Secondary Logo

Journal Logo


Permanent Deformation of Posterior Leaf-Spring Ankle-Foot Orthoses: A Comparison of Different Materials

Kerr, Euan BMSc (Hons); Moyes, Kenny BSc (Hons); Arnold, Graham MSc, PhD; Drew, Tim BEng, PhD, CEng, MIPEM

Author Information
JPO Journal of Prosthetics and Orthotics: July 2011 - Volume 23 - Issue 3 - p 144-148
doi: 10.1097/JPO.0b013e3182272941
  • Free

An orthosis is “an externally applied device used to modify the structural or functional characteristics of the neuromusculoskeletal system” (International Organization for Standardization, ISO 8549-1). Orthoses can be used to help manage the loss of limb function caused by multiple orthopedic and neuromuscular conditions such as cerebral palsy, poliomyelitis, or neuropathies. Such conditions can cause a varying degree and pattern of functional impairment including the phenomenon, foot drop. Foot drop occurs due to a weakness or absence of dorsiflexor muscle power. When dorsiflexor power is impaired, plantarflexion moments are not controlled as effectively by their antagonistic actions. Therefore, during gait from heel-strike to foot flat, there is rapid plantarflexion often ending with a characteristic audible foot slap.1 With complete absence of dorsiflexor power, these muscles are unable to control the ankle-foot complex, resulting in the foot hanging inferiorly when lifted off the ground, hence the name foot drop. This leads to the danger of toe contact with the ground during midswing and a consequent tripping hazard.2 A patient can avoid this by altering his or her gait. The gait pattern known as “steppage gait” (deliberate high hip and knee flexion during the swing phase) is a consequence of foot drop.3 It is thought a lesion of the (common) peroneal nerve is, in general, the most likely cause of foot drop.4 A lesion of this nerve will cause the classical clinical picture of paralysis of dorsiflexion and eversion of the foot and of extension of the toes.4

Ankle-foot orthoses (AFOs) are unable to treat the great number of causes of foot drop, but they can help improve the gait of a patient and function of his or her ankle-foot complex. One of the most common designs of AFO used in the management of foot drop is the posterior leaf spring (PLS). PLS AFOs are a type of dynamic AFO prescribed for patients who predominantly require dorsiflexion assistance.5 The PLS AFO is designed to resist plantarflexion, thus constraining the foot-drop motion.6 The magnitude of the resistance to plantarflexion directly affects the amount of toe clearance in midswing.7 PLS AFOs not only help avoid toe contact with the ground by holding the foot in a neutral position during swing phase, but they also control plantarflexion rate of the foot between heel-strike and foot flat, thus eliminating foot slap.

Various types of materials can be used in the manufacture of orthoses. Thermoplastics such as polypropylene and polyethylene are more commonly used as they are light, manageable, cosmetically acceptable, and can achieve an accurate fit to provide better control. However, more modern materials such as carbon fiber are now becoming popular in orthotics. Carbon fiber is even lighter, has a greater strength-to-weight ratio, and is much stiffer in comparison to the thermoplastics. Carbon fiber also has energy-storing capabilities that reduces energy expenditure in movements such as gait.8 All materials used are molded and trimmed into specific designs, depending on the type of orthosis. PLS AFOs consist of a calf cuff that tapers to a narrow band behind the ankle (the “leaf”) and widens back out to capture the heel and extend to the tip of the toes.8 The flexibility of the AFO must match the degree of paralysis of the patient it is prescribed for.9 A variable amount of change in stiffness can be made by altering the trim line10 as a PLS AFO's resistance to dorsi- and plantarflexion decreases in proportion to the reduction of the posterior upright width.5 Degradation in AFO stiffness can undermine control of the impaired lower limb.11 Failure of the leaf section of PLS AFOs is common.12 Because of their design, PLS AFOs have reduced material in the leaf section and so are more susceptible to failure than other designs. The narrower the leaf, the higher the stress, thus increasing the risk of failure.1

Permanent deformation of PLS AFOs have been increasingly observed at Ninewells Hospital, Dundee. This local experience suggests it is a relatively common occurrence among foot-drop patients, but to date there is a lack of evidence to suggest which material used in the manufacture of these orthoses is most resistant to permanent deformation. This study aims to determine whether carbon fiber, polypropylene, or polyethylene has the best deformation resistance.


Six PLS AFOs were manufactured. Two of each material were used to account for differences in the properties of each AFO due to the manufacturing process. All AFOs were manufactured by the same orthotic technician in the same department as though they were being prescribed to a foot-drop patient. Both polypropylene AFOs were manufactured from the same 4.7 mm sheet and both polyethylene AFOs from the same 5 mm sheet. Both carbon fiber AFOs used the same “two cross weave-eight unidirectional-two cross weave” sheet lay-up. All AFOs produced were a PLS design from the same positive cast. For the AFOs to be tested on the machines used, modifications were made to their structures (Figure 1). Trimlines for each AFO were altered to find a balance between standardizing properties without detracting from PLS AFO structure that would normally be used in the clinical setting. A Zwick/Roell Z050 universal testing machine (UTM) was used to produce force/displacement graphs before and after cyclical loading. A custom-built, pneumatically driven cyclical loading machine (CLM) was used to cyclically load each AFO by plantarflexing them from 0° to 10° plantarflexion (Figure 2). An electromagnetic counter was used to count the number of cycles during testing.

Figure 1.:
Modified posterior leaf-spring ankle-foot orthoses.
Figure 2.:
Custom-made cyclical loading machine.

The method used was an original protocol as there are no definitive methods for measuring orthotic load/deflection characteristics.13 Each AFO was subjected to the same testing method. First, the AFO was aligned on a plate which was fixed to the CLM. It was positioned so that the pneumatic piston flexed the AFO from 0° to 10° plantarflexion. These angles were verified with a laser attachment and angular scale (Figure 2). The plate and attached AFO were then placed on the UTM, and its mobile platform position was altered so that its bracket was touching the AFO footplate with no slack present. As soon as the UTM moved, it began to plantarflex the AFO to 10°, and a force/displacement graph was produced. This was repeated five times and average values taken. The AFO and plate were subsequently reattached to the CLM, and the programmed number of cycles was carried out at a rate of 1.5 Hz. Each AFO underwent the following number of cycles: 30,000, 90,000, 150,000, and 210,000. This equates to a total of 480,000 cycles over four tests. Once the loading had finished, the AFO was immediately transferred back to the UTM, the platform positioned at the exact place as before, regardless of whether it was touching the AFO or not, and another force/displacement graph produced. Again this was carried out five times to produce an average graph. Once these “before” and “after” graphs had been produced for each AFO for each number of cycles, the data were normalized using Microsoft Visual Basic 2008 Express Edition in order for the graphs to be analyzed and compared.


None of the pairs of AFOs produced similar force/displacement graphs before testing began. This was despite each pair being made from the same sheet of thermoplastic or having the same carbon fiber lay-up, and all AFOs being manufactured by the same technician in the same department. The main variability seen from initial force/displacement graphs produced showed that for all three types of materials, neither of the pairs of AFOs had similar stiffness. These differences in stiffness were shown as discrepancies in the gradients of each AFOs graph in relation to that of the other AFO of the same material. The most dissimilar result produced before testing began was from the two polyethylene AFOs as there appeared to be no real resemblance between the two graphs. The pairs of polypropylene and carbon fiber AFOs, however, showed slightly more association in comparison. The second carbon fiber AFO was stiffer throughout all degrees of plantarflexion with a resultant 2.1 N more being applied at 10° plantarflexion. Both polypropylene AFO graphs were relatively similar at initial plantarflexion although by 10°, the second AFO was also stiffer than the first, with an additional 1.4 N being applied.

All six PLS AFOs were tested for four different amounts of cyclical loading. After 30,000 cycles (∼6 days of continuous walking), there were no signs of deformation from any of the materials. However, after the second test of 90,000 cycles, the polypropylene and polyethylene PLS AFOs began to show signs of permanent deformation. Their stiffness began to drop and the force/displacement graphs began to indicate the occurrence of permanent deformation to a plantarflexed position. As the number of loading cycles increased over the next two tests, these signs of permanent deformation became more apparent for both thermoplastics, while the only sign of carbon fiber failure was a drop in stiffness, although much smaller than that experienced by the thermoplastics. Carbon fiber showed no signs of permanent deformation and remained in its manufactured neutral position throughout this study.

The following graphs (Figures 3–5) show overall examples of the changes observed for each material. The graphical data are the mean of each pair of AFOs tested. Each material's “before loading” and “after loading” graphs from every test were combined to produce an overall average “before” and “after” graph for that material. The “after” graphs in Figures 3 to 5 represent the total of 480,000 cycles completed in this study. The initial curves seen in the “polypropylene AFTER” graph and “polyethylene AFTER” graph indicate the presence of permanent deformation to a plantarflexed position. The actual angular deformation measured for the polypropylene and polyethylene PLS AFOs following testing was a mean of 8.6° and 9.3° plantarflexion, respectively. The drop in the final points of the “after” graphs when compared with the “before” graphs represent the loss of stiffness.

Figure 3.:
Overall carbon fiber force/displacement graph.
Figure 4.:
Overall polypropylene force/displacement graph.
Figure 5.:
Overall polyethylene force/displacement graph.


This study demonstrates the difference in the structural properties of the AFOs due to the manufacturing process. It has always been appreciated that the resultant structural properties of orthoses are highly dependent not only on the materials used but also on the manufacturing process itself.14 This is a factor of orthosis production which needs to be carefully controlled to have a greater certainty of the exact properties of the orthosis, resulting in a more accurate manufacture and prescription for each patient's individual requirements. Despite using the same technician, workshop equipment and positive cast to manufacture the AFOs used in this study, there were still differences in each of their properties. To better control this area of variability in carbon fiber AFO manufacture, layering up the carbon fiber sheets should be done with extreme care and attention to detail by specially trained health care staff to avoid areas of nonbonding or void pockets between layers. The curing process is a standardized method to minimize the formation of gas pockets between layers; however, this is difficult to control and may still occur despite best efforts. System fluctuations in vacuum pressure and delivery used in the curing process is another area of variability that is difficult to control in carbon fiber AFO manufacture. The main area of variability in the manufacture of thermoplastic AFOs is when the softened sheet is placed over a positive cast and stretching of the thermoplastic occurs around the heel of the cast as the flat sheet is molded around the curved heel. This is very difficult to avoid; however, once the AFO has been initially produced, it is reasonable to suggest that its properties could be quantified using machinery such as the UTM from this study, and the trimlines of the AFO subsequently altered as required to have better control over the resultant AFO properties.

The findings produced from the testing of both thermoplastic materials explain the observations made locally of PLS AFO failure. The thermoplastic AFOs tested were losing stiffness and also permanently deforming to a plantarflexed position, the same failings observed with PLS AFOs of the foot-drop patients. This study found these changes in all of the polypropylene and polyethylene PLS AFOs tested. Although these changes did not occur until after each AFO had undergone 30,000 and then 90,000 cycles of plantarflexion, their changes indicate a failure in their prescribed functions. This evidence of deformation indicates that the affected thermoplastic AFO is unable to support the foot and keep it in a neutral position. This causes an inability in allowing toe clearance during swing phase. Initial heel contact at the start of stance phase or avoidance of foot slap by supporting the foot and letting it gradually make full contact with the ground are also lost functions due to the thermoplastic AFO deformation to a plantarflexed position. This more plantarflexed ankle position will also affect other lower limb joint positions and the biomechanics of the patient's gait.

It has been commented that polypropylene can endure several million cycles of repetitive flexes.15 This may seem unlikely due to this study's evidence of polypropylene AFO clinical failure after only a total of 480,000 cycles. However, the literature did not state whether polypropylene could endure these several million cycles without any changes to its properties, i.e., no clinical failure, or if it could only endure these cycles without catastrophic failure; i.e., deformation but no fracture of the material. It was also unknown what structural shape the material was in when tested. Another comment that polyethylene has a long fatigue life, again contradictory to the evidence of this study, was found in the same literature.15 Results from one study found that cyclical loading will degrade stiffness and soften a plastic AFO.11 This statement supports the findings of this study as degradation can be shown in both thermoplastic's force/displacement graphs as a drop in stiffness and the softening described in the other study can be shown in this one as evidence of deformation.

In comparison with both thermoplastics tested, carbon fiber was relatively unaffected by testing and can generally be said to have far more resilience to permanent deformation than polypropylene or polyethylene. Evidence shows that neither of the carbon fiber PLS AFOs tested deformed to a plantarflexed position; both remaining in their manufactured neutral position. However, this superior performance and resistance comes at a price. Carbon fiber orthoses cost considerably more than thermoplastic equivalents. They require labor-intensive production methods and are made from expensive raw materials. The financial implications of carbon fiber must be taken into account when considering its use for PLS AFO manufacture over that of polypropylene or polyethylene. Questioning its value for money and whether in the long run it would be cheaper to fit a foot-drop patient with one carbon fiber AFO, rather than several replacement thermoplastic AFOs, is essential. From this study, the clear evidence of carbon fiber's superiority in relation to deformation resistance would suggest that, for the right patient and lifestyle, carbon fiber is worth the extra initial cost.

The evidence and discussion produced from this study can support the prescription of carbon fiber PLS AFOs to foot-drop patients who are more active and more likely to load their AFO frequently, possibly for extended periods of time. The properties of carbon fiber would enable the AFO to cope with these higher demands compared with one made from a thermoplastic, thus reducing the risk of permanent deformation similar to that observed. Thermoplastics are not yet obsolete and should be reserved for the manufacture of PLS AFOs for less active foot-drop patients who will load their AFO less frequently and for lesser time periods. They would also be optimal for children suffering from foot drop as these younger patients weigh less than adults, thus reducing the forces through the AFO. They are also likely to outgrow their close-fitting AFOs regularly, thus requiring frequent replacements with the cheaper material.

This study has made several simplifying assumptions; consequently, the testing method may not accurately replicate real-world conditions. As a result, this may limit the reliability and significance of the findings. Plantarflexion to only 10°, absence of dorsiflexion, no periods of rest during cyclical loading, and uncontrolled test environment and AFO temperature may all alter results if any aspect of these limitations were to be changed. One major improvement to this study or similar future research would be to have a more realistic testing environment. If the CLM and UTM could load and test the PLS AFOs without the need to modify their structures by removing the calf cuff and adding slits and holes, more accurate and reliable results may have been obtained. This may have been possible by fitting the AFOs to a dynamic prosthetic leg during testing. From what was stated in the methods and materials section and shown in Figure 1, the differing trimlines of each AFO are likely to have had some impact on the results found in this study. It was attempted to initially standardize the properties of each AFO by altering their trimlines. However, this action was restrained as not to alter the orthoses so much that they did not resemble the PLS AFO structures used in the clinical setting. The thicker trimlines of the thermoplastics were kept as this area gives the structural strength needed to keep the patient's foot in a neutral position during swing phase of gait. The carbon fiber AFO trimlines were noticeably narrower as their strength comes from the lay-up of carbon fiber sheets. A similar trimline to the thermoplastics would not be seen in the clinical setting of carbon fiber PLS AFO prescription as during dorsiflexion, this wider trimline would undergo compression. As carbon fiber is weakest in compression, this situation is avoided by having the narrower trimline, as shown in Figure 1.

For further AFO studies, it would be highly recommended to have a larger number of identical orthoses to test to minimize the significance of structural property differences due to manufacture.


This study aimed to compare the properties of polypropylene and polyethylene AFOs to that of carbon fiber AFOs under cyclical loading of plantarflexion because observations of an increasing number of PLS AFOs failing due to permanent deformation changes have been made. In addition, there is a lack of evidence to suggest which material used in the manufacture of PLS AFOs has the optimum resistance to permanent deformation.

Evidence from this study indicated that after a total of 120,000 cycles, both thermoplastics tested began to show signs of failure: reduced stiffness and permanent deformation to a plantarflexed position. As cyclical loading of the AFOs increased, these signs of failure increased significantly. Carbon fiber remained comparatively unaffected by the same amount of cyclical loading. We conclude that when used for PLS AFOs in the management of foot-drop patients and cyclically loaded to 10° plantarflexion, carbon fiber has far superior deformation resistance properties than that of polypropylene or polyethylene.

Basing future PLS AFO prescription on the evidence found in this study and the suggestions previously discussed, it is hoped this would result in better use of resources, increased efficiency and patient specificity in PLS AFO prescription for foot drop management, and most importantly, improved patient care.


The authors wish to acknowledge their gratitude to all other staff members involved in the production of this study.


1. Bowker P, Condie DN, Bader DL, Pratt DJ. Biomechanical Basis of Orthotic Management. 1st ed. Oxford: Butterworth-Heinemann; 1993.
2. McHugh B. Analysis of body-device interface forces in the sagittal plane for patients wearing ankle-foot orthoses. Prosthet Orthot Int 1999;23:75–81.
3. Bishop D, Moore A, Chandrashekar N. A new ankle foot orthosis for running. Prosthet Orthot Int 2009;33:192–197.
4. Stewart JD. Foot drop: where, why and what to do? Pract Neurol 2008;8:158–169.
5. Sumiya T, Suzuki Y, Kasahara T. Stiffness control in posterior-type plastic ankle-foot orthoses: effect of trimline. Part 2: orthosis characteristics and orthosis/patient matching. Prosthet Orthot Int 1996;20:132–137.
6. Chu TT. Biomechanics of ankle-foot orthoses: past, present, and future. Top Stroke Rehabil 2001;7:19–28.
7. DeToro WW. Plantarflexion resistance of selected ankle-foot orthoses: a pilot study of commonly prescribed prefabricated and custom-molded alternatives. J Prosthet Orthot 2001;13:39–44.
8. Novacheck TF, Beattie C, Rozumalski A, et al. Quantifying the spring-like properties of ankle-foot orthoses (AFOs). J Prosthet Orthot 2007;19:98–103.
9. Yamamoto S, Ebina M, Iwasaki M, et al. Comparative study of mechanical characteristics of plastic AFOs. J Prosthet Orthot 1993;5:59–64.
10. Singerman R, Hoy DJ, Mansour JM. Design changes in ankle-foot orthosis intended to alter stiffness also alter orthosis kinematics. J Prosthet Orthot 1999;11:48–56.
11. Lunsford TR, Ramm T, Miller JA. Viscoelastic properties of plastic pediatric AFOs. J Prosthet Orthot 1994;6:3–9.
12. Chu T, Feng R. Determination of stress distribution in various ankle-foot orthoses: experimental stress analysis. J Prosthet Orthot 1998;10:11–16.
13. Sumiya T, Suzuki Y, Kasahara T. Stiffness control in posterior-type plastic ankle-foot orthoses: effect of trimline. Part 1: a device for measuring ankle moment. Prosthet Orthot Int 1996;20:129–131.
14. Condie DN. The modern era of orthotics. Prosthet Orthot Int 2008;32:313–323.
15. Lusardi MM, Nielsen CC. Orthotics and Prosthetics in Rehabilitation. 1st ed. Oxford: Butterworth-Heinemann; 2000.

permanent deformation; ankle-foot orthosis; carbon fiber; thermoplastic

© 2011 American Academy of Orthotists & Prosthetists