Gait Assessment of Fixed Ankle-Foot Orthoses in Children ... · Gait Assessment of Fixed Ankle-Foot...

126

Gait Assessment of Fixed Ankle-Foot Orthoses in Children With Spastic Diplegia Mark F. Abel, MD, Gregory A. Juhl, Christopher L. Vaughan, PhD, Diane L. Damiano, PhD, PT

ABSTRACT. Abel MF, Juhl GA, Vaughan CL, Damiano DL. Gait assessment of fixed ankle-foot orthoses in children with spastic diplegia. Arch Phys Med Rehabil 1998;79:126-33.

Objective: To evaluate the effectiveness of ankle-foot orthoses (AFOs) in spastic diplegic cerebral palsy patients for whom orthoses were indicated to control equinus or pes planovalgus deformities.

Design: A retrospective, cross-sectional assessment was performed on diplegic subjects who had suitable barefoot and AFO gait trials on the same day.

Patients: Thirty-five subjects with a mean age of 8.7yrs were included. Eighteen wore braces to control equinus and 17 to control pes planovalgus and crouch.

Outcome Measures: Gait data assessed in all subjects included temporal-distance factors and sagittal kinematics. Force plate data to determine joint moments and powers were obtained in 20. Repeated measures analysis of variance was used to compare across conditions and indications.

Results: The cohort demonstrated increased velocity (lOcm/ set; p < .OOl), stride length (1Ocm; p < .OOl), and percent single-limb support (1.8%; p < .002) using AFOs compared with barefoot gait. In braces, ankle excursion was reduced (p < .OOOl), while pelvic, hip, and knee excursions were increased to account for the temporal changes (p < .009). Effects were similar in both indication groups. In neither indication group did the AFO significantly alter knee position in stance. Kinetic analysis showed a reduction of abnormal power burst (p CC .05) in early stance and an increase in late stance ankle moment (p < .05) with AFOs. Differences in gait characteristics and bracing effects are shown for both indication groups.

Conclusion: Compared with barefoot gait, AFOs enhanced gait function in diplegic subjects. Benefits resulted from elimination of premature plantar flexion and improved progression of foot contact during stance. Effects on proximal joint alignment were not significant.

0 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabili- tation

c EREBRAL PALSY (CP) is the most common motor disor- der originating in childhood.’ Classification of patients with

spastic CP is based on the body region involved.* Those with spastic diplegia have greater lower extremity than upper extremity involvement and eventually become ambulatory. Because

From the Department of Orthopaedics, Motion Analysis Laboratory, Kluge Children’s Rehabilitation Center, University of Virginia, Charlottesville (Dr. Abel, Mr. Juhl, Dr. Damiano); and the Department of Biomedical Engineering, Univer- sity of Cape Town Medical School, Observatory, Cape, South Africa (Dr. Vaughan).

Submitted for publication December 16, 1997. Accepted in revised form June 3, 1997.

No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.

Reprint requests to Mark F. Abel, MD, Kluge Children’s Rehabilitation Center, 2270 Ivy Road, Charlottesville, VA 22903.

0 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

0003-9993/98/7902-4304$3.00/O

Arch Phys Med Rehabil Vol79, February 1998

of the improved survival rate of very-low-birth-weight infants, spastic diplegia is becoming more prevalent and now accounts for approximately 50% of the total CP population.’

In diplegia, motor deficits and spasticity typically produce a walking pattern characterized by an equinus ankle position at floor contact (fig lA), exaggerated stance phase knee flexion (crouch), and increased hip adduction and internal rotation.4.4 Persistent equinus and crouch are often associated with pes planovalgus deformities (fig 1B). This foot deformity theoreti- cally reduces stability because the ground reaction force shifts posteriorly and laterally, increasing the flexion, valgus, and external rotation moments (external moments) at the knee.6,7 These gait abnormalities are present in proportion to the severity of neurologic involvementss

Commonly prescribed treatments to address these gait abnormalities include physical therapy, surgery (orthopedic and rhi- zotomy), and braces.” The orthotic approach is conservative; therefore, its application is particularly widespread.” In general, braces are prescribed to prevent deformity and control joint position and, thus, improve stability. In the context of improving the gait of patients with spastic diplegia, the ankle-foot orthosis (AFO) (with a rigid ankle) is designed to prevent or eliminate an equinus position; while the AFO may have a direct effect only on foot and ankle joint alignment, this distal control is believed to exert a positive effect on more proximal joints as well.” For example, by controlling the position of the shank and restraining anterior tibia1 translation, the AFO may reduce midstance knee flexion or crouch.4,6

The effects of AFOs on walking function have been studied using both simple and more sophisticated gait analysis tech- niques. Using video technology alone, Powell and colleagues” showed that AFOs had no significant effect on cadence, velocity, or stride length. Thomas and othersI used kinematic and electromyographic information to evaluate 17 children with CP walking barefoot and in AFOs. They found an improvement in ankle motion for all children and an increase in hip and knee motions for at least 80% of the patients.

Condie and Meadows” compared normal pediatric gait dynamics with gait dynamics in children who have spastic diplegia with and without AFOs. They demonstrated that the use of an appropriate AFO in CP reduced the high-impact forces seen in early stance with CP, while the vertical reaction forces in late stance were increased. They concluded that the AFOs improved ability to support body weight and to generate push-off.

While there is limited evidence to support the general effectiveness of AFOs for improving gait in cerebral palsy, clearly not all patients benefit and the specific clinical indications for their use are poorly specified.‘4-‘6 The purpose of this investigation was to evaluate the effects of AFOs versus barefoot gait in 35 patients with spastic diplegia who were prescribed orthoses for one of two primary indications: to control equinus or to control pes planovalgus. The biomechanical effects of bracing were assessed by kinematic and kinetic analyses. Temporal- distance parameters of gait have been shown to correlate with overall gross motor function: therefore, changes in temporal- distance factors were considered functional outcome measures.9 The assumptions were that the AFOs would (1) enhance gait

GAIT ASSESSMENT OF FIXED ANKLE-FOOT ORTHOSES, Abel

Fig 1. Patients with spastic diplegia showing two different orthotic indications. (A) Child with a dynamic ankle equinus pattern. (Bl Child with a pes planovalgus deformity.

function and stability in both clinical groups as reflected by increased single limb stance time, velocity, and stride length when compared with barefoot gait and (2) reduce stance phase knee flexion in the group with pes planovalgus by controlling foot and ankle position.

METHODS

Subjects A search of the Motion Analysis Laboratory database yielded

35 subjects (mean age 8.7yrs, range 2.5 to 19yrs) since 1993 who met the following criteria: a diagnosis of spastic diplegia; use of fixed, bilateral AFOs; and analyses of gait barefoot and with AFOs on the same day. Furthermore, any subject who had a history of orthopedic surgery was at least 11 months postoperative. The subjects had passive forefoot dorsiflexion with the knee in extension to achieve neutral position (zero). The AFOs were polypropylene, custom molded, with the foot and ankle in several degrees of dorsiflexion and brace trim lines anterior to the malleoli.

The sample was further subdivided into one of two orthotic indication groups. The first group consisted of 18 children who were braced to control equinus, and the second group consisted of 17 who presented with planovalgus foot deformities. Of the 18 children in the equinus group, 12 were independent community ambulators and 6 were limited community ambulators who used assistive devices to ambulate when outside the home.*s9 In contrast, the 17 children in the pes planovalgus group included 15 limited community ambulators and only 2 independent ambulators.

Procedures The following assessment protocol was followed for all sub-

jects. Three-dimensional gait data were collected with an Ex- pertvision system,” or the Vicon 370 system.b All temporal and

127

kinematic data were processed using Vicon Clinical Manager software (version 1.21).b Fifteen reflective markers were placed on specific anatomic landmarks and six cameras recorded (at 60Hz) the three-dimensional spatial location of each marker as the subject walked at his/her freely chosen speed.17 Two juxtaposed force plates” were concealed in the walking path to quantify forces coincident with joint positions. Generally, a total of 10 to 20 walking trials was performed during data collection, with barefoot walks preceding AFO walks. A minimum of three gait cycles (mean of 5, range 3 to 10) were analyzed, with data for the left and right sides averaged. Problems of fatigue were not noted with any of the patients; however, subjects were al- lowed to rest as needed between trials.

The variables studied were temporal-distance factors (velocity, stride length, cadence, and stance times) and sagittal plane kinematics (joint excursions) at the pelvis, hip, knee, and ankle. For ankle motion, negative values signify plantarflexion and positive values dorsiflexion. Sagittal plane joint moments were also obtained in 20 subjects. Use of assistive devices or inade- quate step lengths interfered with collection of kinetic data in the remaining subjects. Moments are reported as internal, muscle- generated moments to counter the external joint moments cre- ated by the ground reaction forces.6 Repeated measures analysis of variance (ANOVA) procedures were performed using SPSS for Windowsd to compare data across conditions (barefoot and AFO) and across indications (equinus versus planovalgus). For intergroup comparisons, velocity and stride length were normalized by dividing them by the respective leg lengths. A p < .05 was set as the cut-off for statistical significance.

The gait studies for this investigation were obtained in the course of routine clinical evaluations. Our laboratory protocol received approval from the Investigational Review Board of the institution and all subjects gave their signed consent. Normal values for comparisons were obtained from our laboratory data base of age-matched children.

RESULTS

Use of an AFO during gait, as compared with the barefoot condition, produced an increase in velocity (p < .OOl), stride length @ < .OOl), and single support time (p < .002), a decrease in double support time (p < .OOl), and no change in cadence (table 1). These changes in temporal parameters were accompa- nied by increased excursions at the pelvis, hip, and knee, while motion at the ankle decreased. As expected, the ankle position was more dorsiflexed at floor contact and plantarflexion (maximum ankle plantarflexion) was restricted at push-off with AFOs. At the knee, maximum flexion in swing was increased but no change was noted in maximum knee extension in stance or the flexed knee position at floor contact in the AFOs.

When comparing the barefoot walks of the two bracing indication groups, gait characteristics were shown to be quite different (table 2). The planovalgus group had a longer double limb stance time, a shorter normalized stride length, and a lower cadence than the equinus group. Joint excursions were also more restricted at the hip and knee in the planovalgus group. Those with planovalgus had hyperdorsiflexion in midstance (mean of 19.9”) and less plantarflexion at push-off as compared with the equinus group. Despite their differences in the barefoot baseline condition, similar improvements at- tributable to the AFOs were seen in both groups (fig 2). How- ever, the planovalgus group in particular showed improved stability as single limb stance increased while double limb support and total stance decreased significantly. In neither group did the AFOs alter knee position at foot contact or in midstance.


128 GAIT ASSESSMENT OF FIXED ANKLE-FOOT ORTHOSES, Abel

Table 1: Gait Parameters

Parameters Barefoot AFO

(Mean e SD) (Mean 2 SDJ FVdlJe p Value

Temporal Factors Velocity (m/set) .72 2 .30 0.82 + .32 16.0 .OOOl Stride length (ml .69 ? .22 0.79 2 .25 37.5 .OOOl Cadence tstep/min) 123.5 + 32 121.2 t- 31 1.19 ,283 Single stance (%) 33.2% -c 6.2 35.0% + 5.0 11.9 .002 Double stance 1%) 33.9% f 12 30.0% t 10 13.8 ,001 Stance 1%) 67.0% + 6.2 65.0% 2 5.1 13.9 .OOl

Joint Excursions f”) Pelvic excursion 6.96 2 2.7 8.13 + 2.9 7.59 .009 Hip excursion 40.8 ? 9.7 45.2 c 9.5 30.8 .OOOl Knee excursion 36.8 c 13 42.0 2 13 13.4 .OOl Ankle excursion 25.8 -c 15 13.7 + 8.5 57.5 .OOOl

Joint Angles (“1 Knee flexion at contact 36.1 ? 15 34.7 2 15 .77 ,386 Knee maximum flexion 56.3 + 13 59.9 2 13 5.98 ,020 Knee maximum extension* -19.5 -c 14 -17.3 c 14 2.87 ,100 Ankle position at contact 1.26 k 12 6.91 -c 7.9 7.92 ,008 Ankle maximum dorsiflexion’ 14.3 + 13 15.7 2 8.2 .62 .438 Ankle maximum plantarflexion’ -9.05 2 15 2.6 k 7.4 75.9 .OOOl

n = 35. Mean age, 9.76 t 4.37yrs (range, 2.5 to 19). * Minus denotes the degree from full extension or the residual knee flexion in midstance. ’ Minus denotes the movements in the plantarflexion range.

Kinetic data were collected on only 7 patients with planovalgus and 13 patients with equinus, so the reduced sample made intragroup and intergroup comparisons problematic (table 3). In general, sagittal ankle moments were higher in the equinus group than the planovalgus group. Compared with barefoot gait, the AFOs did not reduce the abnormally increased ankle plan- tarflexor moment in early stance, but the ankle power generation at this point in the cycle was reduced. In late stance, the ankle plantarflexion moment was slightly higher in the AFO while power generation was reduced.

Ensemble average curves for both groups demonstrate the kinematic and kinetic changes graphically (fig 3). The relation of the ground reaction force vector to the joints is shown for two representative patients: one with equinus pattern (fig 4A) and one with pes planovalgus pattern (fig 4B).

DISCUSSION In diplegic CP, spasticity, abnormal muscle activation pat-

terns, and muscle tightness disrupt the normal interaction of the

ankle and foot with the supporting surface. The patients with equinus have a forefoot strike that is reflected by an increase in the first peak vertical force or an increase in the early plantar flexion moment.‘X.‘9 Increased knee flexion at foot strike com- pensates for the equinus by keeping the body center of mass over the base of support; the forefoot (figs 3A, 4A). With weight acceptance, the ankle and foot move in the direction of dorsiflexion, but this movement is interrupted by premature plantarflexion as reflected by the power generation spike in early stance. Plantarflexion persists through the remainder of stance, and the late plantar flexion moment and power generation are reduced. In effect, heel strike, early to midstance ankle dorsiflexion, and metatarsal roll-off are eliminated, resulting in a reduction in stride length.‘,” The premature ankle power burst results in an exaggerated vertical oscillation of the body center of mass rather than the normal, low-amplitude sinusoidal excursion. Consequently, the gait of the diplegic person with equinus is less energy efficient.“.‘8*20

Diplegic patients with pes planovalgus deformities have a

Table 2: Barefoot Gait Characteristics by Indication

Parameters

Temporal Factors Normalized* velocity (m/set/m) Normalized stride length (m/m) Cadence fsteps/min) Single stance (%) Double stance f%) Stance (%)

Joint Excursions (degrees) Pelvic excursion Hip excursion Knee excursion Ankle excursion

Joint Angles (degrees) Knee flexion at contact Knee maximum flexion Knee maximum extension Ankle position at contact Ankle maximum dorsiflexion Ankle maximum plantarflexion

Equinus (Mean f SD)

1.57 f .35 1.33 k .16

140.4 i- 22 36.8 2 3.4 26.8 + 6.7 63.6 -t 3.5

6.67 2 2.9 45.1 ? 7.1 43.8 2 11 22.8 ? 9.5

34.4 t 14 60.1 f 12

-16.3 + 12 -.74 c 11 6.13 + 12

-13.5 + 15

Pes Phlovalgus (Mean 2 SD)

.83 + .38

.91 ? .26 105.6 -c 32

29.3 k 6.3 41.3 c 13 70.6 k 6.6

7.26 +- 2.4 36.2 2 IO 29.3 2 12 28.9 + 19

34.9 + 15 50.6 ? 13

-22.2 2 15 3.54 f 13 19.9 2 11

-3.42 2 11

F Value p Value

31.0 ,001 26.7 .OOl 14.2 .0007 19.8 .OOOl 17.8 .0002 15.6 .0004

.42 .5218 9.27 .0045

14.4 .0006 1.47 .2347

.Ol .9176 5.04 .0316 1.67 .2059 1.10 .3021 9.19 .0047 5.12 .0304

Equinus group n = 18 (mean age, 7.6yrsl; pes planovalgus group n = 17 (mean age, 9.8yrs). * Normalized refers to dividing value by leg lengths.


GAIT ASSESSMENT OF FIXED ANKLE-FOOT ORTHOSES. Abel 129

Temporal Factors

1 RBarefoot DAFO L

% Stance Time

Single Double

Joint Excursions

Fig 2. Histogram showing effects of AFOs on kinematic gait 3 30 parameters for the two indication groups (*significant differ-

g

ences, p < ,051. (A) Equinus ; 10

group. (B) Pes planovalgus -10 group. IC, initial contact; Max KF, maximum knee flexion during .30 the gait cycle; Max KE, maximum knee extension; Max ADF,

Pelvis Hip Knee

Joint Positions

I

Knee Max Max Ankle Max Max IC KF KE IC ADF APF

maximum ankle dorsiflexion; Max APF, maximum ankle plantarflexion.

A

I lBarefoot OAF0

Single Double Total

Eli Pelvis Hip Knee Ankle

J

Knee Max Max Ankle Max Max IC KF KE IC ADF APF

B

foot-flat strike in association with increased knee flexion at equinus group. The ground reaction force remains behind the contact (figs 3B, 4B). These patients, however, have more quadriceps and gastrocnemius weakness2’ so that they do not have

knee during stance phase, producing an external flexion moment and perpetuating a crouched position. Stress concentration at

the early heel rise and power generation characteristic of the the midfoot results in subtahu eversion and forefoot abducton

Table 3: Kinetic Analysis

Moments (Nm/kg) and Powers hvatts/kg) (Mean + SD)

Ankle plantar flexor moment, 0% to 30% cycle Ankle plantar flexor moment, 30% to 80% Ankle generation, 0% to 30% power Ankle generation, 30% to 80% power Knee extensor 0% to 30% moment, Knee extensor moment, 30% to 80%

Bold pairs are significant at p < .05.

Equinus Pee Planovalgus (n = 13) In = 7) Normal

Barefoot P!d

AFO Barefoot AFO (n = 18)

.89 + .36 .89 k .33 .70 2 .54 .79 2 22 .67

.94 t .42 1.10 ? 91 .80 + .26 .91 + .40 1.10

.68 i- .60 .38 + .43 .69 2 1.1 .I7 c .I5 0 1.38 i .35 .89 -+ .35 1.11 t .72 1.17 i 1.2 2.29

.50 ? .45 .44 t .33 .53 t .27 .52 + .28 .23

.41 + .42 .36 k .22 .57 F .28 .60 t .I9 .I5



A Hip Knee Ankle

*O I/ *O 77 3 60

ic 40

Sagittal Motion (degrees)

20

01 -20 ' I I 0 100 0 100

2 I 1

1 ,,*w. .“1.

k

', 0 I \

-1

:;

Sagittal Moments W-N9

Total Joint Power (watt/kg)

1 0

0 ,*... : I’ . .

-1 w

* ., **

-2 0 100 0 100

% Gait Cycle

Hip Knee B Ankle 20 , 1

Sagittal Motion (degrees)

Sagittal Moments (Nmk$

Fig 3. Ensemble average curves of the sagittal plane kinematic and kinetic date, and the total joint powers at the hip, knee, and ankle in barefoot (----I and with AFOs (- - -1 by indication group as compared with normal barefoot pediatric gait f-1. (A) Equinus group. (B) Pes planovalgus group.

Total Joint Power WatWd

0 100

% Gait Cycle

(planovalgus). In effect, tibia1 advancement in these patients is ankle moments and power generation are lower in this group accomplished not only with ankle dorsiflexion, but with motion than in the equinus group. that occurs within the subtalar and midtarsal joints.‘* The sagit- If passive ankle dorsiflexion with the knee in extension is tal plane ankle moment arm in terminal stance is shortened near neutral (O”), then AFOs are commonly prescribed to control because of the external rotation of the foot with respect to the some of the aforementioned abnormalities. In the equinus pa- line of forward progression, and roll-off occurs across the in- tients, the AFO effectively reduced the early stance ankle power step rather than the metatarsal heads. Consequently, late stance burst, and in both indication groups, it increased late stance


GAIT ASSESSMENT OF FIXED ANKLE-FOOT ORTMOSES, Abel 131

B

Fig 4. Stick figure recreations of lower extremity sagittal plane joint positions in barefoot and AFO gait, with the magnitude and direction of the vertical ground reaction force included, at five positions across the gait cycle for an individual patient in each of the two indication groups. (A) Patient with ankle equinus. (B) Patient with pes planovalgus. FS, foot strike; OFO, opposite foot off; MS, midstance; OFS, opposite foot strike; FO, foot off.

Barefoot

FS OF0 MS OFS FO

FS OF0 MS OFS FO

ankle moment. These changes reflect a more normal center of pressure transition from the midfoot to the forefoot. However, the AFO produced a reduction of the late stance power generation in the equinus group by restricting ankle motion (angular velocity) (table 3). In contrast, the pes planovalgus patients, with weakness and midfoot laxity, demonstrated reduced late stance power generation when barefoot. The AFOs appeared to increase the mean late stance moment by partially correcting the planovalgus and excessive dorsiflexion, but because of the small numbers with adequate force plate data the changes did not reach statistical significance. No appreciable change in ankle power generation was noted in the pes planovalgus group.

The AFO also facilitates foot clearance through elimination of the swing phase plantarflexion in the equinus group so that the ankle is more dorsiflexed at initial contact (Figs 3A, 4A).18 In addition, the AFO is presumed to increase the weight-bearing area in stance by eliminating equinus and planovalgus. This

assumption could be more thoroughly assessed utilizing a dynamic plantar pressure distribution system.

The current findings are supported by a previous investigation from our laboratory in which we compared two kinds of braces, supramalleolar orthoses (SMOs) and fixed AFOs, to shoes alone.” No barefoot condition was evaluated. The AFOs showed similar benefits as in the present study, but the magnitude of the changes was less, suggesting that some positive effects may be provided by shoes alone.

Joint positions in diplegic gait are not dictated by mechanical factors alone, yet it has been suggested that proper positioning of the foot and ankle with AFOs will produce beneficial effects in the more proximal joints. 23 This study clearly shows that the deleterious effects of spastic muscles on the proximal joints are not eliminated by the use of the AFOs. The increased early stance and midstance knee flexion were not reduced significantly. This abnormal stance phase knee flexion can be caused



by spasticity or tightness of muscles that cross the knee joint, including hamstrings and/or the gastrocnemius, or can be compensation for hip flexion deformitiesz4 In diplegia, the passive examination of joint motion does not always correlate with the positions seen dynamically in gait. If the dynamic activity of the hamstrings or gastrocnemius muscles is severe enough, use of the AFO may increase stance phase hip or knee flexion because the brace eliminates compensation that may occur at the ankle.

Although gait analysis provides objective quantification of gait parameters, some potential sources of error using 3-D sys- tems need to be examined. First, the foot is assumed to be a rigid segment and all motion occurring within the foot is recorded as ankle motion. For example, a person with pes planovalgus typically has some restriction of ankle dorsiflexion due to gastrocsoleus contracture. However, the gastrocsoleus contracture is not obvious on the ankle sagittal plane kinematic graph because the excessive motion that may occur within the midfoot is recorded as dorsiflexion. Differentiation of forefoot motion relative to the hindfoot could be improved by increasing the number of markers about the foot, but this would also necessi- tate decreasing the data collection volume, which could compro- mise the ability to collect complete gait cycles.”

A second potential source of error is the change in marker placement across conditions. 26 Placing markers over an orthosis plus a shoe is problematic because the anatomic landmarks must be estimated, and the appliance increases the width of the ankle markers and the height of the metatarsal markers.

A third consideration is that the AFO, while it rigidly blocks plantarflexion, is designed to allow some bending in the direction of dorsiflexion. Brace bending is a function of the thickness and flexibility of the fabrication material, the weight of the wearer, and the amount of stress applied to the orthosis. Chil- dren with pes planovalgus have greater knee flexion in stance so they have greater bending stress at the ankle. This may partially explain the shift in the ankle curve towards dorsiflexion in that group.

Several study design limitations need to be addressed. Our standard lab protocol was to collect barefoot data first, followed by AFO data collection. If patient fatigue was an influencing factor, we might expect to find fewer differences between barefoot and AFO conditions than those reported. Conversely, it is conceivable that children became more comfortable (confident) during later trials, which would falsely enhance the differences between the conditions. However, children with diplegia typically utilize increases in cadence as the primary means of increasing walking speed, as opposed to normal subjects who use concomitant increases in stride length and cadence.’ Therefore, we believe that the elimination or reduction of the foot and ankle deformities by the AFOs did result in enhanced stability and improvements in gait function. Nonetheless, a randomly chosen sequence of conditions should be utilized in future analyses.

All subjects in this study had acclimated to the use of AFOs so that this conditioning effect may partially explain some of the reduced gait function in their absence. However, all subjects in this study usually spend a portion of their day walking without braces; therefore, the magnitude of the acclimating effect is difficult to ascertain. The widespread prescription of AFOs to children with diplegia creates this type of selection bias. A prospective trial including subjects who have never experienced bracing is clearly needed to sort out the true benefits of AFOs and reasons for brace rejection.

One final consideration is that this investigation was limited to the evaluation of a single orthosis; although positive effects were found, different designs may be preferable. For instance,


an articulated AFO with plantarflexion stop at 0” may allow better late stance power generation in the equinus patient. Alter- natively, a rear-entry, dorsiflexion control or floor reaction AFO may be better for the pes planovalgus patient in whom crouch is caused by quadriceps weakness.

CONCLUSION The current standard treatment of diplegic patients without

significant ankle contractures is to use AFOs to improve ankle and foot patterns during walking. This study illustrates the beneficial effects on gait function in two groups of diplegic patients, equinus and pes planovalgus, by use of AFO versus barefoot. The mechanical effects of the brace were found to be confined primarily to the ankle while proximal joint alignment was not changed significantly. Therefore, if significant dynamic or static deformity is present in proximal joints, these should be addressed to optimize outcome. The AFOs did improved gait function for the cohort studied, as reflected by an increase in single- limb stance, stride length, and velocity. Further investigation should focus on identifying other patient and orthotic factors that maximize functional outcomes.

1.

2.

3.

4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

References Molnar GE. Rehabilitation in cerebral palsy. West J Med 1991; 154569-72.

Bleck EE. Neurological and orthopaedic assessment. In: Bleck E, editor. Orthopaedic management in cerebral palsy. Oxford: Mac- Keith Press, 1987. p. 17-64. Rosen MG, Dickinson JC. The incidence of cerebral palsy. Am J Obstet Gynecol 1992; 167:417-23. Binder H, Eng GD. Rehabilitation management of children with spastic diplegic cerebral palsy. Arch Phys Med Rehabil 1989;70: 482-9. Perry J. Pathologic gait. Instr Course Lect 1989;39:325-31. Gage JR. Gait analysis in cerebral palsy. Oxford: Blackwell Scien- tific, MacKeith Press; 1991. Gage JR, DeLuca PA, Renshaw TS. Gait analysis: principles and applications. J Bone Joint Surg Am 1995;77A:l607-23. Abel MF, Damiano DL. Strategies for increasing walking speed in diplegic cerebral palsy. J Pediatr Orthop 1996; 16:753-8. Damiano DL, Abel MF. Relationship of gait analysis to gross motor function in cerebral palsy. Dev Med Child Neural 1996;38:389-96. Park TS, Owens JH. Surgical management of spastic diplegia in cerebral oalsv. New Engl J Med 1992:326:745-9. Condie DN,-MeadowsCB. Ankle-foot orthoses. In: Bowker P, Condie DN, Bader DL, Pratt DJ, editors. Biomechanical basis of orthotic management. Oxford: Butterworth-Heinemann; 1993. p. 99-123. Powell MM, Silva PD, Grindeland T. Effects of two types of ankle- foot orthoses on the gait of children with spastic diplegia [abstract]. Dev Med Child Neurol 1989; 31 Suppl 59:8-9. Thomas SS, Mazur JM, Wright N. Quantitative assessment of ankle foot orthoses for children with cerebral palsy. Dev Med Child Neu- rol 1992;34:547-55. Butler PB, Nene AV. The biomechanics of fixed ankle foot orthosis and their potential in the management of cerebral palsied children. Physiotherapy 1991;77:81-8. Middleton EA, Hurley GRB, McIlwain JS. The role of rigid and hinged polypropylene ankle-foot orthoses in the management of cerebral palsy: a case study. Prosthet Orthot Int 1988; 12:129-35. Radta S, Skinner SR, Dixon D. A comparison of gait with solid and dynamic ankle-foot orthoses in children with spastic cerebral palsy [abstract]. Devel Med Child Neurol 1994;36 Suppl 70:26. Ramakrishnan HK, Kadaba MP. On the estimation of joint kinematics during gait. J Biomech 1991;24:969-77. Perry J. Gait analysis: normal and pathological function. Thorofare (NJ): Slack Inc; 1992. Sutherland DH. Gait analysis in neuromuscular diseases. Instr Course Lect 1989; 39:333-41. Inman VT, Ralston HJ, Todd F. Human walking. Baltimore (MD): Williams & Wilkins: 1981.

GAIT ASSESSMENT OF FIXED ANKLE-FOOT ORTHOSES. Abel 133

21. Damiano DL, Kelly LE, Vaughan CL. Effects of quadriceps femoris 25. Kidder SM, Abuzzahab FS; Harris GF. A system for the analysis muscle strengthening on crouch gait in children with spastic diple- of foot and ankle kinematics during gait. IEEE Trans Rehabil Eng gia. Phys Ther 1995;75:658-71. 1996;4:1-7.

22. Carlson WE, Damiano DL, Abel MF, Vaughan CL. Biomechanics of orthotic management of gait in spastic diplegia. Am J Phys Med Rehabil 1997;76:219-25.

26. Kadaba MP, Ramakrishnan ME, Wooten JG, Cochran GVB. Re- peatability of kinematic, kinetic and electromyographic data in normal adult gait. J Orthop Res 1989;7:849-60.

23. Cusick BD. Managing foot deformity in children with neuromotor disorders. Phys Ther 1988; 68: 1903-12.

24. Delp SL, Arnold AS, Speers RA, Moore CA. Hamstrings and psoas length during normal and crouch gait: implications for muscle ten- don surgery. .I Orthop Res 1996; 14:144-51.

Suppliers a. Motion Analysis Corp., Santa Rosa, CA. b. Oxford Metrics Limited, Oxford, England. c. Kistler Inc., Winterthur, Switzerland. d. SPSS Inc., Chicago, IL.


Gait Assessment of Fixed Ankle-Foot Orthoses in Children ... · Gait Assessment of Fixed Ankle-Foot...

Documents

Transcript of Gait Assessment of Fixed Ankle-Foot Orthoses in Children ... · Gait Assessment of Fixed Ankle-Foot...