Quantitative research from The Netherlands suggests that for every ankle foot orthosis, there is an optimal stiffness associated with the lowest energy cost of walking for a given set of gait-related impairments. Achieving this optimal device stiffness in practice, however, may require clinicians to rethink conventional approaches to AFO prescription.
By Daan J.J. Bregman, PhD
Stroke, multiple sclerosis, and partial spinal cord injury are central neurological disorders with a high prevalence in Western society. Many patients suffering from these central neurological disorders have a reduced walking ability, which has a high impact on daily living. To restore walking ability, ankle foot orthoses (AFOs) are prescribed frequently. It is not known, however, what mechanical properties an AFO should hold to promote optimal walking ability.
My colleagues and I have proposed a new method to quantify mechanical AFO properties that enables evaluation of the AFO’s mechanical functioning during gait. We have performed such an evaluation for spring-like AFOs and compliant AFOs, and have identified clear links between mechanical AFO functioning and the benefits for the patient in terms of energy cost and walking speed, or activity level. Subsequently, with model simulations and physical experiments, we have demonstrated that with variations in the mechanical AFO properties, the mechanical functioning and the benefits of the AFO at activity level can be optimized.
Quantifying AFO mechanical properties
We have shown that the BRUCE (Bi-articular Reciprocating Universal Compliance Estimator) device can measure the mechanical properties of an AFO (i.e., the stiffness and neutral angle around the ankle joint and forefoot joint), in a reliable yet clinically feasible manner (Figure 1).1 Operation of the device requires no special skills or knowledge, making the BRUCE a useful tool for research purposes as well as for everyday use in clinical practice. Several studies have attempted to quantify mechanical AFO properties.2-10 However, apart from the designs by Cappa et al3,4 and Novacheck et al,5 the reliability of these quantifications are unknown.
The BRUCE device was the first to characterize both the stiffness and neutral angle of AFOs. The neutral angle of an AFO gives a good indication of its positioning independently of the patient for whom it has been prescribed. However, the effect of certain patient characteristics (e.g., calf circumference) on the same AFO neutral angle may result in a slightly different AFO alignment (i.e., its global orientation when worn by the patient). Recent literature suggests that AFO alignment makes an important contribution to potential benefits conferred by the device.11-13 Specifically, it has been suggested that the shank-to-vertical angle should be tuned to individual patient deficits to ensure appropriate AFO function and avoid excessive moments at the knee. Clinicians can obtain patients’ shank-to-vertical angle from the orientation of an individual’s lower leg in the AFO (when no forces are applied on the AFO) and shoe inclination. To measure the shoe inclination, fixate the shoe in the BRUCE and acquire the ankle angle with the dummy shank fixed in a vertical position.
The BRUCE was the first device developed to measure AFO and shoe properties around the forefoot joint. Recent literature indicates that AFOs with full-length forefoot plates produce higher moments at the ankle than those with three quarter-length forefoot plates.14 Moreover, it’s known that shoes worn in combination with an AFO affect kinematics and kinetics during walking differently than barefoot walking.15,16 We have not previously explored the influence of the AFO’s forefoot properties on gait (with or without a shoe).
The AFO properties measured with the BRUCE device have varied in our research. In a study of seven patients with drop foot secondary to stroke or multiple sclerosis, average AFO stiffness was approximately 0.2 Nm/deg.17 In a study of seven patients with drop foot secondary to stroke or multiple sclerosis, average AFO stiffness was 6.36 Nm/deg (including one outlier with a stiffness of 23.5 Nm/deg).18 Stiffnesses reported in the literature are in line with these numbers, with values ranging from 0.07 Nm/deg for articulated AFOs10 to up to 4.60 Nm/deg for spring-like carbon AFOs.19 Consistent with our work, studies have shown that identical mechanical AFO properties can be achieved using different materials and designs.5 This emphasizes that reporting on the mechanical AFO properties, not just on the design or material, is a prerequisite for comparison across studies.20
The BRUCE device was designed to replicate ankle and foot joint motion within the AFO and we imposed an ankle and forefoot joint to the AFO. Moreover, the location of the imposed ankle and forefoot joints was identical to the joint rotation centers used in instrumented gait analysis.21 This made it possible to calculate the moment and power delivered by the AFO throughout the gait phase by multiplying AFO stiffness by the migration of the AFO from its neutral angle. Furthermore, the moments and power provided by the patient could also be determined by subtracting the AFO kinetics from the net joint moment and power. We applied this method17 to gain insight into the mechanical functioning of compliant polypropylene AFOs to overcome a drop-foot gait, and found that AFOs had sufficient stiffness to support the foot throughout swing without hampering ankle function during stance. Use of this method in a subsequent study18 revealed that spring-like carbon-composite AFOs account for 60% of the positive ankle work, and that these AFOs reduce the energy cost of walking by partially taking over active ankle work.
Two studies have addressed the AFO kinetics in pathological gait. Stanhope et al19 performed a double case study and found that a carbon AFO accounted for approximately half the ankle moment during stance in patients with postpolio syndrome. Wolf et al22 analyzed AFO kinetics in five children with spina bifida and found that an AFO accounted for 62% of peak ankle power in late stance, which is consistent with the 60% positive ankle power we found.3 Unfortunately Wolf et al22 did not report on the stiffness of the analyzed AFOs, hindering further comparison. Furthermore, Stanhope et al19 and Wolf et al22 did not impose an ankle joint to the AFO while measuring its stiffness, which raises the question of whether their obtained AFO moments can be compared to moments obtained from clinical gait analysis.
Benefits at activity level
In our research, we evaluated the functioning of AFOs by mechanical analysis of the AFO throughout the gait cycle. We also evaluated the effect of the AFO at activity level. By combining these two types of evaluations, the working mechanisms of compliant AFOs and spring-like AFOs can be determined.
In our seven-patient study,17 we evaluated both the mechanical effects and the effect at activity level of compliant AFOs used to overcome a drop-foot gait. We observed clinically relevant improvements in walking energy cost only when the AFO corrected the drop-foot position during swing. In these patients, the AFO was stiff enough to resist an ankle moment large enough to support the foot during swing, but also flexible enough that it did not hamper the ankle in stance. In patients who did not benefit at activity level, the analysis showed that dorsiflexion in swing was already sufficient, and that the AFO did not add additional support to the foot in swing (Figure 2).
This analysis indicated a clear link between the mechanical functioning of the AFO and the benefits perceived by the patient at activity level. Although various research has investigated the effects of polypropylene AFOs using a combined evaluation of kinematics and kinetics and the energy cost of walking,23-25 our study was the first to relate the mechanical functioning of the AFO to the effects of the AFO at activity level.
In a study of 10 patients with stroke or multiple sclerosis18 we analyzed the effects of the spring-like AFOs on patients with reduced ankle push-off. Prior to the study, we expected the spring-like AFOs to augment the ankle push-off, increase knee flexion in initial swing, and reduce the energy cost of walking. However, the combined mechanical and functional evaluation revealed that reductions in the energy cost of walking are achieved by the AFO substituting for, rather than augmenting, ankle work (Figure 3). Specifically, the AFO accounted for 60% of the positive ankle work, resulting in an 11% reduction in the total work in the affected leg and we observed a consistent 10% reduction in the energy cost of walking.
Various studies have described the kinematics and kinetics of walking with a spring-like AFO.16,22,26,27 As in our study, the kinematics and kinetics in these studies did not find an augmenting effect for the AFO. However, none of the studies assessed the energy cost of walking or performed a mechanical evaluation of AFO function. Information about working mechanisms of AFOs is essential, as we aim to optimize the functional benefit obtained from AFOs by varying their mechanical properties.
The optimal AFO
The next phase of our research focused on trying to optimize AFO benefits at activity level by varying the mechanical properties and, thereby, the mechanical behavior of the AFO during gait.
In a recent study,28 we used a forward-dynamic simulation model to study the effects of variations in AFO stiffness on the energy cost of walking. In the simulation model, the optimal AFO stiffness in terms of the energy costs appeared to be a function of both the amount of energy returned by the AFO and the timing of that energy return (Figures 4 and 5). Subsequently,29 we studied the effect of AFO stiffness on the energy cost of walking in patients with reduced ankle push-off. Again, we observed an AFO stiffness at which the energy cost was minimal that appeared to be the best compromise among an increase in work done by the AFO, a decrease in the ability to generate ankle push-off, and an increase in the need to perform work at the hip with increasing AFO stiffness.
The optimal AFO stiffness for the simulation model was higher than the optimal stiffness found in the patient trials, possibly because of the absence of ankle muscles in the simulation model. Consistently, the optimal AFO stiffness in the simulation model resembled the quasi-stiffness of the ankle during nonpathological gait.30
Both the simulation study and the test with patients indicated the presence of an optimal AFO stiffness, which provides maximal function at a minimal energy cost to the patient. In the literature, no studies have yet reported on the changes in energy cost with systematic variations in mechanical AFO properties. However, different types of AFO have been compared. Buckon et al23 studied the effects of three AFOs (hinged, solid, and posterior leaf spring) on the energy cost of walking in children suffering from cerebral palsy. In their study all three AFOs resulted in a significant reduction in energy cost compared with walking without the AFO, though differences in energy cost among the three AFOs were marginal. Due to the absence of an evaluation of the mechanical functioning of the AFOs, however, it is unclear to what extent each AFO contributed to the patients’ gait.
Based on our collective findings, we have proposed an AFO design for patients with reduced ankle push-off to augment push-off compared with walking without an AFO.31 Unlike a typical spring-type AFO, this device would allow for unrestricted ankle plantar flexion; because of its similarity in this regard to the klap skate used in speed skating, we have called it the Klap-AFO. A key feature of the design is that its stiffness during plantar flexion would be lower than during dorsiflexion. This type of device would be expected to increase ankle push-off power while lowering energy costs of walking. Our proposed design is currently theoretical; we are not aware of any existing devices of this type.
With the introduction of the BRUCE device, it is now possible to characterize AFOs mechanically in the clinical setting. This characterization is expected to further objectify the AFO prescription process. In clinical practice, AFOs are commonly characterized by their form and materials rather than by their function (i.e., their mechanical properties), despite the fact that the same mechanical properties of an AFO can be achieved with different forms and materials. Objective information about an AFO’s properties transforms the AFO prescription from an experience-based process to a more explicit one, allowing for better communication among practitioners, accurate documentation, and improved education for future practitioners.
In our research we observed that low-stiffness AFOs (0.2 Nm/deg) adequately support the foot throughout the swing phase without interfering with the patient’s ankle kinetics during the stance. Therefore, in multiple sclerosis or stroke patients with weak ankle dorsiflexor muscles and absence of spasticity an AFO with 0.2 Nm/deg of stiffness is recommended. Individual variations in AFO stiffness appear not to be needed in these low-stiffness AFOs prescribed to overcome drop-foot gait. Because beneficial effects in patients are present only when a drop foot in the swing phase is observed, clinicians should conduct 3D video gait analysis, even in “obvious” gait-related disorders. When prescribing a 0.2 Nm/deg stiffness AFO in this patient group, practitioners can expect both a clinically relevant improvement in walking speed and in energy costs.
Based on model simulations and patient trials, we conclude that the energy cost of walking can be optimized by choosing an appropriate stiffness for spring-like AFOs. For patients with central neurologic disorders with reduced ankle push off power who do not have balance problems or severe spasticity, an average AFO stiffness of 1.4 (+/-0.4) Nm/deg was the most beneficial stiffness we observed. Hence, for this clinical population, that 1.4 Nm/deg value may serve as a guideline for the optimal AFO stiffness. To diagnose a reduced ankle push off power we recommend the use of 3D gait analysis in addition evaluation of clinical signs, such as a delayed heel rise in stance and a reduced capacity to produce force in the calf muscles. Because stiff AFOs (3-6 Nm/deg) are less beneficial in terms of energy cost of walking, they should be prescribed only when other possible beneficial effects of the AFO, such as control of the effects of spasticity or the prevention of knee hyperextension in midstance, outweigh the benefit of energy cost reduction.
Recommendations for future research
Our research suggests that to fully understand AFO functioning, future studies should include a mechanical evaluation of AFO function throughout the gait phase. To achieve this investigators will need to measure mechanical AFO properties with a device such as the BRUCE. Moreover, the orthopedic and O&P industries should classify AFOs according to their stiffness. In addition to a mechanical evaluation of the AFO, an evaluation of the effects of the AFO at activity level is needed to fully understand AFO functioning.32 Specifically, the significance of changes in kinematics and kinetics as a result of applying an AFO cannot be known when no information is available on changes at activity level. Therefore, we recommend that future orthotic research should include both an evaluation of mechanical AFO function and an assessment of AFO functioning at activity level.32
To ensure valid comparisons could be made among different degrees of AFO stiffness, we kept the neutral ankle angle and the forefoot properties of the AFOs constant during these experiments. We recommend that future studies evaluate the effects of variations in these properties.
With regard to the neutral angle of the AFO around the ankle, it is known that this may influence the kinematics and kinetics of walking with an AFO.11-13 For example, undesired knee extension may be introduced by incorrect alignment of the AFO-footwear combination.11 With regard to the neutral angle in spring-like AFOs in particular, we expect that a more plantar-flexed position would allow for equivalent energy storage at a lower AFO stiffness. Consequently, a greater range of motion at the ankle, a higher ankle angular velocity, and greater push-off power could be expected with more plantar-flexed neutral angle. However, it is expected that a more dorsiflexed position of the AFO is needed to induce knee flexion in early stance. In future studies that investigate the interplay between these two mechanisms the AFO neutral angle should be systemically varied. Also, a simulation model such as the one used in our research could be used to gain conceptual insights into the role of the neutral angle of the AFO.
In addition, the mechanical AFO properties at the forefoot are expected to influence AFO function. Recently, Fatone et al14 found that AFOs with a full-length foot-plate resulted in an increased ankle moment compared with AFOs with a three-quarter-length foot-plate. It is also known that part of the benefit of an AFO can be attributed to the shoes compared with walking barefoot.15,16 The mechanism through which the mechanical AFO forefoot properties determine the effects of the AFO is not known, however. We expect that with spring-like AFOs, a high forefoot stiffness and a small AFO neutral angle result in the ground reaction force progressing forward on the foot more easily, resulting in higher ankle moments and capacity to store energy at the AFO ankle component. However, this would also introduce high knee extension moments, and increased momentum would be needed to rotate around the tip of the toe in an AFO with high forefoot and ankle stiffness.
Research in which AFO forefoot properties are systematically varied is needed to gain insight into the effects of the mechanical AFO properties around the forefoot. Simulations models such as ours could be used to investigate the conceptual mechanisms that cause AFO forefoot properties to influence AFO mechanics and to clarify the interplay between mechanical AFO properties at the ankle and forefoot. The AFO in our simulation model would need to be extended with a forefoot joint for this purpose.
Predicting optimal AFO properties
In a group of patients with reduced ankle push-off power, we found the optimal AFO stiffness to be 1.4 Nm/deg on average. However, to generalize this optimal AFO stiffness to other clinical populations, including children, future research should focus on predicting optimal AFO properties based on the patient’s characteristics. For example, in patients with completely absent ankle function, the optimal AFO stiffness may be higher.
We expect a good prediction will require identification of:
1) The relationship between the quantified primary impairments for which the AFO is prescribed (e.g., walking with ankle push-off power reduced by 50%), and the optimal AFO properties required to overcome activity limitations caused by these impairments; and
2) The factors that interfere with this relationship, and how they influence the optimal AFO properties (e.g., if the patient’s passive ankle stiffness is elevated, a more compliant AFO is required to achieve the same net ankle-AFO stiffness). Potential factors that could influence optimal AFO properties include body weight, passive ankle stiffness, spasticity, and force-generating capacity at the knee and hip.
Because experiments aiming to predict optimal AFO properties introduce a high load for participants, simulation models may be a useful alternative; models that include a detailed representation of the human musculoskeletal system appear to be the most appropriate.33
Obviously, the proposed working mechanism of the Klap-AFO should be tested in patients with reduced ankle push-off. We expect that with the Klap-AFO ankle power will be augmented while walking without an AFO and while walking with a typical spring-like AFO. Hence, we expect patients to walk faster and with lower energy costs. In order to verify whether the Klap-AFO takes over ankle function as expected, mechanical evaluation of the device should be extended with electromyographic measurements of the calf muscles.
Daan J.J. Bregman, PhD, is coordinator of sports innovation at the Delft University of Technology in Delft, The Netherlands. This article was adapted from his 2011 doctoral dissertation, which he completed while at VU University Medical Center in Amsterdam.
- Bregman DJ, Rozumalski A, Koops D, et al. A new method for evaluating ankle foot orthosis characteristics: BRUCE. Gait Posture 2009;30(2):144-149.
- Sumiya T, Suzuki Y, Kasahara T. Stiffness control in posterior-type plastic ankle-foot orthoses: effect of ankle trimline. Part 1: A device for measuring ankle moment. Prosthet Orthot Int 1996;20(2):129-131.
- Cappa P, Patané F, Pierro MM. A novel device to evaluate the stiffness of ankle-foot orthosis devices. J Biomech Eng 2003;125(6):913-917.
- Cappa P, Patané F, Di Rosa G. A continuous loading apparatus for measuring three-dimensional stiffness of ankle-foot orthoses. J Biomech Eng 2005;127(6):1025-1029.
- Novacheck TF, Beattie C, Rozumalski A, Gent G, Kroll G. Quantifying the spring-like properties of ankle-foot orthoses (AFOs). Prosthet Orthot Int 2007;19(4):98-103.
- Lunsford TR, Ramm T, Miller JA. Viscoelastic properties of plastic pediatric AFOs. J Prosthet Orthot 1994;6(1):3-9.
- Polliack AA, Swanson C, Landsberger SE, McNeal DR. Development of a testing apparatus for structural stiffness evaluation of ankle-foot orthoses. J Prosthet Orthot 2001;13(3):74-82.
- Singerman R, Hoy MG, Mansour JM. Design changes in ankle-foot orthosis intended to alter stiffness also alter orthosis kinematics. J Prosthet Orthot 1999;11(3):48-56.
- Yamamoto S, Ebina M, Iwasaki M, et al. Comparative study of mechanical characteristics of plastic AFOs. J Prosthet Orthot 1993;5(2):59-64.
- Kobayashi T, Leung AK, Akazawa Y, et al. Design of an automated device to measure sagittal plane stiffness of an articulated ankle-foot orthosis. Prosthet Orthot Int 2010;34(4):439-448.
- Jagadamma KC, Owen E, Coutts FJ, et al. The effects of tuning an ankle-foot orthosis footwear combination on kinematics and kinetics of the knee joint of an adult with hemiplegia. Prosthet Orthot Int 2010;34(3):270-276 .
- Miyazaki S, Yamamoto S, Kubota T. Effect of ankle-foot orthosis on active ankle moment in patients with hemiparesis. Med Biol Eng Comput 1997;35(4):381-385.
- Owen E. The importance of being earnest about shank and thigh kinematics especially when using ankle-foot orthoses. Prosthet Orthot Int 2010;34(3):254-269
- Fatone S, Gard SA, Malas BS. Effect of ankle-foot orthosis alignment and foot-plate length on the gait of adults with poststroke hemiplegia. Arch Phys Med Rehabil 2009;90(5):810-818.
- Churchill AJ, Halligan PW, Wade DT. Relative contribution of footwear to the efficacy of ankle-foot orthoses. Clin Rehabil 2003;17(5):553-557.
- Desloovere K, Molenaers G, Van Gestel L, et al. How can push-off be preserved during use of an ankle foot orthosis in children with hemiplegia? A prospective controlled study. Gait Posture 2006;24(2):142-151.
- Bregman DJ, De Groot V, Van Diggele P, et al. Polypropylene ankle foot orthoses to overcome drop-foot gait in central neurological patients: a mechanical and functional evaluation. Prosthet Orthot Int 2010;34(3):293-304.
- Bregman DJ, Harlaar J, Meskers CG, de Groot V. Spring-like ankle foot orthoses reduce the energy cost of walking by taking over ankle work. Gait Posture 2012;35(1):148-153.
- Stanhope SJ, Siegel KL, Halstead LS. Contribution of dynamic ankle-foot orthoses to ankle moments during stance in gait. Presented at 12th ISPO World Congress, Vancouver, Canada, August 2007.
- Ridgewell E, Dobson F, Bach T, Baker R. A systematic review to determine best practice reporting guidelines for AFO interventions in studies involving children with cerebral palsy. Prosthet Orthot Int 2010;34(2):129-145.
- Cappozzo A, Catani F, Croce UD, Leardini A. Position and orientation in space of bones during movement: anatomical frame definition and determination. Clin Biomech 1995;10(4):171-178.
- Wolf SI, Alimusaj M, Rettig O, Döderlein L. Dynamic assist by carbon fiber spring AFOs for patients with myelomeningocele. Gait Posture 2008;28(1):175-177.
- Buckon CE, Thomas SS, Jakobson- Huston S,et al. Comparison of three ankle-foot orthosis configurations for children with spastic diplegia. Dev Med Child Neurol 2004;46(9):590-598.
- Balaban B, Yasar E, Dal U, et al. The effect of hinged ankle-foot orthosis on gait and energy expenditure in spastic hemiplegic cerebral palsy. Disabil Rehabil 2007;29(2):139-144.
- Brehm MA, Harlaar J, Schwartz M. Effect of ankle-foot orthoses on walking efficiency and gait in children with cerebral palsy. J Rehabil Med 2008;40(7):529-534.
- Van Gestel L, Molenaers G, Huenaerts C, et al. Effect of dynamic orthoses on gait: a retrospective control study in children with hemiplegia. Dev Med Child Neurol 2008;50(1):63-67.
- Ounpuu S, Bell KJ, Davis RB III, DeLuca PA. An evaluation of the posterior leaf spring orthosis using joint kinematics and kinetics. J Pediatr Orthop 1996;16(3):378-384.
- Bregman DJ, van der Krogt MM, de Groot V, et al. The effect of ankle foot orthosis stiffness on the energy cost of walking: a simulation study. Clin Biomech 2011;26(9):955-961.
- Bregman DJJ. Spring-like ankle foot orthoses reduce the energy cost of walking in patients with reduced ankle push-off only when their stiffness is appropriate. In: The optimal ankle-foot orthosis: The influence of mechanical properties of Ankle Foot Orthoses on the walking ability of patients with central neurological disorders [dissertation]. Amsterdam: VU University Amsterdam, The Netherlands. 2011;105-126.
- Frigo C, Crenna P, Jensen LM. Moment-angle relationship at lower limb joints during human walking at different velocities. J Electromyogr Kinesiol 1996;6(3):177-190.
- Bregman DJJ. An ankle foot orthosis to augment ankle push-off: the Klap-AFO. In: The optimal ankle-foot orthosis: The influence of mechanical properties of Ankle Foot Orthoses on the walking ability of patients with central neurological disorders [dissertation]. Amsterdam: VU University Amsterdam, The Netherlands. 2011;127-136.
- Harlaar J, Brehm M, Becher JG, et al. Studies examining the efficacy of ankle foot orthoses should report activity level and mechanical evidence. Prosthet Orthot Int 2010;34(3):327-335.
- Crabtree CA, Higginson JS. Modeling neuromuscular effects of ankle foot orthoses (AFOs) in computer simulations of gait. Gait Posture 2009;29(1):65-70.