middle of the back of the orthosis from lower-neck to upper-heel region (see Figure 3) .

A total 10 motions of the stance phase and the swing phase were implemented as testing conditions for the experiment. Each motion was tested three times. The resultant stress was calculated as the average of the three trials from the vertical and horizontal components of the strain gage. The loading force on the AFO was approximately 170 lbs.

Four walking motions (slow forward walk, fast forward walk, running and backward walk) and three in position motions (jumping, standing up/sitting down and lifting object) were selected for the stance phase. Three coupled movements were selected for the swing phase. They were Dorsiflexion/Planterflexion, Inversion/Eversion, and Abduction/Adduction. These motions are common daily activities usually occurring with the elderly, the disabled who need support and a patient who is in or after the recovering process.

Results

The stresses measured in the present study were either tensile or compressive stresses due to bending during motions. Motion of slow forward walk is the most common human movement in nature. Results from this motion were used as the base for comparison with other motions of the stance phase. Both heel-strike and toe-off produced high tensile and compressive stresses, but at different locations. The maximum stress generated during heel-strike was compressive stress (see Table B) . The maximum stress generated during toe-off was tensile stress (Table B). Specifically, the maximum tensile stresses of the Varus AFO and the Standard AFO were 0.35MPa and 0.84MPa, respectively. The stresses were located at the upper neck of the lateral side (see Figure 4 and Table B ). The maximum compressive stresses (0.5MPa and 0.6MPa, respectively) were located at the lower neck (see Figure 5 and Table B ). For the Moderate and the Solid AFOs, the maximum tensile (0.74MPa and 0.39MPa, respectively) and compressive (0.55MPa and 0.49MPa, respectively) stresses were found at the upper-neck of both lateral and medial side (see Figure 4 , Figure 5 and Table B ). For the Flex orthosis, the tensile APS (0.9MPa) was highly concentrated at G2, G6 and G3 (see Figure 4) . For the Standard orthosis, stress was concentrated at the neck and upper-neck region of the lateral side. It clearly showed a shear failure. This phenomena was also observed during clinical study for the Flex orthosis. It confirmed that this is the easiest point to break.

Standing up/sitting down is a common activity for everyone, especially for patients in the recovery process. It is the earliest and yet the hardest motion for patients. However, there was no obvious difference in the stress distribution between each orthosis in standing up/sitting down. The magnitudes of the tensil APS (Varus: 0.7 MPa, Solid: 0.8 MPa, Moderate: 0.84 MPa, Standard: 0.5 MPa, and Flex: 0.62 MPa) indicated that the patient exerted more force on the wider-necked orthosis in order to stand up.

Lifting objects (i.e., 25 lbs) did not significantly change the stress distribution in three AFOs. Varus, Solid and Moderate orthoses had the similar results. The tensile APS was concentrated at upper-neck region and the magnitudes were approximately 0.2 MPa. However, for the Standard orthosis, the tensile APS, which was located at G1 and G2 (0.4 MPa), increased by 100 percent. For the Flex orthosis, the stress (0.75 MPa) was increased by 275 percent compared with the first three orthoses. While the flexibility of the orthosis was enhanced, the tensile peak stress increased sharply.

Dorsiflexion/planterflexion is the motion of the ankle joint. It is the most important motion in the swing phase in order to determine the flexibility of different types of AFOs (4). It is also a common activity for people who drive. The tensile and compressive APS profiles were clearly pointed out that the magnitude of the maximum peak stress of different orthoses was rely on the width of their necks. The magnitude of compressive stress (at the neck region) increased when the width of the AFO's neck decreased (see Figure 6) . Although the width of the Varus and the Solid orthoses were approximately the same at neck region, the support part on the lateral side of the Varus orthosis caused the stress to decrease. Except the Varus orthosis, these two parameters (width of the neck and the compressive APS) had linear relationship (see Figure 6) . The narrower the neck, the higher the stress, thus, the easier to fail. Therefore, the narrow-necked orthosis is not suitable for a person who needs to drive.

Inversion/eversion is the motion of the subtalar joint. Abnormality usually occurs to the patient with spasticity, an involuntary movement. Simulation of an normal inversion/eversion generated a tensile peak stresses at the neck (medial) region of the Solid and the Moderate orthoses, as well as in the neck (lateral) region of the Standard and the Flex orthoses. The tensile peak stress distribution profile implied that the medial side of the Solid and the Moderate AFOs and the lateral side of the Standard and the Flex AFOs were more susceptible than the opposite side during this motion. However, the above movement had no significant influence on the Varus orthosis. It demonstrated that inversion/eversion was prohibited by the additional support material added to the lateral side of the Varus AFO. Thus, the Varus AFO should be used by patient who has a sever muscle contract problem. Abduction/adduction is the motion of the ankle and the subtalar joints. Abnormal heel strike is due to toe drag and foot abduction.

Abduction and adduction produced a distinct tensile APS distribution contour in all tested orthoses except the Varus AFO (see Figure 7) . This confirmed the earlier result (1,2). The maximum tensile stress for the Varus AFO was located at the neck of the lateral side (see Figure 7) . For the rest of orthoses, the tensile APS was located at the lateral side arc edge and medial side upper-neck edge (Figure 7) . It was noted that abduction/adduction had a major effect on the lateral side arc edge of the orthoses. The AFO prescribed to a patient who has a server toe drag condition should have reinforcement at the lateral side arc edge.

Discussion

The magnitude and the location of the stress concentration determined from this experiment clearly indicated the average peak stress was dependent on the geometry of the AFOs. More important, it depended on different patients and their activities. Activities such as slow walk, fast walk, running, jumping, all significantly altered the magnitude and distribution of the peak stress.

It was obvious that the APS increased with an increase in the flexibility of the orthosis (see Figure 6) . In addition, each type of the orthosis had its own distinct stress distribution contour (Figure 4 , Figure 5 , and Table B ). The peak stress concentration was primarily located at the lateral side neck region. However, the stress concentration shifted from the upper-neck to the lower-neck when the width of the neck decreases.

Compared with the slow forward walk, fast walk did not significantly affect the stress distribution on the Varus and the Flex orthoses. However, it significantly affected the other three AFOs. The peak stress concentration of the Standard, the Moderate and the Solid AFOs shifted from the neck of both sides in slow walk to the lower-neck lateral side in fast walk (see Figure 8) . This was due to an increasing in walking velocity. Thus, the magnitude of the loading forces was increased. In addition, this was due to the slightly shift of the ground contact point as well.

Compared with walking, results from running showed a difference in tensile APS distribution (see Figure 8) . It demonstrated running had intensive effect on the middle of the lower-neck region of the orthosis in addition to the upper neck region (Figure 8c) . Running also had intensive effect on the middle of the leg portion of the orthosis during heel strike.

Since backward walk was the reverse movement of forward walk, the APS concentration was almost the same as that of slow forward walk indicated in Figure 4 and Figure 5 . In addition, results from the fast walk simulation were similar to the slow walk in terms of location of stress concentration. The magnitude of tensile and compressive stresses was higher. This was due to the fact the measurement cable was not long enough for the subject to take more than three steps for each motion.

Although in position jumping was not a common motion for the patients. It may occur while doing exercises in recovering process. All five orthoses demonstrated this motion had a significant effect on the lower-neck and heel region. It was observed that the compressive peak stresses shifted from the neck (in slow walk) to the lower-neck and heel region of all orthoses. It was also observed that a shear failure of the AFO would occur after jumping just about several times. Thus, the excessive and intensive loads, such as jumping, had a significant effect on the orthosis failure.

Standing up/sitting down, unlike walking, jumping and/or other motions, the stress distribution at the neck of the orthosis was not significantly influenced by this movement. Moreover, the magnitude of the peak stress also depended on how fast the person standing up/sitting down. The faster of standing up/sitting down, the higher the stresses exerted on the neck region. Thus, the wider neck would provide a better balance for the patient. Therefore, the wider-necked orthosis should be prescribed to the earlier or slow-recovered patients.

Simulation of motion of lifting objects was similar to the motion of stand up/sitting down with the exception of additional weight need to be lifted. Results indicated this motion did not significantly change the stress distribution in three AFOs, namely, the Varus, the Solid and the Moderate orthoses. However, for patients who have to lift material during rehabilitation exercises, this will alter the stress distribution in the Flex and the Standard AFOs. If increased, the lifting weight would increase the stress concentration in AFOs.

Parameter analysis revealed for some motions, such as running and jumping, the tensile peak stresses in the neck region on the lateral side of the Flex orthosis were above material yield strength. The plastic deformation occurred and an unrecoverable strain was produced. No compressive peak stresses exceeded the yield strength during simulation. There were two primary factors caused material to fail: high-stress/low-cycle failure that happened with the Flex orthosis and low-stress/high-cycle failure that happened with most other orthoses, especially the Varus orthosis. The magnitude of peak stress from high to low for most movements was in the sequence of the Flex, Standard, Moderate, Solid and Varus orthosis. It further confirmed that the more flexible, the higher the stress will be generated during loading and motion.

Overall analytical results from the dynamic analysis confirmed the former FEA results (1,2) that the maximum peak stress occurred in the neighborhood of the neck region of the AFOs and the stress distributions were asymmetric. Results from the study suggested using the asymmetric orthosis with a wider or thicker neck and, specially, at the lateral side would provide the same flexibility and stability but reduce stress concentration. For further analysis, an energy consumption parameter could be used to indicate the support ability of different AFOs. Like the peak stress, the amount of energy absorbed by the orthosis might cause its shear failure in certain regions.

Conclusion

In the present investigation, extensive experimental analysis was performed on five types of AFOs with 10 common daily activities related to patient recovery process. The peak stress concentration in the orthoses varied significantly with a change of the orthosis geometry. However, each type of orthosis had its own distinct stress distribution contour during motion. These stress information can be used to predicted the fatigue life or failure point of the orthosis (6).

Specific findings are 1) for the Varus orthosis, the peak stress was concentrated at the neck of the lateral side; 2) for the Solid and the Moderated orthoses, the peak stress was concentrated at the upper-neck area of both side; 3) for the Standard orthosis, it was located at the neck and the upper-neck region of the lateral side; and 4) for the Flex orthosis, the peak stress was located at the neck and the lower-neck region of the lateral side. In addition, results clearly showed a shear failure, the same as most failed Flex orthosis in clinical observations 5) and FEA (1,2).

Acknowledgements

The authors wish to express their sincere gratitude to Jack Hodgins, director of prosthetic/orthotic department, and Gus Eppinger, orthotist, of Kessler Institute of Rehabilitation for their endless resources and support. This project is supported by the research funding grant number 175 from the Henry H. Kessler Foundation.

References:

1. Chu T, Reddy NP, Padovan J. Three-dimensional finite element stress analysis of the polypropylene, ankle-foot orthosis: Static analysis. Med. Eng. Phys. 1995;17:5:372-9.
2. Chu T, Reddy NP. Stress distribution in the ankle-foot orthosis used to correct pathological gait. J Rehab Rsc and Devel 1995;32:4: 349-60.
3. Wu KK. Foot orthoses principles and clinical applications. Baltimore, MD: Williams & Wilkins, 1990.
4. Yamamoto S, Ebina M, I#wasaki M, Kubo S, Kawai H, Hayashi T. Comparative study of mechanical characteristics of plastic AFOs. JPO 1993;5:3:59-64.
5. Chu T, Gent A. Bonding methods of strain gages to the polypropylene AFO. Experimental Techniques 1996;20:5.
6. Collins JA. Failure of materials in mechanical design. New York: John Wiley & Sons, 1981.
7. Bunch W, Keagy R. Principles of orthotic treatment. St. Louis: C.V. Mosby, 1976.

Website built by oandp.com