INTRODUCTION AND PURPOSE
Ankle-foot orthoses (AFOs) are commonly prescribed to improve gait in children with cerebral palsy (CP). Ankle-foot orthoses can improve gait kinematics and temporal-spatial parameters,1–5 activity levels,6 and metabolic costs.7,8 However, many different types of AFOs are available and little evidence exists to guide orthosis selection for individual patients. Consequently, improvements in gait with AFOs are highly variable and often unsatisfactory.9 Limitations of AFO design and inappropriate prescription for an individual's gait pattern can hinder the effectiveness of AFOs.10
Traditional AFOs are usually molded from plastic. Common styles include solid AFOs and hinged AFOs with dorsiflexion or plantar-flexion stops. Unfortunately, many traditional AFOs have detrimental effects such as reducing push-off power because of the AFO restricting plantar flexion.4,10,11 This is problematic because ankle plantar flexor power generation in preswing is 1 of the 3 primary mechanisms used to increase walking speed.12,13 Some orthosis companies provide additional or alternate designs, for example, dynamic AFOs (DAFOs) and adjustable dynamic response (ADR) AFOs. Posterior leaf spring AFOs (DAFO 3.5) are offered by Cascade DAFO, Inc (Ferndale, WA), with a trimmed posterior strut to allow greater flexibility than solid AFOs, in addition to more traditional solid (DAFO 3) and hinged (DAFO 2) orthoses and other designs. Ultraflex Systems, Inc (Pottstown, PA), promotes ADR-AFOs featuring joints with an elastomer component that produces variable resistance to dorsiflexion and plantar flexion. Dorsiflexion and plantar-flexion stiffness can be adjusted separately by an orthotist or physical therapist to “tune” the brace for the individual patient. The purpose of this study was to compare the Cascade DAFO and Ultraflex ADR-AFO approaches to orthotic prescription in children with CP and crouch and/or equinus gait.
Participants wore braces prescribed using the DAFO and ADR-AFO approaches during a randomized cross-over design. The participants wore each brace for 4 weeks in randomized order. They were assessed barefoot before they started wearing the first brace (baseline), barefoot and braced after wearing the first brace for 4 weeks, and barefoot and braced after wearing the second brace for 4 weeks. The braces were worn consecutively, with no “washout” period between the 2 brace wear periods. Differences in gait while wearing braces or while barefoot were examined. Walking activity and parent-reported outcomes were compared between the 2 bracing methods.
The study participants were 10 children with CP (age 4-12 years); 6 were classified at Gross Motor Function Classification System (GMFCS) level I, and 4 were at GMFCS level III. All had crouch and/or equinus gait. Crouch was defined as excessive knee flexion throughout stance phase and equinus as excessive static or dynamic plantar flexion throughout stance phase on the basis of the physical therapists' visual observation of the child's gait.14 Five participants had bilateral involvement, and 5 had unilateral involvement (3 right, 2 left; Table 1). All participants and/or parents provided written assent and consent after approval of the study by our hospital's Institutional Review Board.
Orthoses Fabrication and Fitting
A single certified and experienced pediatric orthotist molded and fit each child for both DAFO and ADR-AFOs (5 unilateral and 5 bilateral) (Figure 1). The manufacturers of the orthoses determined the specifications of the orthoses provided to each participant (including the type of DAFO) on the basis of a physical examination performed by the motion analysis laboratory physical therapist before molding, specifically range of motion, muscle tone, and a videotape of the participant walking barefoot. The same shoes designed for use with AFOs (Apis Answer2 Shoes, Apis Footwear Company, South El Monte, CA) were worn with both orthoses. These were sized to accommodate the ADR-AFO brace, which tends to be bulkier than the DAFO; shims were provided to ensure good fit when the DAFOs were worn in the shoes. All participants were prior orthosis users and were instructed not to use their previous AFOs during the course of the study. No gait training specific to use of the new orthoses was provided, but participants continued with their regular physical therapy programs.
Final fitting of both orthoses and tuning of the ADR-AFO braces was done by the orthotist investigator. The concept underlying the ADR-AFO approach is that dorsiflexion and plantar-flexion stiffness of the brace can be separately adjusted (after the brace has been made) by using an Allen wrench to compress anterior and posterior elastomer channels on joints built into the brace. The orthotist's objective was to optimize knee and ankle position during loading response, terminal stance, and preswing. Heel wedges placing the foot in a more plantar flexed position in the shoe were used to enable initial contact with the heel and to promote the first rocker. The height of the heel wedge was selected by the manufacturer to allow loading of the heel with the tibia vertical in a weight-bearing position (creating a 90° angle between the tibia and sole of the shoe); the heel wedge height was measured with the child in the supine position, knee fully extended. The DAFOs follow a more conventional approach with the brace having a fixed stiffness and optional dorsi-/plantar-flexion stops. The level of support was adjusted by selecting the appropriate DAFO style, which was done by the manufacturer in this study.
Walking Activity Measures
Walking activity while wearing the braces was measured during each wear period using a StepWatch activity monitor (Orthocare Innovations, Mountlake Terrace, WA) that was mounted on the brace to ensure compliance with the monitoring (Figure 1A and B). The StepWatch was individually calibrated for each participant at each visit and set to record for up to 30 days; the first and last days were not included in the analysis because they were partial days. Days where the StepWatch recorded less than 5% of the participant's average daily step count were also excluded from the analysis under the assumption that the brace was not worn that day. Compliance with brace wear was assessed by the percentage of days the brace was worn (number of days with step count ≥5% of average daily step count/number of full days in recording period). The number of hours the brace was worn was estimated as the time between the first recorded activity (step) and the last recorded activity (step) in a day, and active time was calculated as the total number of minutes in which at least 1 step was recorded.
Instrumented Gait Analysis
Gait analysis using a Vicon 612 motion capture system (Vicon Motion Systems Ltd, Oxford, UK) with 4 in-floor triaxial force plates (Advanced Mechanical Technology, Inc, Watertown, MA) was performed using the conventional gait model15 at baseline and at the end of each 4-week brace wear period. The testing was completed barefoot at baseline and barefoot and braced (using the brace worn for the previous 4 weeks) after each bracing period. Testing was performed without assistive devices for all children at GMFCS level I. Children in GMFCS level III were tested either with the assistive device they used in everyday life or with hands held if data could not be collected with the assistive device because of marker obstruction. Two children had different assistance between tests because they forgot to bring their assistive device to 1 of the test sessions (Table 1). Data from 4 to 10 strides were averaged to calculate kinematic, kinetic, and temporal-spatial parameters. Asymmetry of gait was assessed by examining the magnitude of difference in step length, step time, and single-limb stance time between the left and right sides for each participant. For subjects who could stand independently, standing balance was assessed at the same time points by analyzing motion of the center or pressure (COP) while the participant stood on a force plate on both feet with eyes open for 30 seconds. Twenty seconds of data from the middle of the trial were analyzed to determine the range and standard deviation of anterior-posterior and medial-lateral motion, as well as the average speed of COP motion (ie, path length/time).
Parent Satisfaction Measures
Parent-reported outcomes were assessed at the end of each wear period using the Orthotics and Prosthetics Users' Survey (OPUS)16 and the Pediatric Outcomes Data Collection Instrument (PODCI) questionnaires.17 The OPUS is a validated set of questionnaires assessing functional status, quality of life, and satisfaction with orthotic and prosthetic devices and services.16 The PODCI is a validated instrument commonly used to assess function and quality of life in pediatric orthopedics.18 The questionnaires were completed by a parent and were scored following the instructions for each instrument. Raw scores are reported for the OPUS, and standardized scores (0-100) are reported for the PODCI.
Outcomes were compared between braces and between braced and barefoot conditions using paired t tests. The pairing was within the same session for braced versus barefoot and between sessions for DAFO versus ADR-AFO. Because different styles of DAFOs were used, analysis was also performed subdividing the DAFOs according to the type of stance-phase dorsiflexion (free vs block/resist) and plantar-flexion (block vs resist) control they provided. All statistical analysis was performed using Stata (version 12.1, StataCorp LP, College Station, TX).
In terms of overall gait, children took longer strides (11-12 cm difference; P < .0001) at a reduced cadence (8-11 steps/min difference; P ≤ .005) when wearing both types of orthoses compared with walking barefoot (Tables 2–3). Velocity increased significantly when walking in DAFOs (P = .006), and a trend toward increased velocity in the ADR-AFOs did not reach statistical significance. The differences in temporal-spatial parameters were smaller for DAFOs that blocked plantar flexion and allowed free dorsiflexion. Temporal-spatial parameters did not differ significantly between the 2 orthotic approaches (Table 3). No significant effects of bracing or type of brace were observed on gait asymmetry (P > .3) or standing balance (P > .51) (Tables 4 and 5).
Both orthotic approaches improved dorsiflexion (restricted plantar flexion) in swing compared with barefoot (9°-17° difference vs barefoot; P ≤ .01). Dorsiflexion also increased significantly in stance for the DAFOs (13° difference from barefoot; P = .003) and showed a trend toward increasing in the ADR-AFOs, but did not reach statistical significance (6° difference from barefoot; P = .08). Dorsiflexion in stance and swing was greater in the DAFOs than in the ADR-AFOs (9°-10° difference; P ≤ .003) because of wedging of the ADR-AFOs and greater plantar-flexion stiffness of the DAFOs. Knee extension in stance was better in the ADR-AFOs than in the DAFOs (5° difference; P = .006), but neither brace differed significantly from barefoot. Hip extension in stance was improved in both types of braces compared with barefoot (2°-4° difference; P ≤ .002), but no difference was found between the 2 brace approaches.
Push-off power was significantly reduced in both brace approaches (0.4-0.6 W/kg difference; P ≤ .03) but was significantly higher in the ADR-AFOs than in the DAFOs (0.3 ± 0.2 W/kg difference; P = .04). Peak plantar-flexion moment in late stance was higher in both types of braces compared with barefoot (0.17-0.19 Nm/kg), reaching statistical significance for the DAFOs (P = .03). During the loading response, a dorsiflexion moment was not seen for any limb barefoot, but was seen in 3 of the 7 limbs in ADR-AFOs and 5 of the 7 limbs in DAFOs, with the difference being significantly different from barefoot for the DAFOs (P = .02). We were not able to identify any clear pattern of differences in kinematics or kinetics associated with the style of DAFO used.
Parent-reported patient satisfaction with the orthoses was significantly higher for the DAFO-style orthoses compared with the ADR-AFOs (9.2 ± 6.7 point difference in OPUS device satisfaction; P = .002) because of higher satisfaction with fit, weight, comfort/pain, ease of use, and cosmesis (P ≤ .05) (Table 6). No differences were found between orthotic approaches in terms of function, service, or quality of life as measured by the OPUS and PODCI questionnaires.
Participants spent more time walking (3.9 vs 3.2 hours per day; P = .02) and recorded more steps per day in the DAFO-style braces (5952 steps per day) compared with the ADR-AFOs (5224 steps per day) with an average within-participant difference of 729 ± 1119 steps per day (P = .07; Table 6). All 10 participants recorded more walking time in the DAFO-style braces than in the ADR-AFOs, and 8 of the 10 participants recorded more steps per day when using DAFOs (Figure 2). No difference was found between orthotic approaches for the percentage of days used, but there was a trend toward a longer time from the first recorded activity to the last in DAFOs compared with ADR-AFOs (P = .12, Table 6).
Both the DAFO and ADR-AFO approaches improved hip extension in stance and dorsiflexion in swing during gait compared with the barefoot condition. The DAFO braces all blocked or restricted plantar-flexion motion and produced more normal first rocker as evidenced by a more frequent and greater magnitude internal dorsiflexion moment in loading response. They also allowed more normal dorsiflexion motion in stance phase with associated improvement in plantar flexor moment production in stance, indicating a more normal second rocker. By design, the ADR-AFOs were set to provide customized resistance to plantar flexion in stance. This allowed for some preservation of push-off power at preswing, which theoretically should be beneficial for energy-efficient gait; though this was not reflected in our outcomes for gait speed, number of steps taken per day, or parent-reported patient satisfaction. Follow-up studies measuring oxygen consumption could provide a more direct measure of energy efficiency. Although knee extension in stance was not significantly changed from the barefoot condition with either orthotic approach, knee extension in stance phase was better in the ADR-AFOs compared with the DAFOs, possibly as a result of more plantar flexion at the ankle. Because all but 1 of the participants had crouch gait with or without equinus, improved knee extension in stance is an important goal of bracing in addition to the control of ankle position. The small number of subjects precluded comparisons of subjects with equinus versus crouched gait.
The ADR-AFO braces positioned the ankle in a plantar flexed position for most participants in the current study, which could raise concerns about worsening plantar flexor spasticity or contracture over time for these participants. During the month-long period of ADR-AFO use in the current study, no participants experienced a clinically significant change in values for the gastrocnemius or soleus of either the angle of catch after fast velocity stretch (R1) or the angle of full passive range of motion (R2),19 indicating no effect of plantar flexed walking on spasticity or range of motion. The DAFO-style orthoses tended to promote a more flexed posture (more dorsiflexion and knee flexion during stance). Because crouch gait tends to worsen over time in patients with CP,14,20 allowing children with equinus and crouch to walk in more relative plantar flexion to promote a more upright posture may be appropriate and safe. Additional research with more subjects and longer follow-up is needed to further investigate this issue.
We noted that the children used the DAFO-style orthoses more than the ADR-AFOs and recorded more steps per day when wearing the DAFOs, and they expressed a preference for the DAFO on the OPUS questionnaire. Five children maintained similar walking times regardless of the brace approach. However, the other 5 were more active in the DAFO-style braces, averaging 1 hour and 23 minutes more active time in the DAFOs than in the ADR-AFOs. Although 5 children took about the same number of steps in both brace types, 3 walked at least 2000 steps per day more in the DAFOs, 1 walked 855 steps per day more in the DAFOs, and only 1 walked significantly more (974 steps per day) in the ADR-AFOs. These may be clinically significant differences because children at GMFCS level I normally average approximately 11 000 steps per day whereas children at GMFCS level III average approximately 3000 steps per day.21 Because the step monitor was mounted on the AFO, however, whether the higher recorded active time and number of steps was due to the children walking more during the DAFO period or better compliance with brace wear when using the DAFOs cannot be determined. Both compliance with brace wear and walking activity are clinically important. Patients must wear their braces to receive any benefit in function from their AFOs. It is an added benefit if bracing can also increase activity, participation, and quality of life. Although we did not observe any patterns suggesting an accommodation period for either brace approach, these benefits may require a longer time to occur for the ADR-AFO braces because most of the children wore either traditional AFOs or DAFOs before the study. A longer wear period may be needed to adapt to the ADR-AFO braces, which represent a bigger change from the children's previous braces.
This study has several limitations. First, only 10 children with CP with a mix of crouch and/or equinus gait participated. Because of the small sample size and heterogeneity of the participants, power of the statistical tests was generally low; additional differences between braces may exist but not have been detected in the current analysis (type 2 error). Only children in GMFCS levels I and III are represented although the inclusion criteria allowed for the participation of children in GMFCS level II. The study results should therefore be considered preliminary until they can be confirmed and refined by larger follow-up studies. Also, 2 children had different forms of assistance during tests because they forgot to bring their assistive devices to 1 of the test sessions. Ideally, assistive devices would have remained consistent across all test sessions because the change in assistance may have affected the gait parameters measured. All but 1 of the ADR-AFO braces had a heel wedge attached, whereas none of the DAFOs were wedged. The wedging rather than other aspects of brace design possibly caused some of the differences seen between the 2 brace types. Although the use of shoes without braces can alter gait, a “shoes alone” condition was not tested because the shoes used in the study do not fit well without orthoses. In addition, because of the age of the participants, this study assessed parent-reported satisfaction and functional ability; views on functionality and cosmesis may differ between the parent and the child.
Finally, because each child has a different clinical presentation, bracing was individualized rather than standardized. Bracing must be customized to the individual to be effective, and we believe this provides the fairest representation of how bracing is applied in actual clinical practice. Both manufacturers were given the information they requested to be able to provide the braces they deemed most appropriate for each child. Future research should focus on identifying the brace characteristics most appropriate for individuals with different gait patterns and clinical presentations.
Both Cascade DAFO and Ultraflex ADR-AFO orthotic approaches improve stride length, hip extension in stance, and dorsiflexion in swing compared with barefoot walking. The DAFO approach produces more normal ankle kinematics and kinetics through late stance, but the ADR-AFO method allows more normal push-off power at preswing, which may improve the energy efficiency of gait. Stance-phase knee extension is better in the ADR-AFO compared with the DAFO orthoses. Parent-reported patient satisfaction and walking activity are higher for the DAFOs because of lower weight and better comfort, fit, cosmesis, and ease of use.
1. Dursun E, Dursun N, Alican D. Ankle-foot
orthoses: effect on gait
in children with cerebral palsy. Disabil Rehabil. 2002;24(7):345–347.
2. Lam WK, Leong JC, Li YH, Hu Y, Lu WW. Biomechanical and electromyographic evaluation of ankle foot
orthosis and dynamic ankle foot
orthosis in spastic cerebral palsy. Gait
3. Radtka SA, Skinner SR, Johanson ME. A comparison of gait
with solid and hinged ankle-foot
orthoses in children with spastic diplegic cerebral palsy. Gait
4. Rethlefsen S, Kay R, Dennis S, Forstein M, Tolo V. The effects of fixed and articulated ankle-foot
orthoses on gait
patterns in subjects with cerebral palsy. J Pediatr Orthop. 1999;19(4):470–474.
5. Figueiredo EM, Ferreira GB, Maia Moreira RC, Kirkwood RN, Fetters L. Efficacy of ankle-foot
orthoses on gait
of children with cerebral palsy: systematic review of literature. Pediatr Phys Ther. 2008;20(3):207–223.
6. Bowman TE. Single-case study: traditional thermoplastic AFO versus adjustable dynamic response–a crossover single-case study. J Prosthet Orthot. 2010;22:84–90.
7. Maltais D, Bar-Or O, Galea V, Pierrynowski M. Use of orthoses lowers the O(2) cost of walking
in children with spastic cerebral palsy. Med Sci Sports Exerc. 2001;33(2):320–325.
8. Mossberg KA, Linton KA, Friske K. Ankle-foot
orthoses: effect on energy expenditure of gait
in spastic diplegic children. Arch Phys Med Rehabil. 1990;71(7):490–494.
9. Novacheck TF, Kroll GJ, Gent G, Rozumalski A, Beattie C, Schwartz MH. Orthoses. In: Gage JR, Schwartz MH, Koop SE, Novacheck TF, eds. The Identification and Treatment of Gait
Problems in Cerebral Palsy. Vol 180–181. London: Mac Keith Press; 2009:327–348.
10. Ounpuu S, Bell KJ, Davis RB 3rd, DeLuca PA. An evaluation of the posterior leaf spring orthosis using joint kinematics and kinetics. J Pediatr Orthop. 1996;16(3):378–384.
11. Romkes J, Brunner R. Comparison of a dynamic and a hinged ankle-foot
orthosis by fait analysis in patients with hemiplegic cerebral palsy. Gait
12. Jonkers I, Delp S, Patten C. Capacity to increase walking
speed is limited byimparied hip and ankle power generation in lower functioning persons post-stroke. Gait
13. Schwartz M, Rozumalski A, Trost JP. The effect of walking
speed on the gait
of typically developing children. J Biomech. 2008;41(1639–1650):1639.
14. Wren TA, Rethlefsen S, Kay RM. Prevalence of specific gait
abnormalities in children with cerebral palsy: influence of cerebral palsy subtype, age, and previous surgery. J Pediatr Orthop. 2005;25(1):79–83.
15. Davis R, Ounpuu S, Thyburski D, Gage J. A gait
analysis data collection and reduction technique. Hum Movement Sci. 1991;10:575–587.
16. Heinemann AW, Bode RK, O'Reilly C. Development and measurement properties of the Orthotics and Prosthetics Users' Survey (OPUS): a comprehensive set of clinical outcome instruments. Prosthet Orthot Int. 2003;27(3):191–206.
17. Daltroy LH, Liang MH, Fossel AH, Goldberg MJ. The POSNA pediatric musculoskeletal functional health questionnaire: report on reliability, validity, and sensitivity to change. Pediatric Outcomes Instrument Development Group. Pediatric Orthopaedic Society of North America. J Pediatr Orthop. 1998;18(5):561–571.
18. McCarthy ML, Silberstein CE, Atkins EA, Harryman SE, Sponseller PD, Hadley-Miller NA. Comparing reliability and validity of pediatric instruments for measuring health and well-being of children with spastic cerebral palsy. Dev Med Child
19. Boyd RN, Graham HK. Objective measurement of clinical findings in the use of botulinum toxin type A for the management of children with cerebral palsy. Eur J Neurol. 1999;6:S23–S35.
20. Bell KJ, Ounpuu S, DeLuca PA, Romness MJ. Natural progression of gait
in children with cerebral palsy. J Pediatr Orthop. 2005;22:667–682.
21. Bjornson KF, Belza B, Kartin D, Logsdon R, McLaughlin JF. Ambulatory physical activity performance in youth with cerebral palsy and youth who are developing typically. Phys Ther. 2007;87(3):248–257.