knee in healthy subjects.8 Their investigation showed that setting an AFO in 5 degrees of dorsiflexion prolonged the flexion moment at the knee during early ...
Gait Comparison of Subjects with Hemiplegia Walking Unbraced, with Ankle-Foot Orthosis, and with Air-Stirrup® Brace RAY G. BURDETT, DIANE BORELLO-FRANCE, CATHLEEN BLATCHLY, and CYNTHIA POTTER The effects of the Air-Stirrup® (AS) standard ankle brace on the gait of 19 subjects with hemiplegia resulting from a cerebrovascular accident who exhibited excessive subtalar joint motion were studied. Videotaped trials and footprint analyses were used to measure subjects' hip, knee, and ankle sagittal plane angles; inversion and eversion of the calcaneus; and time-distance gait characteristics. A one-way analysis of variance for repeated measures was used to compare the gait of 19 subjects with the AS brace and unbraced and 11 subjects with the AS brace, unbraced, and with an ankle-foot orthosis. The AS brace was associated with more calcaneal stability during standing than the unbraced condition. The ankle-foot orthosis was associated with less plantar flexion at foot-strike than either the AS brace or unbraced condition. Both the AS brace and the ankle-foot orthosis were associated with less mid-swing plantar flexion and increased step length on the paretic side compared with no brace. These results support the effectiveness of the AS brace in controlling inversion and eversion instability in patients with hemiplegia. Key Words: Hemiplegia, orthoses and supports; Kinesiology/biomechanics, lower extremity; Lower extremity, ankle and foot; Orthotics/splints/casts, lower extremity.
Many patients with hemiplegia regain lower limb function; therefore, retraining patients with hemiplegia to walk is a major goal of the rehabilitation program. Separate studies have indicated that 60% to 75% of patients with hemiplegia can walk unaided at discharge from the hospital.1-3 Not all patients with hemiplegia, however, can walk unaided, and many patients show gait deviations resulting from abnormal motion at the hip, knee, ankle, and subtalar joints.4,5 To lessen gait deviation and subsequently improve the patient's ability to walk, the use of orthoses and walking aids has been suggested.4,6 Metal ankle-foot orthoses with double uprights and ankle joints that control dorsiflexion and plantar flexion have been recommended to minimize gait deviations caused by muscle tone changes or weakness around the knee and ankle after a cerebrovascular accident (CVA).7,8 Lehmann et al studied the R. Burdett, PhD, is Assistant Professor, Program in Physical Therapy, University of Pittsburgh, 4200 Fifth Ave, Pittsburgh, PA 15261 (USA). Address correspondence to Dr. Burdett. D. Borello-France, MS, is Assistant Director, Department of Physical Therapy, Harmarville Rehabilitation Center, Inc, PO Box 11460, Guys Run Rd, Pittsburgh, PA 15238-0460. C. Blatchly, BS, is Staff Physical Therapist, Comprehensive Rehabilitation Hospital, Department of Physical Therapy, Medical College of Ohio at Toledo, CF 10008, Toledo, OH 43699. She was Staff Physical Therapist, Harmarville Rehabilitation Center, Inc, when this study was completed. C. Potter, MS, is in private practice in Pittsburgh, PA. She was Research Assistant, Program in Physical Therapy, University of Pittsburgh, when this study was completed. This article was submitted February 4, 1987; was with the authors for revision 35 weeks; and was accepted February 2, 1988. Potential Conflict of Interest: 1.
Volume 68 / Number 8, August 1988
effects of altering the dorsiflexion-plantar flexion angle of conventional AFOs on flexion and extension moments at the knee in healthy subjects.8 Their investigation showed that setting an AFO in 5 degrees of dorsiflexion prolonged the flexion moment at the knee during early stance. When the AFO was set in 5 degrees of plantar flexion, the knee-flexion moment decreased in early stance, and the knee-extension moment increased at mid-stance. Plastic AFOs have also been recommended for patients with hemiplegia who have knee and ankle instability.9-11 Lehmann et al compared healthy subjects and subjects with hemiplegia while ambulating with various plastic AFOs.12 All of the plastic AFOs they tested limited plantar flexion at heelstrike through foot-flat, resulting in potential knee instability in both healthy subjects and subjects with hemiplegia. Subjects with hemiplegia had less dorsiflexion during push-off, a longer mid-stance phase, and a shorter push-off phase than healthy subjects. All but one of the subjects with hemiplegia using the plastic AFOs had less than 1 degree of dorsiflexion during the swing phase of gait, which was comparable to the kinematic characteristics of healthy elderly men. The plantar flexion of subjects with hemiplegia was not excessive, and the plastic AFOs were strong enough to prevent toe drag. Other studies have also examined the effect of AFOs on the gait of healthy and hemiplegic subjects. Opara et al found that a double-upright metal AFO reduced stride and step length and caused widening of step width and toe-out angles in healthy subjects.13 Simkin et al suggested that abnormal muscle activity and orthosis weight may influence gait.14 1197
Magora et al reported that rigid AFOs increased electromyographic activity in the unbraced limb that could result in limb discomfort and fatigue.15 Ofir and Sell, in a 10-year retrospective study of 843 patients with hemiplegia, found that improvement in functional ambulation was not related to the type of orthosis.16 The Air-Stirrup® (AS) ankle brace* has been shown to effectively allow adequate ankle movement in dorsiflexion and plantarflexionwhile providing mediolateral stability after inversion sprains.17,18 Its use has been suggested for patients with hemiplegia to control excessive foot inversion while allowing dorsiflexion and plantar flexion.19 The purpose of this study was to examine the effects of the AS brace on gait characteristics of patients with hemiplegia and to compare its effectiveness with plastic and metal AFOs. Our expectation was that the AS brace would be as effective as AFOs in limiting inversion and eversion, but that there would be no effect on dorsiflexion or plantar flexion.
TABLE 1 Subject Characteristics Sex Group
M All subjects Subjects with AFOb Subjects without AFO a b
Age (yr)
n F
Time Since CVAa (d)
s
19 10 9 61.9 11 5 6 64.6 8 5 3 58.2
10.7 8.1 12.9
s 114.5 133.5 88.4
108.5 127.4 76.6
CVA = cerebrovascular accident. AFO = ankle-foot orthosis.
METHOD Subjects Nineteen adults with a diagnosis of hemiplegia caused by a CVA participated as subjects after signing an informed consent form approved by the Harmarville Rehabilitation Center Institutional Review Board for biomedical research studies. Criteria for inclusion in the study required that subjects could ambulate unassisted or with a conventional or quad cane. Fifteen of the 19 subjects exhibited increased plantar flexion or plantar flexion and inversion tone, 3 subjects had normal plantar flexion and inversion tone, and 1 subject had decreased plantarflexionand inversion tone. Two subjects could dorsiflex actively through a full range of motion, and 10 subjects could initiate partial dorsiflexion. Tone and active movement at the ankle were tested in sitting and supine positions for all subjects. Tone was determined by the resistance felt by the physical therapist (C.B. or D.B.) to passive movements of the ankle. Active movement was evaluated by asking the subjects to dorsiflex their ankle. A physical therapist (C.B. or D.B.) assisted subjects with balance during the testing procedure and used her hands to stabilize the subject's pelvis during weight shifting to the stance leg if necessary. Eleven subjects wore either a metal or a plastic AFO before this study began, and 8 subjects did not wear an AFO. Characteristics for all subjects are shown in Table 1. The metal AFOs used in this study had double-action ankle joints adjusted so that the plantar-flexion stop prevented plantar flexion beyond neutral (90°). The adjustment to the dorsiflexion stop varied from subject to subject but generally allowed 5 to 10 degrees of movement in the direction of dorsiflexion. Plastic AFOs were fabricated from a rigid, hightemperature plastic material to ensure full control of ankle dorsiflexion and plantarflexion.The angle of the plastic AFOs varied; however, they were set either at neutral (90°) or at 5 degrees of ankle dorsiflexion. The AS brace consists of premolded side supports lined with inflatable air cells that adjust to conform to the individual (Fig. 1). A strap connecting the bases of the two side supports passes under the sole of the foot when the AS brace is worn. The side supports are positioned over the malleoli to limit inversion and eversion while permitting dorsiflexion and plantar flexion. The AS
Fig. 1. Air-Stirrup® standard ankle brace with premolded side supports.
brace fits comfortably into the shoe and is fastened by two cloth adhesive straps. Procedure We videotaped each subject from the side and rear while the subject walked 1) without a brace and 2) with the AS. The 11 subjects who wore an AFO were also videotaped walking with the AFO. The same pair of shoes was worn by a subject for each bracing condition. If a metal AFO was permanently attached to the subject's shoe, shoes with the same heel height were worn for each bracing condition. During each trial, the subject walked along a 3- × 15-ft† strip of brown paper taped to thefloor.On successive trials, a Panasonic Model WV32308AF color video camera‡ was positioned either to the side of the subject to obtain a sagittal plane view or to the rear of the subject so that calcaneal inversion and eversion could be observed in the frontal plane. The video camera was equipped with a digital stopwatch that superimposed time to the nearest one-hundredth of a second onto the video image. The order of bracing conditions (no brace, AS brace, and AFO) and the order of videotaping from the side or rear were randomly chosen for each subject. Subjects wore shorts that exposed their knees and ankle joints but covered their hip joints. Small squares of white adhesive tape with a 1.9-cm black dot were positioned on † 1ft= 0.3048 m.
* Aircast Inc, PO Box T, Summit, NJ 07901.
1198
‡ Panasonic Industrial Co, One Panasonic Way, Secaucus, NJ 07094.
PHYSICAL THERAPY
RESEARCH
Fig. 2. Placement of adhesive tape used to measure sagittal angles of subjects' paretic leg.
subjects' paretic side. For sagittal plane reference, tape was placed on the subjects' shorts over the greater trochanter, on the skin over the lateral midline of the knee joint space and lateral malleolus, and along the lateral border of the shoe at the base of the heel and head of the fifth metatarsal (Fig. 2). For frontal plane reference, tape was placed on the skin on the midline of the back of the leg just below the knee joint, on the midline of the Achilles tendon, and on the superior and inferior midline of the heel (Fig. 3). The tape allowed subsequent measurements ofjoint angles from the videotapes. For AFOs that covered the lateral malleolus or Achilles tendon, tape was applied to the AFO over these landmarks. The AS brace was applied uninflated and was then inflated by a physical therapist (C.B. or D.B.) who subjectively determined when the AS brace fit snugly. We affixed moleskin pads to the entire bottom of the sole of each subject's shoes. White latex paint diluted with an equal amount of water was brushed on the moleskin to produce footprints on the brown paper during walking trials. Footprints were used to measure step width, step length, stride length, and toe-out angle. Stride length and stride time measured from the videotape were used to calculate walking velocity. Data Analysis Four footprints consisting of two successive right and left footprints from the middle portion of each walking trial were analyzed (Fig. 4). These footprints were used to obtain two measurements of stride length (left heel to left heel and right heel to right heel), two measurements of step width (base of support between first and second footprints and between second and third footprints), one measurement of right step length (first left heel to subsequent right heel), one measurement of left step length (first right heel to subsequent left Volume 68 / Number 8, August 1988
Fig. 3. Placement of adhesive tape used to measure calcaneal inversion or eversion angle of subjects' paretic leg.
heel), and two measurements of left and right toe-out angle. The two measurements of stride length, step width, and toeout angle were averaged for the statistical analysis. To measure gait characteristics, the midline of each footprint was marked by bisecting the footprint with a straight line through the center of the heel and the most anterior part of the toes (Fig. 4). The left and right lines of progression were estimated by drawing lines between the middle of two successive right heels and two successive left heels. Toe-out angle was defined as the angle between the line of progression and the midline of the same foot. Right step length was determined by drawing a line (W) from the base of the right heel perpendicular to the left line of progression and then measuring the distance between this line and the base of the left heel. Right step width was defined as the length of line W. Left step length and step width were similarly determined. For analysis of joint angles, a rear view and a side view stride typical of a subject's gait were chosen from the videotapes for each bracing condition. Because between-stride variability existed within gait, the average of several strides would more accurately have characterized the gait. The limited endurance of the subjects, however, did not allow collection of several trials for each bracing condition. 1199
intervals using the stop-action videotape recorder and the goniometer. The subtalar joint axis is oblique to the sagittal plane and the transverse plane so that motion at this joint is triplanar.20 One component of this motion is inversion or eversion of the calcaneus relative to the lower leg within the frontal plane. The angle of inversion or eversion is measured clinically as the angle between the calcaneus and the lower leg, using a goniometer with one arm aligned along the posterior midline of the leg and the other arm along the midline of the calcaneus.21 In this study, the inversion or eversion angle was measured by a similar method except that the measurements were made from the videotapes using the tape as reference points to find the midline. This method of measuring inversion or eversion has been used in previous studies.22,23 The measurements were used to obtain the angle of the calcaneus at foot-strike and heel-off and the maximum change in calcaneal angle after foot-strike, whether it was inversion or eversion. The calcaneal angle change indicated mediolateral stability of the subtalar joint. We used a one-way analysis of variance (ANOVA) for repeated measures to compare the gait characteristics of all 19 subjects with an AS brace and with no brace. Eleven of the subjects also walked with a metal or plastic AFO. A oneway ANOVA for repeated measures was used to compare gait characteristics of all three bracing conditions for 11 subjects. When the ANOVA showed a difference among the three conditions, the Newman-Keuls multiple comparison post hoc test was used to determine which conditions differed from each other. RESULTS
Fig. 4. Schematic view of measurements obtained with footprint analysis. (TOL = left toe-out angle, TOR = right toe-out angle, SLL = left step length, SLR = right step length, W = step width.)
The videotapes were analyzed using a Panasonic Model NV-8950 stop-action video recorder‡ and a Panasonic Model CT-1930V color video monitor.‡ One complete stride from the paretic-side foot-strike until that foot-strike occurred again was analyzed from the side view. Sagittal angles of the paretic side were measured on the video monitor with a standard small clinical goniometer (Fig. 2). Hip flexion and extension were measured relative to the trunk by aligning the goniometer arm along the midline of the trunk and thigh, using the greater trochanter as the axis of rotation. Knee flexion was measured by aligning the goniometer arms along the midlines of the thigh and lower leg using the tape on the lateral knee as the axis. Ankle plantar flexion and dorsiflexion were measured by aligning one goniometer arm parallel to the line between the tape on the knee and lateral malleolus and on the other arm parallel to the line between the tape on the heel and metatarsal on the side of the foot. Angles during pareticside foot-strike, mid-stance, heel-off, toe-off, and mid-swing were measured. The rear-view videotape was used to analyze one paretic side gait cycle from foot-strike to heel-off at 0.03-second 1200
Tables 2 through 7 show averages of the various gait characteristics measured during this study and the results of the one-way ANOVA and Neuman-Keuls multiple comparison post hoc test. Gait characteristics associated with speed, distance, and time are shown in Tables 2 and 3. Except for paretic-side step length, which increased with use of either brace, the AS brace and AFO did not affect gait speed, distance, or time. Sagittal plane joint angles at various points in the gait cycle were unchanged by use of the AS brace or AFO except at the ankle joint. Use of the AS brace compared with no brace resulted in less plantar flexion at toe-off (Tab. 4). Use of the AFO resulted in less plantar flexion at footstrike than either the AS brace or unbraced condition, and TABLE 2 Gait Characteristics for Subjects Wearing Air-Stirrup® Brace and Unbraced (N = 19) Condition Characteristic
Air-Stirrup®
Unbraced s
s Stride time (sec) Stride length (cm) Speed (cm/sec) Base of support (cm) Step length (cm) Nonparetic side Paretic side Toe-out angle (°) Nonparetic side Paretic side a
P
2.6 62.9 26.8 13.6
1.4 27.4 17.3 3.8
2.7 57.9 22.5 12.9
.9 19.0 10.8 4.3
.768 .126 .113 .207
29.3 33.6a
14.2 9.1
27.2 30.7a
9.9 9.4
.278 .006
7.2 8.3
5.6 7.0
7.5 10.7
6.5 8.0
.840 .320
Significantly different from other condition at p < .05.
PHYSICAL THERAPY
RESEARCH TABLE 3 Gait Characteristics for Subjects Wearing Air-Stirrup® Brace, Unbraced, and Wearing Ankle-Foot Orthosis (AFO) (n = 11) Condition AirStirrup®
Characteristic
Condition (°)
a
AFO
Unbraced s
s Stride time (sec) Stride length (cm) Speed (cm/sec) Base of support (cm) Step length (cm) Nonparetic side Paretic side Toe-out angle (°) Nonparetic side Paretic side
P
29.8 17.3 26.4 11.9 26.2 11.5 .465 33.3 11.2 29.8a 11.4 33.4 9.8 .049 5.5 8.8 6.7 13.3
6.6 9.4 9.0 15.6
6.0 .655 6.1 .089
Significantly different from other conditions at p < .05.
TABLE 4 Sagittal-Plane Joint Angles of Subjects Wearing Air-Stirrup® Brace and Unbraced (N = 19) Condition (°) Joint
Air Stirrup®
a b c
Joint
Air-Stirrup
Unbraced
s
s
Anklea Foot-strike -12.8 Mid-stance -4.5 Heel-off 6.3 0.7 Toe-off Mid-swing -2.0 Kneec Foot-strike 16.3 11.2 Mid-stance 18.8 Heel-off Toe-off 36.1 Mid-swing 34.3 Hipc 20.9 Foot-strike Mid-stance 7.8 0.5 Heel-off 8.5 Toe-off Mid-swing 15.3 a
Unbraced
s Anklea Foot-strike Mid-stance Heel-off Toe-off Mid-swing Kneec Foot-strike Mid-stance Heel-off Toe-off Mid-swing Hipc Foot-strike Mid-stance Heel-off Toe-off Mid-swing
®
s
.7 .642 .8 2.5 22.3 56.6 18.2 .150 8.2 21.9 6.9 .428 4.1 12.3 3.4 .568
2.8 1.6 2.7 65.3 35.0 56.8 21.5 9.1 19.1 12.9 4.4 12.0
8.0 11.7
TABLE 5 Sagittal Plane Joint Angles of Subjects Wearing Air-Stirrup® Brace, Unbraced, and Wearing Ankle-Foot Orthosis (AFO) (n = 11)
P s
b c
-9.5 -2.1 7.3 2.6b 0.8
6.6 7.1 8.4 7.8 6.9
-10.4 -3.1 6.8 -2.0 b -1.9
8.5 8.0 7.5 9.5 7.1
.745 .584 .784 .040 .068
15.3 10.2 17.8 39.0 36.2
7.7 10.1 9.8 14.8 16.5
15.4 11.9 20.8 41.1 37.9
9.6 8.4 9.4 10.9 13.3
.994 .440 .442 .755 .823
21.2 8.5 2.9 10.2 18.2
10.5 7.4 10.0 8.4 10.1
19.7 7.8 0.5 10.1 15.9
10.9 7.4 9.1 9.4 11.0
.587 .554 .224 .941 .371
Positive value = dorsiflexion. Significantly different from other condition at p < .05. Positive value = flexion.
5.5 -13.3 7.7 -1.9 9.5 6.6 -5.4 8.9 7.3 -5.6 b
AFO
P s
9.4 -4.6 b 10.5 -1.2 9.4 6.8 10.5 -1.3 6.0 -0.1
5.1 7.8 6.6 6.1 5.9
.019 .475 .971 .122 .036
5.9 9.8 9.1 10.2 13.9
14.2 15.7 18.4 37.0 31.8
11.3 12.2 7.9 12.0 16.3
17.0 16.9 19.3 38.8 34.7
10.2 9.7 8.6 9.8 10.7
.774 .162 .889 .506 .667
12.3 7.4 8.0 8.9 12.0
18.9 7.6 -0.8 8.4 12.5
13.1 8.6 7.9 9.5 11.5
20.6 8.7 1.0 9.1 14.3
11.4 8.6 9.0 9.1 9.6
.287 .738 .482 .838 .259
Positive value = dorsiflexion. Significantly different from other conditions at p < .05. Positive value = flexion.
TABLE 6 Calcaneal Anglea Relative to Paretic Leg of Subjects Wearing Air-Stirrup® Brace and Unbraced (N = 19) Condition (°) Event
Foot-strike Maximum angle change after foot-strike Heel-off a b
Air-Stirrup®
Unbraced
s
s
P
2.9b
5.1
5.8b
5.7
.034
6.3b -1.5
3.9 6.2
10.0b -1.1
7.8 6.3
.023 .670
Negative value = eversion. Significantly different from other condition at p < .05.
TABLE 7 Calcaneal Anglea Relative to Paretic Leg of Subjects Wearing Air-Stirrup® Brace, Unbraced, and Wearing Ankle-Foot Orthosis (AFO)(n = 11) Condition (°)
use of either brace resulted in less plantar flexion during midswing (Tab. 5). Tables 6 and 7 show the average calcaneal angle at footstrike and heel-off and the average maximum change in calcaneal angle after foot-strike. The AS brace was associated with less inversion at foot-strike and less angular change than the unbraced condition (Tab. 6). No significant differences in calcaneal angle existed among the three bracing conditions (Tab. 7). DISCUSSION Hemiplegia results in patterns of gait that differ from normal gait in various ways. Speed, time, and distance characteristics generally differ drastically between healthy and Volume 68 / Number 8, August 1988
Event
AirStirrup®
Unbraced
s 4.7 Foot-strike Maximum angle change after foot-strike 5.5 0.7 Heel-off a
AFO
P
s
s
4.7
6.9
6.7
3.9
2.7
.315
3.9 5.4
8.9 0.4
9.3 6.0
4.5 1.4
3.0 3.5
.092 .824
Negative value = eversion.
hemiplegic adults. Murray reported the following mean gait components for healthy men aged 20 to 65 years walking at a freely chosen speed: 1) speed of 151 cm/sec, 2) stride length of 156 cm, 3) stride time of 1.06 seconds, 4) base of support 1201
of 7.7 cm, and 5) toe-out angle of 8 degrees.24 Comparing these characteristics with data in our study, hemiplegic gait was characterized by a much slower speed (15% of normal speed), much shorter stride length (37% of normal length), much longer stride time (250% of normal time), moderately wider base of support, and slightly greater toe-out angle. These time, speed, and distance measurements generally were similar to those reported by Holden et al.25 Use of the AS brace or an AFO did not change speed, temporal, or distance components significantly, except that the paretic-side step length increased a small, but statistically significant, amount with both braces (Tabs. 2, 3). Use of the AS brace or AFO, therefore, should not be expected to improve speed, distance, or temporal gait components of patients with hemiplegia. Sagittal-plane joint motions are very different between normal and hemiplegic gait. Comparing our results with normal gait as measured by Murray et al,26 hemiplegic gait was characterized by the following differences: 1) decreased hip flexion at foot-strike, increased hip flexion at toe-off, and decreased hip flexion during mid-swing; 2) more knee flexion at foot-strike and less knee flexion at toe-off and mid-swing; and 3) more ankle plantar flexion at foot-strike and midswing and less ankle plantar flexion at toe-off. Simon et al described the effects of an AFO worn by children with spastic cerebral palsy.27 The AFO decreased knee hyperextension during standing for subjects with marked spasticity of the plantar flexor muscles. In our study, however, the hip and knee angles were not changed in any significant way by use of either orthosis. Because our subjects were adults who did not exhibit severe spasticity of the plantar flexor muscles when resistance to passive dorsiflexion was tested in the sitting and supine positions, this result was expected. Lehmann et al described the effects of various plastic AFOs on the ankle joint of subjects with hemiplegia.12 All of the AFOs limited plantar flexion during heel-strike and midswing. In our study, use of the AFOs was also associated with a statistically significant reduction in plantar flexion at footstrike and mid-swing. The AFOs helped to correct the weakness of the dorsiflexor or abnormally active plantar flexor muscles during foot-strike and mid-swing by providing mechanical resistance to plantar flexion. The AFO provided mechanical resistance either by metal uprights along the side of the ankle joint or by rigid plastic supports along the back of the lower leg and under the mid-foot and hind-foot. We expected this mechanical resistance to also control subtalar joint motion. Although use of AFOs resulted in a trend toward subtalar joint stability (Tab. 7), no statistically significant difference occurred in inversion at foot-strike or in the maximum angular change of the calcaneus after foot-strike. The AS brace is designed to provide mediolateral stability for the subtalar joint without resistance to plantar flexion and dorsiflexion by means of bilateral inflatable air cells within premolded plastic supports. In our study, the AS brace was associated with statistically significant decreases in 1) plantar flexion during swing phase and at toe-off, 2) inversion at footstrike, and 3) maximum calcaneal angular change after footstrike (Tabs. 4, 6). The decrease in plantar flexion with the AS brace during certain phases of gait could be attributed to a decrease in extension synergy that included both inversion and plantar-flexion components. By limiting the amount of foot inversion during certain phases of gait, the AS brace may have disrupted the extensor muscle synergy and prevented excessive plantar flexion. This explanation, however, remains 1202
speculative without EMG data. Another possible explanation is that the limitation of inversion by the AS brace allows the tibialis anterior muscle to function only at the ankle joint rather than at both the ankle and subtalar joints. Limitation of inversion, therefore, could result in increased dorsiflexion, especially if the dorsiflexor muscles are very weak or are resisted by abnormally active plantar flexor muscles. Both the AFO and the AS brace are designed to provide mediolateral stability for the subtalar joint. An AFO brace provides stability with either metal uprights along the sides of the ankle or with rigid plastic supports along the back of the lower leg and under the midfoot and hindfoot that resist mediolateral movement. The AS brace stabilizes by means of bilateral inflatable air cells in the premolded plastic supports. This study shows that the AS brace was as effective as an AFO in controlling mediolateral instability for the subjects with hemiplegia in this study. Most AFOs are custom-made, but the AS brace is prefabricated in one size to fit most adults. The AS brace, therefore, is useful as a temporary or long-term inexpensive substitute for an AFO for selected hemiplegic patients with inversion or eversion instabilities. An AFO, because of its greater rigidity, may be more effective than the AS for instability more severe than that exhibited in this study. Further studies of subjects with severe instabilities are needed to identify the limits of instability that the AS brace can control. CONCLUSIONS
This study examined the effects of the AS brace on gait characteristics of subjects with hemiplegia and compared the effectiveness of the AS brace with metal and plastic AFOs. Gait speed, time, and distance characteristics were not affected by either type of brace, except that paretic-side step length increased a small, but statistically significant, amount with use of either brace. Hip- and knee-joint angles were not affected by either brace. At the ankle joint, use of the AFO resulted in statistically significant decreases in plantar flexion at foot-strike and mid-swing. Use of the AS brace resulted in statistically significant decreases in plantar flexion at toe-off and mid-swing. Gait comparison of the 19 subjects with and without the AS brace showed significantly less inversion at foot-strike and less calcaneal motion during stance phase when the AS brace was used. No significant differences in subtalar joint motion existed among the bracing conditions for the 11 subjects who used an AFO. REFERENCES 1. Lehmann JF, DeLateur BJ, Fowler RS, et al: Stroke: Does rehabilitation affect outcome? Arch Phys Med Rehabil 56:375-382, 1975 2. Marquardsen J: Natural history of acute cerebrovascular disease: Retrospective study of 769 patients. Acta Neurol Scand 45(Suppl 38):56-59, 1969 3. Moskowitz E, Lightbody FE, Freitag S: Long-term follow-up of the poststroke patient. Arch Phys Med Rehabil 53:167-172, 1972 4. Montgomery J, Inaba M: Physical therapy techniques in stroke rehabilitation. Clin Orthop 63:60-66, 1969 5. Davis PM: Steps to Follow: A Guide to the Treatment of Adult Hemiplegia Based on the Concept of K. and B. Bobath. Berlin, Federal Republic of Germany, Springer-Verlag, 1985, pp 147-153 6. Perry J: Lower extremity bracing in hemiplegia. Clin Orthop 63:32-36, 1969 7. Lehmann JF, Warren CG, DeLateur BJ: A biomechanical evaluation of knee stability in below-knee braces. Arch Phys Med Rehabil 51:688-695, 1970 8. Lehmann JF, Ko MJ, DeLateur BJ: Knee moments: Origin in normal ambulation and their modification by double-stopped ankle-foot orthoses. Arch Phys Med Rehabil 63:345-351, 1982
PHYSICAL THERAPY
RESEARCH 9. Engen TJ: TIRR polypropylene orthoses. Orthotics and Prosthetics 26(4): 1-15, 1972 10. Sarno JE, Lehneis HR: Prescription considerations for plastic below-knee orthoses. Arch Phys Med Rehabil 52:503-510, 1971 11. Simons BC, Jebsen RH, Wildman LE: Plastic short-leg brace fabrication. Journal of Orthotics and Prosthetic Appliances 21:215-218, 1967 12. Lehmann JF, Esselman PC, Ko MJ, et al: Plastic ankle-foot orthoses: Evaluation of function. Arch Phys Med Rehabil 64:402-407, 1983 13. Opara CU, Levangie PK, Nelson DL: Effects of selected assistive devices on normal distance gait characteristics. Phys Ther 65:1188-1191, 1985 14. Simkin A, Robin GC, Magora A, et al: Investigation of gait: Relationship between muscle action and mechanical stresses in below-knee braces. Electromyogr Clin Neurophysiol 13:495-503, 1973 15. Magora A, Robin GC, Rozin R, et al: Investigation of gait: Effect of a belowknee brace on the contralateral unbraced leg. Electromyogr Clin Neurophysiol 13:355-361, 1973 16. Ofir R, Sell H: Orthoses and ambulation in hemiplegia: A ten-year retrospective study. Arch Phys Med Rehabil 61:216-220, 1980 17. Hamilton W: Sprained ankles in ballet dancers. Foot Ankle 3(2):99-102, 1982 18. Stover CN, York JM: Air-stirrup management of ankle injuries in athletes. Am J Sports Med 8:360-365, 1980
Volume 68 / Number 8, August 1988
19. Hayes K: Air-stirrup bracing for patients with hemiparesis. Clinical Management in Physical Therapy 3(4):50, 1983 20. Inman VT: The Joints of the Ankle. Baltimore, MD, Williams & Wilkins, 1976, pp 57-62 21. Norkin CC, White DJ: Measurement of Joint Motion: A Guide to Goniometry. Philadelphia, PA, F A Davis Co, 1985, p 98 22. Bates BT, Osternig LR, Mason BR, et al: Functional variability of the lower extremity during the support phase of running. Med Sci Sports 11:328331, 1979 23. Nigg BM, Eberle G, Frey D, et al: Gait analysis and sport-shoe construction. In Asmussen E, Jorgensen K (eds): Biomechanics Vl-A. Baltimore, MD, University Park Press, 1978, pp 303-309 24. Murray MP: Gait as a total pattern of movement. Am J Phys Med 46:290333, 1967 25. Holden MK, Gill KM, Magliozzi MR: Gait assessment for neurologically impaired patients: Standards for outcome assessment. Phys Ther 66:1530-1539, 1986 26. Murray MP, Drought AB, Kory RC: Walking patterns of normal men. J Bone Joint Surg [Am] 46:355-360, 1964 27. Simon SR, Deutsch SD, Nuzzo RM, et al: Genu recurvatum in spastic cerebral palsy. J Bone Joint Surg [Am] 60:882-894, 1978
1203
