Individualized orthotic alignment and footwear for balance and mobility in children with bilateral spastic cerebral palsy: A randomized trial

Abstract

Aim:

To examine whether designed-to-be-rigid ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs (AFO-FC/IAFD) would be more effective than designed-to-be-rigid AFO with non-individualized alignment and footwear designs (AFO-FC/NAFD) in children with cerebral palsy (CP).

Method:

Nineteen children with bilateral spastic CP were randomized to AFO-FC/NAFD (n = 10) or AFO-FC/IAFD (n = 9) groups. Fifteen were male, average age 6 years 11 months (range 4 years 2 months–9 years 11 months –9.8), classified in Gross Motor Function Classification System levels II (n = 15) and III (n = 4). The Pediatric Balance Scale (PBS), Gait Outcomes Assessment List (GOAL), Patient-Reported Outcomes Measurement Information System (PROMIS), and Orthotic and Prosthetic Users’ Survey (OPUS) measures of satisfaction were collected at baseline and after 3 months’ wear.

Results:

Compared with the AFO-FC/NAFD group, those with AFO-FC/IAFD demonstrated greater change in PBS total scores (mean 12.8 [standard deviation 10.5] vs 3.5 [5.8]; p = 0.03) and GOAL total scores (3.5 [5.8] vs −0.44 [5.5]; p = 0.03). There were no significant changes in OPUS or PROMIS scores.

Interpretation:

After 3 months, individualized orthosis alignment and footwear designs had a greater positive effect on balance and parent-reported mobility than a non-individualized approach. No effect was documented for the PROMIS and OPUS. Results may inform orthotic management for ambulatory children with bilateral spastic CP.

Ambulatory children with cerebral palsy (CP) demonstrate altered lower-limb biomechanical alignment and they experience balance and activity limitations that negatively influence their mobility and ability to participate in daily life.^1,2 Ankle–foot orthoses (AFOs) are considered the standard of care to facilitate static and dynamic balance in children with CP. However, systematic reviews suggest that, in this population, the efficacy of rigid or solid AFOs with non-individualized ankle alignment of 90° and non-individualized footwear design remains equivocal or of modest effect to optimize walking,^3,4 with inconsistent effect on balance.⁵ A clinical algorithm⁶ to guide orthotic prescription suggests a patient-specific, individualized method for adjusting alignment of fixed AFOs and integrating footwear modifications (ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs; AFO-FC/IAFD). This clinical algorithm proposes that individualized attention to lower leg (shank) and thigh alignment in orthotic prescription has the potential to support therapeutic goals of improved balance and walking.

Clinically, orthotic management in children with CP is based on correcting biomechanical misalignment at the ankle and subsequent abnormal joint motion associated with balance and walking impairments.⁷ The published evidence to date has conflicting support for the efficacy or effectiveness of orthotic management. A review by Lintanf et al. suggests there is strong evidence that AFOs of varying designs improve gait speed with a small to moderate effect on balance and gross motor skills.⁵ However, in ambulatory children with CP, reviews of the evidence for efficacy of rigid AFOs aligned at 90° but with no documentation of footwear type, sole profile, and/or individual modifications remain equivocal.^3,4 This literature focuses on laboratory-based outcomes to assess the immediate effects on walking without taking into account adaptations that may occur from wear in daily life.⁷ For example, solid AFOs with no documentation of footwear type, sole profile, and/or modification have been reported to improve ankle/knee/hip kinematics compared with barefoot in children with CP in a laboratory setting only.⁸

Ambulatory children with spastic CP also demonstrate impaired balance compared with typically developing children.² Impairments of static and dynamic balance in children with CP are negatively associated with walking performance.⁹ Although AFOs are considered the standard of practice to enhance standing balance in children with CP, there are conflicting reports about their effect in children with bilateral CP.^10–13 A recent small study of individualized distal control orthoses with optimized shank-to-vertical alignment (consistent with the AFO-FC/IAFD approach) reported positive effects on standing duration, loss of balance, and center of gravity velocity in children with bilateral CP.¹⁴

The AFO-FC/IAFD approach is based on optimizing the sagittal plane orientation of the ground reaction force vector with respect to knee and hip joints during the stance phase of gait. This is achieved by individually aligning a non-articulated rigid AFO and footwear design and represents a major development in orthotic management as it accommodates ankle equinus contractures rather than forcing every ankle into a 90° alignment. Additionally, the AFO holds the midfoot in a neutral triplanar boney alignment and wedging under the AFO may be used to reorient the tibial segment with respect to vertical. Different heel, midsole, and forefoot shoe modifications are added to restore the lost ankle–foot rockers required for walking.⁶ Specifically, the AFO-FC/IAFD approach is hypothesized to improve stability by orienting the more proximal segments over the middle of the foot in standing and in single-limb support during walking.

As part of a clinical trial (National Institutes of Health R21HD094823; Clinicaltrials.gov NCT03756571), this study aimed to compare the efficacy and effectiveness of the AFO-FC/IAFD with the standard-of-practice combinations with non-individualized alignment and footwear designs (AFO-FC/NAFD) in ambulatory children with bilateral spastic CP during daily life. We assessed balance and parent-reported outcomes at baseline and after 3 months of daily wear. Parent-reported outcomes addressed community-based functional mobility, pain, gait priorities, and satisfaction with device.

METHOD

A randomized cross-over waitlist design was used, with participants randomized after consent to either AFO-FC/IAFD (intervention group) or AFO-FC/NAFD (waitlist group). All AFO-FC/IAFD were non-articulated and designed-to-be-rigid,¹⁵ with individualized ankle angles, shank-to-vertical angles, and standardized footwear with modifications. All AFO-FC/NAFD were non-articulated and designed-to-be-rigid, with ankle angles of 90°, and standardized, unmodified footwear (except to adjust for leg length discrepancy if needed). Here we report outcomes for the first phase of the study (baseline and 3-month follow-up), which is equivalent to a two-arm parallel randomized-control trial because the AFO-FC/NAFD (randomized waitlist) group had not yet crossed-over to receiving the AFO-FC/IAFD.

Participants

Study participants were approached by direct mailings and blast emails to regional care providers through a regional pediatric specialty care hospital and local site participants in the Cerebral Palsy Research Network (https://cprn.org/). Per local human participant approval, and after obtaining written informed consent from parents and assent from children with receptive language age of 7 years, ambulatory children with bilateral spastic CP were enrolled.

Inclusion criteria were: 4 to 10 years of age, bilateral impairment, classification in Gross Motor Function Classification System (GMFCS) levels II or III,¹⁶ and use of walking as a primary mode of independent mobility. Requirements for physical examination and barefoot video gait analysis were insufficient gastrocnemius length to allow knee extension with ankle dorsiflexion of 10° and an uncompromised foot arch; tone of calf muscle that did not interfere with dorsiflexion during stance; and insufficient calf muscle strength to prevent excessive dorsiflexion in stance or insufficient triplanar boney stability of the foot during stance-phase dorsiflexion.⁶ Potential participants were excluded if visual impairment limited independent mobility, if they had orthopedic/neurosurgery in the 6 months before the study or injection therapy to lower extremities in the 3 months before the study, or if they had plans for lower extremity surgery or a movement disorder medication change during the study period.

Study sample characteristics captured by parental report, medical record review, and/or direct observation by the investigators included age, sex, Hispanic ethnicity, racial group, GMFCS level, type of previous orthosis, and mother’s level of education as a proxy for socioeconomic status.¹⁷

Physical examination documented height; foot length; passive range of motion; spastic catch; hip, knee, and ankle range of motion; and supine/standing leg length. Although spastic catch (R1) was recorded, it was not used as part of the algorithm⁶ for defining the AFO-FC/IAFD intervention.

Interventions

During the first phase of the study, participants attended four study visits: baseline assessment split over two study visits (visits 1 and 2), an orthosis and gait check (visit 3), and outcome measure assessments repeated at the 3-month follow-up (visit 4). The first baseline visit included consent, randomization, video gait assessment, physical examination, orthotic prescription development, and casting for devices. The second baseline visit included fitting and delivery of orthoses, balance assessment by the study’s physical therapist, and completion of parent-reported outcomes (community-based functional mobility, pain, gait priorities, and satisfaction with device). Orthotic prescription for the AFO-FC/IAFD group was guided by the published clinical algorithms⁶ specifically, angle of the ankle in the AFO determined on the basis of combined assessment of length and stiffness of the gastrocnemius, triplanar boney alignment of the foot, and estimated risk of loss or over-lengthening of the gastrocnemius muscle.⁶ All data collection took place in an outpatient research laboratory. To ensure internal validity of interventions and address any parental concerns (e.g. skin intolerance), all orthoses were cast, fitted, and tuned by a single study orthotist with input from the study’s physical therapist who had 8 years’ experience using the published algorithms⁶ to implement the AFO-FC/IAFD intervention in clinical practice. Cascade Orthotics (Bellingham, WA, USA) fabricated all orthoses. Standardized shoes (DC High-top Skate shoes) were provided to minimize the challenge of finding shoes to fit orthoses and to improve internal validity (https://www.dcshoes.com/kids).

For participants randomized to receiving the AFO-FC/IAFD first, the second baseline visit included fitting and tuning of new orthoses and shoes that were modified according to the AFO-FC/IAFD algorithm.⁶ Shoes were modified either internally or externally to optimize shank-to-vertical angle in static standing and timing of heel off at terminal stance⁶ (see Figure 1a and Table 1 for device/footwear prescriptions/modifications). For participants randomized to receiving the AFO-FC/NAFD first, the second baseline visit consisted of fitting a new AFO-FC/NAFD with unmodified shoes (except to adjust for leg length discrepancy if needed) (Figure 1b and Table 1 for device/footwear prescriptions/modifications). Where a leg length discrepancy was identified for any participant, it was addressed by internal and/or external full shoe lifts. If foot length was short for the child’s height, this was also addressed.¹⁸ Both groups were instructed to wear the devices for at least 8 hours per day.

FIGURE 1.

Example of bench alignment and shoe modifications for (a) the designed-to-be-rigid ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs (AFO-FC/IAFD), example of a rigid flat sole profile with no rocker and plain heel; and (b) the designed-to-be-rigid ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs (AFO-FC/NAFD), example of a flexible sole, rounded rocker, and plain heel.

TABLE 1.

Baseline clinical presentation and ankle–foot orthosis and footwear prescription for individual participants by group

Participant	Treatment group	Shank anglea	Deviceb	AAAFO (degrees)	SVA alignment (degrees)	MTPJ	LLD	Shoes	Sole (stiffness/profile/length)
1	AFO-FC/IAFD	Insufficient	RAFO-B	10 plantarflexed	11 incline	Fixed	Y-int	Adapted	Stiff, flat, PLR 85%, PH
2	AFO-FC/IAFD	Insufficient	RAFO-B	5 plantarflexed	12 incline	Fixed	Y-int	Adapted	Stiff, flat, PLR 85%, PH
3	AFO-FC/IAFD	Insufficient	RAFO-B	5 plantarflexed	11 incline	Fixed		Adapted	Stiff, flat, PLR 85%, PH
4	AFO-FC/IAFD	Excessive	RAFO-C	10 plantarflexed	15 incline	Fixed	Y-ext	Adapted	Stiff, flat, 100%, PH
5	AFO-FC/IAFD	Insufficient	RAFO-B	10 plantarflexed	15 incline	Fixed		Adapted	Stiff, flat, 100%, PH
6	AFO-FC/IAFD	Insufficient	RAFO-B	5 plantarflexed	12 incline	Fixed		Adapted	Stiff, flat, 100%, PH
7	AFO-FC/NAFD	Insufficient	RAFO-A	90 neutral ankle	5 incline	Free	Y-int	Not adapted	Flexible, rounded, PH
8	AFO-FC/NAFD	L, insufficient; R, excessive	RAFO-A	90 neutral ankle	L, 3 incline; R, 5 incline	Free		Not adapted	Flexible, rounded, PH
9	AFO-FC/NAFD	Insufficient	RAFO-A	90 neutral ankle	5 incline	Free		Not adapted	Flexible, rounded, PH
10	AFO-FC/NAFD	Insufficient	RAFO-A	90 neutral ankle	4 incline	Free		Not adapted	Flexible, rounded, PH
11	AFO-FC/IAFD	Insufficient	RAFO-B	10 plantarflexed	12 incline	Fixed		Adapted	Stiff, flat, PLR 90%, PH
12	AFO-FC/NAFD	Excessive	RAFO-A	90 neutral ankle	6 incline	Free	Y-int	Not adapted	Flexible, rounded, PH
13	AFO-FC/NAFD	L, excessive, R, insufficient	RAFO-A	90 neutral ankle	L, 6 incline; R, 4 incline	Free	Y-int	Not adapted	Flexible, rounded, PH
14	AFO-FC/NAFD	Excessive	RAFO-A	90 neutral ankle	6 incline	Free		Not adapted	Flexible, rounded, PH
15	AFO-FC/NAFD	L, excessive, R, insufficient	RAFO-A	90 neutral ankle	L, 6 incline; R, 4 incline	Free	Y-int	Not adapted	Flexible, rounded, PH
16	AFO-FC/NAFD	Excessive	RAFO-A	90 neutral ankle	6 incline	Free	Y-int	Not adapted	Flexible, rounded, PH
17	AFO-FC/IAFD	Excessive	RAFO-C	10 plantarflexed	15 incline	Fixed		Adapted	Stiff, flat, 100%, PH
18	AFO-FC/IAFD	Insufficient	RAFO-B	5 plantarflexed	12 incline	Fixed	Y-ext	Adapted	Stiff, flat, PLR 90%, PH
19	AFO-FC/NAFD	Insufficient	RAFO-A	90 neutral ankle	4 incline	Free		Not adapted	Flexible, rounded, PH

^aShank-to-vertical alignment barefoot at temporal midstance observed with video gait analysis.

^bDevice per AFO-FC algorithm.

Abbreviations: AAAFO, angle of the ankle in the ankle–foot orthosis; AFO, ankle–foot orthosis; AFO-FC/IAFD, ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs; AFO-FC/NAFD, ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs; LLD, leg length difference (corrected, yes/no [Y/N]; int, internal to shoe; ext, external to shoe sole); MTPJ, metatarsal phalangeal joints; PH, plain heel; PLR, point loading rocker; RAFO-A, designed-to-be-rigid AFO with trim lines proximal to metatarsal phalangeal joints; RAFO-B, designed-to-be-rigid ankle AFO with trim lines distal to metatarsal phalangeal joints; RAFO-C, designed-to-be-rigid ground reaction solid ankle AFO with trim lines distal to metatarsal phalangeal joints; SVA, shank-to-vertical angle.

A video vector gait assessment combined with two-dimensional video was done 2 weeks after initial fitting to adjust and confirm that a clinically optimal AFO-FC/IAFD was obtained. Optimal AFO-FC/IAFD alignment was defined as thigh alignment in static standing as close to vertical as active and passive hip and knee range of motion allowed (i.e. the child could attain and maintain this position spontaneously, and approximation of this alignment was observed during midstance of gait).⁶ The AFO-FC/NAFD group also had this visit to confirm fit/tolerance of the orthosis, but no video vector assessment was conducted.

Outcomes

Balance was assessed in the laboratory by an unblinded physical therapist using the Pediatric Balance Scale (PBS). Community-based functional mobility, pain, and gait priorities were assessed by parental report using the Gait Outcomes Assessment List (GOAL). Physical function and pain interference were assessed by parental report using the Patient-Reported Outcomes Measurement Information System (PROMIS) Parent Proxy Profile 49 survey. Satisfaction with device was assessed by parental report using the Orthotic and Prosthetic Users’ Survey (OPUS).

The PBS is a performance-based outcome measure intended to assess balance in a clinical setting for children with neurological impairment.¹⁹ Test–retest reliability²⁰ has been established with pediatric norms,²¹ allowing discrimination between functional levels of children with CP.²² The minimal clinically important difference (MCID) for the PBS total score for children with CP ranged from 3.66 to 5.83.¹⁹ The PBS static subtest is made up of six items (e.g. standing unsupported, standing on one foot), while the dynamic subtest comprises eight items (e.g. sit to stand, placing foot on stool); both have an MCID of 2.9.

The GOAL is a parent- or patient-reported outcome measure designed to evaluate gait priorities and functional mobility in the community for ambulant children with CP.²³ The GOAL consists of 48 items grouped into seven domains: (1) activities of daily living and independence; (2) gait function and mobility; (3) pain, discomfort, and fatigue; (4) physical activities, sports, and recreation; (5) gait pattern and appearance; (6) use of braces and mobility aids; and (7) body image and self-esteem. Task-oriented GOAL domains were scored on a 7-point scale (0–6), including a 4-point scale (0–3) documenting the amount of assistance/type of device used. The domains capturing symptoms were rated on a 6-point scale, quantifying the frequency of the symptom, including a 4-point scale quantifying the intensity of the symptom. The scores are additive to generate the domain score. The importance of each item within a domain as a goal for improvement for the child/parent is captured on a 5-point scale from ‘not a goal’ to ‘extremely important’.

The PROMIS item bank was developed by the National Institutes of Health²⁴ from patient-reported outcome measures that indicate the child’s wellbeing or suffering, as well as their ability or lack of ability to function. The physical function mobility, pain intensity, and pain interference subscales were captured for this study through the Parent Proxy Profile 49 survey. Physical function mobility was rated on a 5-point scale of ‘no trouble’ to ‘unable to do’, and pain interference was rated on a 5-point scale of ‘never’ to ‘almost always’ with pain intensity rated on a 10-point scale of ‘no pain’ to ‘worst pain you can think of’.

The OPUS is a collection of patient-reported outcome measures developed specifically for users of prosthetic and orthotic devices to assess lower extremity functional status, health-related quality of life, and satisfaction with device and service.²⁵ Only the 21-item satisfaction with device module was used in this study with statements of satisfaction (e.g. ‘my orthosis fits well’) rated on a 5-point scale from ‘strongly agree’ to ‘strongly disagree’. Previous assessment of children with the OPUS used parental report.²⁵ Since the OPUS only has normative validation for adults, the raw scores were reported and analyzed.

To capture issues of skin/discomfort and device tolerance that are common with orthotic management, adverse events were assessed throughout the protocol and coded for relationship (direct or possible) to the study and severity. ‘Mild’ severity was defined as inconvenient to daily life, ‘moderate’ required medical attention, and a ‘severe’ event was life threatening or required hospitalization.

Statistical analysis

Eligible participants were randomly assigned by the study biostatistician with 1:1 allocation to AFO-FC/IAFD and AFO-FC/NAFD arms by a random number generator. We conducted intent-to-treat analysis using all randomized participants who provided any evaluable data. Continuous variables were summarized by presenting the means, standard deviations (SD), and interquartile ranges for the participants who contributed values. Categorical variables were summarized using frequencies and proportions. Because of the relatively small sample size, we chose non-parametric statistical tests for comparisons. Baseline demographics and outcome values were compared between study arms using Fisher’s exact test on categorical variables and a Wilcoxon rank sum test on continuous variables. For the main analysis, we calculated changes from baseline to 3 months for the outcomes, then compared the change scores between study arms using a Wilcoxon rank sum test. All analyses were performed using R statistical software version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria); p-values less than 0.05 indicate statistically significant differences.

RESULTS

Of the 46 participants approached, 21 enrolled with 19 completing the protocol (Figure S1). Seven of the 19 participants were in protocol during COVID-19 restrictions (March–August 2020). Two children withdrew shortly after device delivery: one child did not tolerate wear of the assigned device (AFO-FC/NAFD) and a second (AFO-FC/IAFD) because of difficulty of shoe donning. Of the 19 children completing the protocol, 10 randomized to the AFO-FC/NAFD group and nine to the AFO-FC/IAFD (Figure S1 and Tables 1 and 2) between December 2018 and September 2020. Fifteen participants were male, average age 6 years 11 months (range 4 years 2 months–9 years 11 months), classified in GMFCS levels II (n = 15) and III (n = 4). The study arms were balanced with respect to GMFCS levels, sex, Hispanic ethnicity, racial group, maternal education, and type of orthosis used immediately before study participation (p = 0.40–1.0). Table 3 shows the average scores for all outcome measures. There were no significant differences in baseline outcomes between treatment groups. Participants in both treatment groups had leg length discrepancies addressed as needed¹⁸ and none required optimization of foot length to height. Before enrolling in the study, five participants wore AFO-FC/IAFD that were exactly the same as the study prescriptions, including the footwear (Table 1). Of the five participants who wore AFO-FC/NAFD before the study, all had the same devices as the protocol (Table 1) and all wore Billy high-top tennis shoes (https://billyfootwear.com/products/grey-jersey-billy-classic-lace-highs) with rounded rockers at 85% shoe length.

TABLE 2.

Characteristics of total sample and groups (AFO-FC/NAFD or AFO-FC/IAFD)

Characteristic	Total (n = 19)	AFO-FC/NAFD (n = 10)	AFO-FC/IAFD (n = 9)	p a
Mean age (SD); median (IQR), years:months	6:11 (1:11); 7:1 (5:5–8:7)	6:7 (2:4); 5:7 (4:11–8:11)	7:4 (1:6); 7:2 (6:5–7:10)	0.5
Female sex, n (%)	4 (21)	2 (20)	2 (22)	0.9
GMFCS level, n (%)
II	15 (79)	8 (80)	7 (78)	0.9
III	4 (21)	2 (20)	2 (22)
Hispanic ethnicity, n (%)	1 (5.3)	0	1 (11)	0.5
Racial group, n (%)
White	10 (53)	4 (40)	6 (67)	0.6
Black	3 (16)	1 (10)	2 (22)
Asian	4 (21)	3 (30)	1 (11)
Other	2 (11)	2 (20)	0
Maternal education, n (%)
Four-year college or lower	10 (53)	7 (70)	3 (33)	0.4
Postgraduate	9 (47)	3 (30)	6 (67)
Previous orthosis type, n (%)
None	4 (21)	2 (20)	2 (22)	>0.99
AFO-FC/NAFD	5 (26)	2 (20)	3 (33)
Hinged AFO	5 (26)	2 (20)	3 (33)
AFO-FC/IAFD	5 (26)	3 (30)	2 (22)
Etiology, n (%)
Preterm birth	8 (42)	5 (50)	3 (33)	0.9
Hypoxic–ischemic encephalopathy	9 (47)	5 (50)	4 (44)
Genetic	1 (5)	0	1 (11)
Multiple birth	1 (5)	0	1 (11)

^aFisher’s exact test; Wilcoxon rank sum test.

Abbreviations: AFO, ankle–foot orthosis; AFO-FC/IAFD, ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs; AFO-FC/NAFD, ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs; GMFCS, Gross Motor Function Classification System; IQR, interquartile range; SD, standard deviation.

TABLE 3.

Change in score between baseline and 3-month follow-up for balance and parent-reported outcomes by group (AFO-FC/NAFD and AFO-FC/IAFD)

	AFO-FC/IAFD Baseline Mean (SD) n = 9	AFO-FC/NAFD Baseline Mean (SD) n = 10	p a	AFO-FC/IAFD Follow-up Mean change (SD)	AFO-FC/NAFD Follow-up Mean change (SD)	p a
Pediatric Balance Scale: total	32.8 (10.6)	33.4 (15.4)	0.69	15.6 (6.5)	4.3 (5.5)	0.004
Pediatric Balance Scale: static subscale	14.2 (5.2)	14.6 (6.2)	0.82	5.1 (6.9)	0.6 (3.2)	0.04
Pediatric Balance Scale: dynamic subscale	18.6 (6.3)	18.0 (8.3)	>0.99	8.8 (3.8)	3.5 (4.2)	0.009
GOAL: total score (T scores)	60.3 (6.7)	57.6 (15.5)	0.90	7.9 (9.2)	−0.4 (5.5)	0.03
GOAL: gait function and mobility (T scores)	64.8 (10.9)	60.0 (27.2)	0.84	3.2 (8.2)	−2.6 (5.6)	0.16
GOAL: physical activity, sports/recreation	40.6 (8.4)	39.5 (19.4)	0.93	11.7 (16.9)	1.4 (12.8)	0.22
GOAL: gait pattern and appearance	44.8 (11.8)	44.2 (17.2)	0.97	13.4 (13.6)	2.0 (12.8)	0.08

^aBased on Wilcoxon rank sum test: bold type indicates significant difference in change over time between groups at p < 0.05.

Abbreviations: AFO, ankle–foot orthosis; AFO-FC/IAFD, ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs; AFO-FC/NAFD, ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs; GOAL: Gait Outcomes Assessment List; SD, standard deviation.

After 3 months of wear, the AFO-FC/IAFD group had significantly greater improvement in balance than the AFO-FC/NAFD group assessed using the PBS total score (Table 3 and Figure 2), as well as the static (p = 0.042) and dynamic (p = 0.009) subscales. The changes in GOAL total score were significantly different between groups after 3 months of wear (p = 0.03): the AFO-FC/IAFD group GOAL total score increased by 7.9 (SD 9.2), while the AFO-FC/NAFD group GOAL total score decreased by 0.44 (SD 5.5). There were no significant differences documented between groups for the remainder of the GOAL subscales, PROMIS physical function, pain intensity, pain interference, or OPUS satisfaction with device (p = 0.08–0.74).

FIGURE 2.

Pediatric Balance Scale (PBS) total scores at baseline and after 3 months of wear for the ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs (AFO-FC/IAFD) and ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs (AFO-FC/NAFD) groups. Box, top/bottom interquartile range; whiskers, minimum/maximum data range; triangles, mean; black dots, individual participants’ data.

There were no severe adverse events. The AFO-FC/NAFD group had a total of 13 adverse events (five mild, eight moderate) and the AFO-FC/IAFD group had a total of 14 adverse events (10 mild, four moderate) (p = 0.13). The adverse events were all skin issues related to device fit (i.e. redness lasting more than 20 minutes after doffing). All parents reported device/shoe wear time of at least 8 hours at the 3-month follow-up.

DISCUSSION

This study compared the efficacy and effectiveness of AFO-FC/IAFD with AFO-FC/NAFD in ambulatory children with bilateral spastic CP after 3 months of wear in daily life. The results indicate that the AFO-FC/IAFD approach had a greater positive effect on balance over 3 months of wear than the AFO-FC/NAFD (mean 12.8, SD 10.5). For the AFO-FC/IAFD approach, the change in PBS total score was twice that of the PBS MCID of 5.8.¹⁹ In contrast, the change in PBS total score for the AFO-FC/NAFD (mean 3.5, SD 5.8) did not exceed the MCID. Our results also suggest that, after 3 months of wear, the AFO-FC/IAFD approach had a significantly greater positive effect on parent-reported community mobility/gait priorities (as assessed by the GOAL) than the AFO-FC/NAFD. While the MCID for the GOAL total or domain scores were not known, increases in GOAL total scores were greater than the mean absolute differences of 5.0 for test–retest of the GOAL, suggesting true change.²³ For all other assessments (i.e. tolerance/skin issues, parent-reported change in functional mobility, pain intensity, pain interference, and satisfaction) there were no significant differences between groups at the 3-month follow-up.

Exploring the static and dynamic subscales of the PBS provides some additional insights into the effect of these orthotic approaches on balance. For the PBS static subscale, the change in score over 3 months of wear in the AFO-FC/IAFD group exceeded the published MCID of 2.9, while the AFO-FC/NAFD group did not exceed the MCID.¹⁹ Static subscale items with the greatest change in the AFO-FC/IAFD group were for tasks of standing with feet together and one foot in front of the other. This is consistent with the work of Liao et al.⁹ who documented a positive relation between standing on one foot and with one foot in front of the other and gross motor skills. For the PBS dynamic subscale, the change in scores over 3 months of wear in both groups exceeded the MCID, with the AFO-FC/IAFD group change greater than twice that of the MCID. Dynamic subscale items with the greatest change for the AFO-FC/IAFD group were sit/stand, stand/sit, transfers, and turning to look behind. We attribute the large effect of the AFO-FC/IAFD approach on balance to the emphasis placed at arriving at an individualized alignment that ensures stability in standing and midstance of walking.

As a limitation, this study was not designed to assess whether the orthoses were optimal for any specific participant given that we compared change over time in independent groups for a single intervention. However, our data do show that the AFO-FC/IAFD group all had shank-to-vertical angles greater than 10° (Table 1) to bring the thigh segment to near vertical at temporal midstance.⁶ In contrast, the AFO-FC/NAFD group had shank-to-vertical angles that were less than the 8° to 12° needed to achieve normal thigh segment alignment at temporal midstance¹⁸ (Table 1).

One study reported increased burden and negative perceptions of shoe weight, size, and appearance for the AFO-FC/IAFD approach.²⁶ Interestingly, the results of this study found no significant difference between groups in parental report of satisfaction with devices/shoes after 3 months of wear, and adverse events were also no different between groups over the study period, suggesting similar tolerance.

Orthoses are one of the top 16 research priority areas for persons with CP reported by an engagement group made up of those living with CP, caregivers, and providers.²⁷ From a broad clinical perspective, this work suggests that these outcomes may be helpful to consider along with clinical (e.g. walking speed, gait analysis) and other quantitative community outcomes (e.g. pedometers, accelerometers, global positioning system data loggers). In particular, patient-/parent-reported perspectives on AFOs for persons with CP are limited. Recently, four common themes around the lived experiences of persons with CP and their parents/caregivers about orthosis use, priorities, and access were documented: (1) AFOs can benefit mobility and independence; (2) more information is needed during device provision; (3) acquiring devices is a lengthy and confusing process; and (4) AFOs can be uncomfortable and difficult to use.²⁸ Similarly, caregivers of children with CP noted that AFOs improved mobility for their children, yet also desired improved comfort and individualization (e.g. colors of device/straps).²⁹

Limitations of this study need to be considered for interpretation and clinical translation. First, the sample size was small and predominately male, but the skew was greater than prevalence data for males of approximately 60% in the CP population.³⁰ Second, quantitative documentation of actual wear time for the devices/shoes was not captured, hence differences in results between groups may have been related to wear time. Future work could potentially use sensors to capture wear time and eliminate it as a confounding variable. Third, the orthoses worn before participation in this study were not controlled and may have biased the treatment effects of either intervention. Also, there were restrictions in opportunities for daily mobility for over one-third of the participants because of the COVID-19 pandemic, and local public health mandates may have influenced parental perceptions of mobility. These limitations in mobility opportunities may also have influenced device wear time and hence the potential motor-learning effect of devices. Finally, the study’s physical therapist who administered the PBS was not blinded and hence there is the possibility of bias.

CONCLUSIONS

There seems to be a greater positive influence on clinic-based balance and parental report of functional mobility and overall gait priorities when AFOs designed-to-be-rigid and footwear are individually aligned/modified and worn for 3 months compared with when AFOs designed-to-be-rigid have the same default alignment and no footwear modifications. Although promising, further study is warranted with a larger sample size and outside the constraints of a pandemic. Inclusion of parent-reported outcomes during orthotic management in ambulatory children with bilateral spastic CP seems feasible and meaningful.

Supplementary Material

Supinfo

Figure S1: CONSORT enrollment flow diagram.

What this paper adds.

• Balance and parent-reported mobility increased more over time for the ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs (AFO-FC/IAFD) group.

• Changes in balance over time suggest a therapeutic effect of the AFO-FC/IAFD approach.

Abbreviations:

AFO

ankle–foot orthosis

AFO-FC/IAFD

ankle–foot orthoses and footwear combinations with individualized alignment and footwear designs

AFO-FC/NAFD

ankle–foot orthoses and footwear combinations with non-individualized alignment and footwear designs

GOAL

Gait Outcomes Assessment List

MCID

minimal clinically important difference

OPUS

Orthotic and Prosthetic Users’ Survey

PBS

Pediatric Balance Scale

PROMIS

Patient-Reported Outcomes Measurement Information System