The Influence of Ankle-Foot Orthoses on Gait and Energy Expenditure in Spina Bifida
by C.M. Duffy, H.K. Grahm, A.P. Cosgrove, Journal of Pediatric Orthopedics, 2000;20;356–361.
Julie Hassler, Northwestern University
It is known that immobilizing a joint, as when using an ankle foot orthosis (AFO), leads to increased energy cost, decreased walking speed and decreased cadence in normal individuals. Children with low-level spina bifida, however, obtain great benefit from wearing orthotics. AFOs worn by children with low lumbar or sacral level spina bifida are designed to control the ankle mortise, control the subtalar joint by preventing collapse of the heel into valgus, control the mid-tarsal joint to prevent forefoot abduction and adduction, prevent primary deformity or recurrence of corrected deformity, counteract internal deforming forces, and improve the appearance of gait. AFOs are hypothesized to decrease energy expenditure in children with spina bifida because the orthoses improve the stability at the ankle so the child does not have to expend excess energy to control the ankle. The purpose of this study was to examine the effects of AFOs on gait and energy expenditure in children with low-level spina bifida.
Fifty-nine children with mid to lower lumbar and sacral level spina bifida underwent gait analysis as part of a larger study. Of those 59, 12 were selected to participate in this study based on their use of AFOs and the ability to walk three minutes with and without AFOs. The patient population included nine boys and three girls aged six to 16 years. Four individuals had L4 level lesions, four had L5 level lesions, and four had sacral level lesions. All individuals had functional (equal to or greater than manual muscle test grade of three) quadriceps, medial hamstrings, and tibialis anterior. Children with L5 level lesions also had functional abductors and lateral hamstrings, while patients with sacral level lesions additionally had functional plantar flexors. Eleven of the 12 children ambulated without an assistive device, and one child used crutches. All children who participated wore bilateral solid, polypropylene AFOs with the ankle at a neutral to slightly dorsiflexed position.
After choosing the subjects, a physical exam was conducted to determine fixed deformity. All children then underwent three-dimensional gait analyses with and without their AFOs. During the gait analysis, kinematic, kintetic, gait parameters and energy expenditure were measured. Kinematic data included pelvic tilt, hip flexion, knee flexion, ankle dorsiflexion, pelvic obliquity, hip abduction, and pelvic rotation. Kinetic data were analyzed using force plates to analyze the force about the ankle. The children were unaware of the presence of the force plates. Gait parameters included cadence, walking speed, duration of single and double support, and stride length. Finally, energy expenditure was measured using the previously validated Cosmet K2 that determined oxygen consumption and ultimately energy cost with and without the AFOs. The differences between the AFO and non-AFO conditions were examined using the paired t test.
Fixed flexion contractures were found in eight hips and 11 knees. Calcaneal deformity was found in both ankles of one subject. The kinematic data from the three-dimensional gait analysis showed no significant difference at the pelvis or knee in the coronal or saggital planes with and without the AFOs. The most significant finding was increased hip flexion at initial contact (P = 0.008), but no increase in hip extension at midstance when using the AFOs. Knee extension at initial contact and knee flexion at terminal stance showed a trend toward improvement with use of the AFO. Kinetic data were used to determine the force generation about the ankle from force plates. The data revealed a significantly greater force generation about the ankle with the use of AFOs at terminal stance, with 0.5 and 1.3 Watts/kg, respectively, being the forces for barefoot and AFO conditions. This result was significant, regardless of the level of the lesion (P = 0.01). Gait parameters while wearing the AFOs demonstrated an increase in walking speed and stride length (both p values < 0.001) and a decrease in double support phase (P < 0.005). The time in single-support phase and the cadence did not differ across the two conditions. Finally, total energy cost was significantly decreased (P < 0.01) when wearing the AFOs. Although the results show that the rate of oxygen consumption was not decreased, the oxygen cost of walking was less when wearing AFOs (0.33 ml/kg/m2) than without the AFOs (0.41 ml/kg/m2).
Many results of the study suggest children with spina bifida attain greater stability when using AFOs, which leads to a decreased energy cost of walking with orthoses. Walking speed increased, the duration of double support decreased and the stride length increased. Increased walking speed combined with the decreased duration of double support suggests the child is able to derive greater stability when wearing the orthoses. The children spend less time in double support indicating they do not have to attain stability from double support for the next forward progression. The decreased time in double support, but same time spent in single-limb support translates into less overall time to finish a gait cycle and therefore increased walking speed. Similarly, since the children are more stable they are able to take steps without loosing balance while wearing AFOs, which results in greater stride length.
The most significant kinetic finding was that of increased power generation about the ankle at terminal stance. None of the children had significant power generation about the ankle while walking barefoot, however all children had significant increase in ankle power while walking with the AFOs. As children with lesions at the sacral level have functional plantar flexors, the groups were divided into groups according to lesion level to analyze the significance of this finding. The analysis revealed statistically significant differences for each lesion level. Therefore, the increase in power generation at terminal stance is an effect of the orthoses, no matter the level of the lesion. It is likely that the pseudo-power generation conferred by the AFO is responsible for pushing the foot into swing, whereas ordinarily for children with weak or paralyzed calf muscles it must be lifted. This increase in power at terminal stance may also be related to the finding of increased hip flexion noted at initial contact and thus the improvements in stride length and walking speed. Therefore, this increase in force generation at terminal stance is an integral aspect of the AFO that allows for a chain reaction of other events that eventually lead to decreased energy expenditure with the use of the AFO.
In terms of energy expenditure, oxygen cost decreased and walking speed increased with the use of AFOs. In previous studies the same researchers found energy conservation in gait to be correlated with kinematic control about the pelvis in the transverse and coronal planes and control of hip abduction. No differences were seen in any of these parameters, which suggest that the energy conservation observed in this study was not due to improvement in pelvic stability. Rather, the energy conservation was due to the stability and the increased force generation about the ankle produced by the AFO.
The authors mentioned that a limitation of the study was that the AFOs prescribed for the children, although described as solid, proved to be too flexible to correct for some kinematic abnormalities. The flexibility of the AFOs allowed kinematic abnormalities to persist but was also responsible for the simulated power generation that was recorded about the ankle in terminal stance. This increase in power generation was responsible ultimately for increasing hip flexion at initial contact, stride length, and walking speed, thus decreasing oxygen cost. Authors of a previous study did not find the increased force generation about the ankle, but instead found reduced peak force generation in preswing about the ankle. It is likely that the differences between the findings of the two studies are attributable to the differences in the orthoses. The difference in the findings of force generation about the ankle between this study and the previous study address the importance of determining how an orthosis works for a specific child. If the child is having trouble with force generation at terminal stance and not gaining the subsequent benefits of the AFO, it may be desirable to have a less rigid AFO to confer the benefits observed in this study.
Other limitations of the study involve the characteristics of the subject pool including the small sample size and thus limited application of the study to other populations. Of the 59 subjects that were evaluated for the larger study, only 12 qualified for this study based on the inclusion criteria. With such a small sample size, only subjects that are very similar to those studied might benefit from the findings, which are not generalizable to broader populations. When evaluating the inclusion criteria of the study, only subjects who were able to walk without their AFOs for three minutes were invited to participate in the study. Though this inclusion criterion was necessary for data collection, children with spina bifida that are able to walk without AFOs for this amount of time are already at a high level of function. Therefore, the application of the study to populations that are unable to walk without AFOs for that amount of time is not addressed.
Despite the question of the exact features of the AFOs and the limitations of the subject selection, this study provides important reinforcement of the need for orthoses for children with low-level spina bifida. The orthoses allow these children to be more stable in ambulation, which allows them to walk more efficiently and, in turn, confers greater benefits of walking.