Incidence of Gait Abnormalities After Traumatic Brain Injury

Article Outline

• Abstract
• Methods
  • Participants
  • Instrumentation
  • Procedures
  • Data Analysis
• Results
• Discussion
  • Study Limitations
• Conclusions
• References
• Copyright

Abstract 

Williams G, Morris ME, Schache A, McCrory PR. Incidence of gait abnormalities after traumatic brain injury.

Objective

To identify the most common gait abnormalities presenting after traumatic brain injury (TBI) and quantify their incidence rate.

Design

Case series.

Setting

Biomechanics laboratory.

Participants

A convenience sample of 41 people with TBI receiving therapy for gait abnormalities, and a sample of 25 healthy controls.

Intervention

Three-dimensional gait analysis.

Main Outcome Measures

Spatiotemporal, kinematic, and kinetic data at a self-selected walking speed.

Results

People with TBI walked with a significantly slower speed than matched healthy controls. There was a significant difference between groups for cadence, step length, stance time on the affected leg, double support phase, and width of base of support. The most frequently observed biomechanical abnormality was excessive knee flexion at initial foot contact. Other significant gait abnormalities were increased trunk anterior/posterior amplitude of movement, increased anterior pelvic tilt, increased peak pelvic obliquity, reduced peak knee flexion at toe-off, and increased lateral center of mass displacement. Ankle equinovarus at foot-contact occurred infrequently.

Conclusions

People with TBI were found to have multijoint gait abnormalities. Many of these abnormalities have not been previously reported in this population.

Key Words: Biomechanics, Brain injuries, Gait, Rehabilitation

List of Abbreviations: BOS, base of support, CI, confidence interval, COM, center of mass, HC, healthy control, HiMAT, high-level mobility assessment tool, 3DGA, three-dimensional gait analysis, TBI, traumatic brain injury

TRAUMATIC BRAIN INJURY is a leading cause of death and disability for adolescents and young adults.¹ The prevalence of TBI in the community is high because of the survival rate, and the demographic group at most risk is adolescents and young adults.¹, ² The sequelae of severe TBI include motor, cognitive, behavioral, and emotional dysfunction.¹ Independent gait is often a major goal of rehabilitation after moderate to severe TBI. Adverse effects associated with gait abnormalities include falls,³ reduced aerobic fitness,⁴ and limited community access.⁵ Because falls are a major cause of TBI,¹, ⁶ and people with TBI are at a heightened risk of reinjury, there is a pressing need to ensure optimal therapy outcomes.

In contrast with the vast literature on neuropsychologic impairments after TBI, surprisingly little is known about the effect on gait.⁷ The key biomechanical abnormalities of gait after TBI are yet to be determined. 3DGA is a method for accurately measuring joint movement and is the current criterion standard for evaluating gait disorders. Despite 3DGA being used in numerous gait studies in TBI, investigations have restricted data reporting to spatiotemporal variables⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹² or focused on a single aspect of gait such as the hip,¹³ knee,¹⁴ ankle,¹⁵ or COM.¹⁶, ¹⁷, ¹⁸ Only 1 study has used 3DGA to compute data for all lower-limb joints. McFadyen et al¹⁹ reported kinematic analyses for the hip, knee, and ankle in obstructed and unobstructed walking for 8 high-functioning adults with TBI. To be included in this study, subjects were required to be able to walk without gait aids at a speed greater than 1m/s and walk independently in the community. These inclusion criteria make the findings difficult to generalize to other TBI populations, such as those who are mobile but are not high-functioning and exhibit severe and complex gait disorders.

3DGA has had a positive impact on clinical evaluation and surgical intervention in cerebral palsy²⁰, ²¹ and stroke.²² In contrast, the use of 3DGA in the treatment of people with TBI has to date been limited. Detection of the gait variables most frequently affected by TBI may streamline assessment procedures, inform clinical decision-making, and direct intervention programs. The aim of this study was therefore to identify the type and incidence of gait abnormalities after TBI. In particular, we attempted to determine the gait abnormalities that are characteristic of TBI and report biomechanical data for the trunk, pelvis, and lower limbs.

Back to Article Outline

Methods 

This project was approved by Epworth Hospital's Human Research Ethics Committee (study no. 340006) and the University of Melbourne (Ethics ID: 060496.1).

Participants 

Subjects with TBI were recruited for this project from Epworth Hospital, Melbourne, Australia. Patients currently attending physiotherapy for gait limitations after TBI were asked to participate in this project. The inclusion criteria were patients who (1) had sustained a TBI and (2) were able to walk independently over a distance of 20m without the use of a gait aid. Exclusion criteria were (1) patients who were unwilling or unable to provide informed consent, (2) those presenting with concurrent central nervous system disorders, and (3) those presenting with severe cognitive or behavioral problems that prevented assessment. All subjects who were invited to participate consented to do so. Data were also collected from a sample of HC subjects matched for sex, age (±5y), height (±7cm), and weight (±15kg). Table 1 summarizes the demographic data for the TBI and HC samples.

Table 1. Participant Characteristics

NOTE. Values are means ± SDs unless otherwise indicated.

Instrumentation 

3DGA was performed at the Centre for Health, Exercise and Sports Medicine, in the School of Physiotherapy at The University of Melbourne. Kinematic data were acquired using a motion analysis system (Vicon 512^a) with 8 cameras sampling at a rate of 120Hz. Ground reaction force data were collected using 3 AMTI forceplates^b sampling at a rate of 1080Hz.

Procedures 

Twenty-five small (14mm diameter) passive reflective markers were mounted on the skin at specific locations on the pelvis and both lower limbs after a previously described protocol.²³ Three markers were also placed on the trunk overlying the spinous processes of T2 and T10 as well as the sternal notch. Subjects initially performed a standing calibration trial, with additional markers placed bilaterally on the medial femoral condyle, medial malleolus, and proximal calcaneum of both legs. These markers were used to define joint center locations and anatomical coordinate systems.²³ The hip joint center was found using the method of Harrington et al.²⁴

Participants performed walking trials over a 12-m walkway while data were collected at their self-selected walking speed. Spatiotemporal, kinematic, and kinetic data for 5 trials were collected for each lower limb to gain a representative sample of each participant's gait pattern. In order to control for the effect of speed on kinematic and kinetic data, the HCs walked at a speed comparable to the mean (±5%) TBI self-selected walking speed. HCs were given verbal feedback regarding the accuracy of the matched speed. Only trials within 5% of the mean TBI self-selected walking speed were included. Five trials with complete data were collected for each variable on each leg for every HC to generate the normative values. Fifteen of the control subjects also performed walking trials at their self-selected speed for comparison of the spatiotemporal gait variables.

Spatiotemporal, kinematic, and kinetic data were calculated for the TBI and HC groups at self-selected and matched gait speeds. Because of the large number of gait variables that could potentially be investigated, a prioritized list of 20 gait variables (table 2) was identified for this study. These variables were generated from a review of known gait abnormalities after TBI, the key gait variables for normal walking, and specific variables that have been identified as problematic in other neurologic populations. Table 2 summarizes the gait variables prioritized for assessment in this study and how each was measured.

Table 2. Gait Variables and Descriptors

Gait Variable	Units	Description	Reference no.
Spatiotemporal
Velocity	Meters/second		15, 17, 19, 25, 26
Cadence	Strides/minute		15
Step length	Meters		15, 17, 19, 25, 26
Stance duration	Seconds		26
Double support	Seconds		26
Base of support	Meters	Perpendicular distance between the distal calcaneal markers	25, 26
Kinematic
Trunk flexion	Degrees	Range between maximum and minimum values in the gait cycle	27, 28
Trunk lateral flexion	Degrees	Range between maximum and minimum values in the gait cycle	27, 28
Anterior pelvic tilt	Degrees	Average value in the gait cycle
Pelvic obliquity	Degrees	Peak value in swing phase	27, 28, 29, 30
Pelvic rotation	Degrees	Average value in the gait cycle	28, 29, 30
Hip extension	Degrees	Peak value at terminal stance	27, 28, 29, 30, 31
Hip adduction	Degrees	Range between initial contact and first peak/trough in loading response	29
Knee flexion initial contact	Degrees	Angle at initial contact	27, 28, 29, 32, 33
Knee flexion mid-stance	Degrees	Angle at mid-stance	27, 32, 33, 34
Knee flexion swing	Degrees	Angle at toe-off	28, 29, 30, 32, 33
Ankle flexion initial contact	Degrees	Angle at initial contact; positive value indicates plantarflexion	27, 28, 29, 30, 34
Foot equinovarus	Degrees	Peak value in stance phase
Lateral COM displacement	Millimeters	Range between maximum and minimum values in the gait cycle, based on the sacral marker	27, 28, 34
Kinetic
Push-off terminal stance	Watts/kg	Peak ankle joint plantarflexor power generation	31

A clinical measure of mobility was also recorded using the HiMAT.²⁵ The HiMAT was chosen because it is the best measure of mobility for people with TBI who are independent of gait aids.²⁶

Data Analysis 

Three-dimensional joint kinematic and kinetic calculations were performed using Bodybuilder software.^a All lower-limb joint kinematic and kinetic data were computed using a previously described approach.²³ The 3 markers mounted on the trunk were used to define a local coordinate system, and the angular orientation of the trunk was described with respect to the pelvis.

Summary statistics (mean, SD, range) were generated for all of the key gait variables. Comparisons to the HC sample were made using a t test. Values for the spatiotemporal variables were compared with data collected from a sample of 15 matched HCs at their self-selected gait speed. Values for the kinematic and kinetic variables were compared with data collected from a sample of 25 matched HCs at a gait speed matched to the mean walking speed for the TBI sample.

In addition to comparing group data, individual results for the TBI sample were compared with the 95% CIs calculated for each gait variable for the HC sample. All values within ±2 SD from the mean were categorized as normal. Those values greater than 2 SD from the mean were categorized as increased, and those values less than 2 SD from the mean were categorized as decreased. Sample size calculations, based on the gait data reported by Ochi et al,²⁷ indicated that 12 participants with TBI would be required at 80% power.

Back to Article Outline

Results 

Table 1 outlines the demographic data of the participants. The 41 participants with TBI were predominantly young (mean, 29.1±9.4y) and male (31 male, 10 female), consistent with the broader TBI population. All participants except 1 had sustained an extremely severe TBI, determined by the length of posttraumatic amnesia.²⁸ This sample varied considerably in the time to 3DGA testing postinjury. No significant difference was identified between the TBI and HC samples for age, height, or weight.

The TBI and HC summary statistics are outlined in Table 3, Table 4. Figure 1 summarizes the kinematic data in each of the 3 planes of movement. The TBI sample was significantly different from matched HCs for 12 of 20 gait variables. A further 3 variables were not significantly different, yet almost half of the TBI sample was classified as abnormal (table 5). All 6 spatiotemporal variables analyzed were significantly different from normal. The TBI sample adopted a slower walking speed because of a reduced cadence and a shorter step length. Although the TBI sample tended to have a greater stance phase, relatively few had an increase in double support time. The TBI sample had a significantly wider BOS. All people who had an abnormal BOS were categorized as increased; no subjects had a reduced BOS because of excessive hip adduction and scissoring.

Table 3. Spatiotemporal Variables at Self-Selected Walking Speeds

NOTE. Data for step length and stance duration are reported for the more affected leg.

Table 4. Kinematic and Kinetic Variables at Matched Walking Speeds

NOTE. Velocity for the HC group, 1.07±.06m/s, was matched to the TBI group, 1.07±.34m/s.

Abbreviation: deg, degrees.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Kinematic data for TBI and healthy controls. These graphs demonstrate joint movement during 1 complete gait cycle in each of the 3 planes. Each graph begins at initial foot strike. The vertical dashed line represents toe-off for the TBI cohort, and the adjacent vertical solid line represents toe-off for the HCs. The portion of the graph from initial foot strike to the toe-off line represents joint movement during stance phase, while swing phase movement is to the right of the toe-off line. The position of the joint or body segment in degrees is represented along the y-axis. The shaded portion of the graph represents the 95% CIs for movement in the HC population. The solid black line and the lightly dashed lines represent the mean and ±2 SD for the TBI population, respectively. Abbreviations: Abd, abduction; Add, adduction; Ant, anterior; DF, dorsiflexion; Flex, flexion; FS, foot strike; Ev, eversion; Extn, extension; Ext, external; Int, internal; Inv, inversion; PF, plantarflexion; Post, posterior; TO, toe-off; Var, varus; Val, valgus.

Table 5. Incidence of Gait Variable Abnormalities (for the More Affected Leg)

Gait Variable	Decreased (%)	Normal (%)	Increased (%)
Velocity (m/s)	25(61.0)	15(36.6)	1(2.4)
Cadence (strides/min)	18(43.9)	22(53.7)	1(2.4)
Step length (m)	7(17.1)	34(82.9)	0(0.0)
Stance duration (s)	1(2.4)	19(46.4)	21(51.2)
Double support (s)	0(0.0)	37(90.2)	4(9.8)
Base of support (m)	0(0.0)	27(65.8)	14(34.2)
Trunk flexion (deg)⁎	0(0)	16(57.1)	12(42.9)
Trunk lateral flexion (deg)⁎	4(14.3)	21(75.0)	3(10.7)
Anterior pelvic tilt (deg)	0(0)	27(65.9)	14(34.1)
Pelvic obliquity (deg)	1(2.4)	32(78.0)	8(19.5)
Pelvic rotation (deg)	11(26.8)	22(53.7)	8(19.5)
Hip extension (deg)	9(22.0)	24(58.5)	8(19.5)
Hip adduction (deg)	1(2.4)	40(97.6)	0(0)
Knee flexion initial contact (deg)	0(0)	16(39.0)	25(61.0)
Knee flexion mid stance (deg)	15(36.6)	21(51.2)	5(12.2)
Knee flexion swing (deg)	21(51.2)	18(43.9)	2(4.9)
Ankle flexion initial contact (deg)	3(7.3)	33(80.5)	5(12.2)
Foot equinovarus (deg)	4(9.8)	36(87.8)	1(2.4)
Lateral COM displacement (mm)	0(0)	23(56.1)	18(43.9)
Push-off terminal stance (W/kg)	19(46.3)	17(41.5)	5(12.2)

NOTE. Spatiotemporal data comparison made at self-selected walking speeds; kinematic and kinetic data comparison made at a matched walking speed.

Frequency of TBI classification beyond the healthy control 95% CI is presented in table 5. Data are presented for the more affected leg. The more affected side was determined by the number of times a given subject's gait variables fell outside the 95% CI for each leg. Classification of abnormality varied considerably depending on the particular variable. Figure 2 demonstrates the variability in pelvic movement in 2 participants with TBI. Some variables were found to be either normal or decreased (eg, velocity, cadence, knee flexion swing, push-off terminal stance), while other variables were either normal or increased (eg, stance duration, BOS, trunk flexion, anterior pelvic tilt, knee flexion at initial foot contact, lateral COM displacement). Finally, some variables were found to be skewed either way (maximal hip extension in stance, pelvic rotation).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Pelvic rotation abnormality. (A) A patient with the clinical presentation of ataxia demonstrating substantial variability in performance between trials, yet no asymmetry. (B) A patient with dense hemiparesis demonstrating significant pelvic asymmetry, but little variability, clinically referred to as left pelvic retraction. The left side is represented by solid lines, the right side by dotted lines, and the grey shadow represents the 95% CI for normal pelvic rotation during gait. Abbreviations: Ext, external; FS, foot strike; Int, internal; TO, toe-off.

In terms of the kinematic and kinetic variables, knee joint angle at initial contact was the most frequently abnormal variable. Most subjects with TBI had excessive knee flexion at initial contact. This crouch position at initial contact was not maintained during stance, because many subjects with TBI were found to hyperextend their knees. Excessive knee extension, along with reduced push-off, was also evident at terminal stance/preswing. Of the 21 subjects with a stiff legged gait pattern, only 12 (57%) had reduced push-off.

Back to Article Outline

Discussion 

This is the first systematic description of the type and incidence of abnormality involving the key biomechanical events in the gait cycle after TBI. Results yielded several surprising outcomes that have not been identified previously in TBI. The most common gait abnormalities identified in this study were related to trunk and pelvic movements, and excessive knee flexion at initial foot contact. Other gait abnormalities associated with TBI that have previously been cited in the literature, such as hip adduction (scissoring)²⁹ and ankle equinovarus (plantarflexion with inversion),¹⁵, ²⁹, ³⁰, ³¹ were found to occur infrequently. This is a surprising result because equinovarus and calf spasticity are common impairments in the TBI population²⁹, ³¹ and may reflect the spasticity management at the rehabilitation facility in the early phase before the patients had become ambulant.

The prevalence of gait abnormalities identified in this cohort may be a product of sampling bias. The inclusion criteria were designed simply to reflect the wide range of gait abnormalities in people receiving therapy for mobility limitations. The sample was quite heterogeneous in terms of clinical presentation, mobility limitations, length of time postinjury, and age. Several methods for classifying movement disorders and clinical syndromes in TBI have been proposed,²⁹, ³¹, ³² yet none have been empirically tested using motion analysis. To our knowledge, only 1 attempt has been made to classify gait patterns after TBI. Gradenigo⁸ classified the gait patterns of 14 people with TBI based on their biomechanical profile, but all 3 classification systems used were developed in stroke or cerebral palsy and were based on a hemiplegic presentation. Although clinical descriptors such as hemiparesis and ataxia are important and may improve homogeneity, the underlying biomechanical deficits of these clinical descriptors have not been established in TBI.

Pelvic and trunk abnormalities have received little attention and were more prevalent than previously described gait patterns.²⁹, ³¹ Figure 2A demonstrates substantial intertrial pelvic rotation variability in a patient with a clinical presentation of ataxia. Each trial fluctuated between excessive internal and external rotation, but no asymmetry, while figure 2B demonstrates significant pelvic rotation asymmetry in a patient with a clinical presentation of dense hemiparesis yet little intertrial variability. Clinical interventions for the patient with ataxia may target improved interlimb coordination for improved pelvic stability,³³ while the patient with hemiparesis may require greater pelvic and trunk strength to achieve symmetry.³⁴ Further investigation into the biomechanical abnormalities associated with common clinical syndromes, such as hemiparesis and ataxia, may allow for subclassification in TBI and greater homogeneity in gait research.

Excessive knee flexion at initial contact was not highly associated with the previously described flexed knee pattern.²⁹, ³¹ This pattern consists of an excessively flexed knee in stance phase, coupled with excessive hip flexion, reduced push-off, excessive contralateral knee flexion in swing, and reduced contralateral step length.²⁹, ³¹ Only 12.2% of participants had an excessively flexed knee at mid-stance, despite 61% commencing stance phase with excessive knee flexion at initial contact. Although 60% of participants with an excessively flexed knee also had an excessively flexed hip, only 20% had reduced push-off. A gait pattern implies that movements in the lower limb are related and reasonably consistent, and that information obtained about movement at 1 joint may lead to assumptions at adjacent joints with reasonable certainty. Careful consideration must be taken with assumptions made about gait abnormalities in TBI, because our data indicate that abnormal gait events occur independently of each other rather than in consistent patterns. Possibly greater consideration needs to be given to the proportion of participants who demonstrated knee hyperextension as stance phase progressed. All subjects had knee flexion (normal or excessive) at initial contact, but over one third then hyperextended their knee by mid-stance. Knee hyperextension in the early part of stance phase is associated with greater external peak knee extensor torques and greater risk of knee joint injury.¹⁴ Greater emphasis may need to be placed on screening people with TBI for this abnormality for the long-term protection of their knee joints.

Most people with TBI walked at a slower speed. Data presented in table 3 would indicate that reduced gait speed was primarily a result of shorter stride length because it demonstrates a higher proportional reduction from normal than cadence. McFadyen et al¹⁹ suggested people with TBI reduce stride length to compensate for reduced postural control and stability, yet our data indicate reduced gait speed may be primarily a result of reduced cadence (see table 5). Cadence may also be reduced to improve postural control and stability, but further examination of our data showed 84% of participants with a slow gait speed had reduced push-off or a stiff-legged pattern. The strength of push-off, the main driver of forward propulsion in gait,³⁵ was restricted in 64% of those who walked with a slow gait speed. A stiff-legged gait pattern, characterized by knee extension in swing phase, was present in 52% of slow walkers. A stiff-legged pattern, primarily caused by knee extensor spasticity,³⁶ slows limb advancement in the longest portion of the gait cycle. Therefore, reduced gait speed after TBI may be a result of the inability to walk at an age-appropriate speed, rather than a compensatory strategy to improve postural control and stability.

The purpose of this study was to benchmark the scope of gait abnormalities present in the TBI population because they had not been fully described. Further investigation is required to determine which of the many potential contributing factors—such as reduced postural control, balance, strength, range of motion, or spasticity and dystonia—are the primary impairment, and whether the biomechanical abnormalities identified are primary deficits or secondary compensatory strategies. We did not seek to minimize the role of these impairments on the results of the gait analysis but rather simply to describe the scope of the biomechanical abnormalities that may be present in the TBI population.

Study Limitations 

This study has several limitations. We recruited a convenience sample of 41 ambulant participants receiving therapy for gait limitations after TBI. We found little evidence to support the commonly described gait patterns that exist after TBI²⁹, ³¹ and little homogeneity. A much larger sample size may identify the existence and prevalence of clinical syndromes and gait patterns, yet this sample of 41 participants is the first comprehensive description of kinematic abnormalities and is considerably larger than most gait studies published in TBI.¹², ¹⁶, ¹⁷, ¹⁹, ³⁷ The gait variables chosen for this study were based on key gait events that are commonly affected in people with neurologic deficits. It is possible that biomechanical abnormalities, particularly kinetic variables, unique to TBI have yet to be considered and may have been overlooked. Although not conducted in this study, electromyography may provide greater insights to the onset and impact of muscle activity on gait variables, particularly in relation to strength deficits and spasticity.

Back to Article Outline

Conclusions 

People with TBI have severe and complex gait abnormalities that limit mobility. The type and incidence of kinematic gait abnormalities after TBI have not been previously reported, but are prevalent throughout the trunk, pelvis, and lower limb. In this heterogeneous cohort, gait abnormalities do not seem to fit into discrete patterns. Slow walking may be a result of inability to increase gait speed rather than impaired postural control.

Suppliers

Back to Article Outline

References 

1. Thurman D, Alverson C, Browne D, et al. Traumatic brain injury in the United States: a report to Congress. 1999;http://www.cdc.gov/NCIPC/tbi/tbi_congress/tbi_congress.htmAccessed December 16, 2008
   • View In Article
2. Hirtz D, Thurman DJ, Gwinn-Hardy K, Mohamed M, Chaudhuri AR, Zalutsky R. How common are the “common” neurologic disorders?. Neurology. 2007;68:326–337
   • View In Article
   • CrossRef
3. Medley A, Thompson M, French J. Predicting the probability of falls in community dwelling persons with brain injury: a pilot study. Brain Inj. 2006;20:1403–1408
   • View In Article
   • MEDLINE
   • CrossRef
4. Bateman A, Culpan FJ, Pickering AD, Powell JH, Scott OM, Greenwood RJ. The effect of aerobic training on rehabilitation outcomes after recent severe brain injury: a randomized controlled evaluation. Arch Phys Med Rehabil. 2001;82:174–182
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (84 KB)
   • CrossRef
5. Powell J, Heslin J, Greenwood R. Community based rehabilitation after severe traumatic brain injury: a randomised controlled trial. J Neurol Neurosurg Psychiatry. 2002;72:193–202
   • View In Article
   • MEDLINE
   • CrossRef
6. Cassidy JD, Carroll LJ, Peloso PM, et al. Incidence, risk factors and prevention of mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury. J Rehabil Med. 2004;43(Suppl):28–60
   • View In Article
7. Cantin J-F, McFadyen BJ, Doyon J, Swaine B, Dumas D, Vallee M. Can measures of cognitive function predict locomotor behaviour in complex environments following a traumatic brain injury?. Brain Inj. 2007;21:327–334
   • View In Article
   • MEDLINE
   • CrossRef
8. Gradenigo B. Gait patterns after traumatic brain injury. Eura Medicophys. 2002;38:215–218
   • View In Article
9. Horn TS, Yablon SA, Stokic DS. Effect of intrathecal baclofen bolus injection on temporospatial gait characteristics in patients with acquired brain injury. Arch Phys Med Rehabil. 2005;86:1127–1133
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (284 KB)
   • CrossRef
10. Molteni F. Botulinum toxin and rehabilitation programs in lower limb spasticity. Eur J Neurol. 1995;2:61–67
    • View In Article
11. Pohl M, Mehrholz J. Immediate effects of an individually designed functional ankle-foot orthosis on stance and gait in hemiparetic patients. Clin Rehabil. 2006;20:324–330
    • View In Article
    • MEDLINE
    • CrossRef
12. Vallee M, McFadyen BJ, Swaine B, Doyon J, Cantin J-F, Dumas D. Effects of environmental demands on locomotion after traumatic brain injury. Arch Phys Med Rehabil. 2006;87:806–813
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (400 KB)
    • CrossRef
13. Lee LW, Kerrigan DC, Della Croce U. Dynamic implications of hip flexion contractures. Am J Phys Med Rehabil. 1997;76:502–508
    • View In Article
    • MEDLINE
    • CrossRef
14. Kerrigan DC, Deming LC, Holden MK. Knee recurvatum in gait: a study of associated knee biomechanics. Arch Phys Med Rehabil. 1996;77:645–650
    • View In Article
    • Abstract
    • Full-Text PDF (718 KB)
    • CrossRef
15. Fock J, Galea MP, Stillman BC, Rawicki B, Clark M. Functional outcome following Botulinum toxin A injection to reduce spastic equinus in adults with traumatic brain injury. Brain Inj. 2004;18:57–63
    • View In Article
    • MEDLINE
    • CrossRef
16. Basford JR, Chou L-S, Kaufman KR, et al. An assessment of gait and balance deficits after traumatic brain injury. Arch Phys Med Rehabil. 2003;84:343–349
    • View In Article
    • Abstract
    • Full-Text PDF (132 KB)
    • CrossRef
17. Chou L-S, Kaufman KR, Walker-Rabatin AE, Brey RH, Basford JR. Dynamic instability during obstacle crossing following traumatic brain injury. Gait Posture. 2004;20:245–254
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (218 KB)
    • CrossRef
18. Kaufman KR, Brey RH, Chou L-S, Rabatin A, Brown AW, Basford JR. Comparison of subjective and objective measurements of balance disorders following traumatic brain injury. Med Eng Phys. 2006;28:234–239
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (176 KB)
    • CrossRef
19. McFadyen BJ, Swaine B, Dumas D, Durand A. Residual effects of a traumatic brain injury on locomotor capacity—a first study of spatiotemporal patterns during unobstructed and obstructed walking. J Head Trauma Rehabil. 2003;18:512–525
    • View In Article
    • MEDLINE
    • CrossRef
20. DeLuca PA, Davis RB, Ounpuu S, Rose S, Sirkin R. Alterations in surgical decision making in patients with cerebral palsy based on three-dimensional gait analysis. J Pediatr Orthop. 1997;17:608–614
    • View In Article
    • MEDLINE
    • CrossRef
21. Kay RM, Dennis S, Rethlefsen S, Reynolds RA, Skaggs DL, Tolo VT. The effect of preoperative gait analysis on orthopaedic decision making. Clin Orthop Relat Res. 2000;217–222
    • View In Article
22. Fuller DA, Keenan MA, Esquenazi A, Whyte J, Mayer NH, Fidler-Sheppard R. The impact of instrumented gait analysis on surgical planning: treatment of spastic equinovarus deformity of the foot and ankle. Foot Ankle Int. 2002;23:738–743
    • View In Article
    • MEDLINE
23. Schache AG, Baker R. On the expression of joint moments during gait. Gait Posture. 2007;25:440–452
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1142 KB)
    • CrossRef
24. Harrington ME, Zavatsky AB, Lawson SE, Yuan Z, Theologis TN. Prediction of the hip joint centre in adults, children, and patients with cerebral palsy based on magnetic resonance imaging. J Biomech. 2007;40:595–602
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (606 KB)
    • CrossRef
25. Williams GP, Robertson V, Greenwood KM, Goldie PA, Morris ME. The high-level mobility assessment tool (HiMAT) for traumatic brain injury: part 2: content validity and discriminability. Brain Inj. 2005;19:833–843
    • View In Article
    • MEDLINE
    • CrossRef
26. Williams G, Robertson V, Greenwood K, Goldie P, Morris ME. The concurrent validity and responsiveness of the high-level mobility assessment tool for measuring the mobility limitations of people with traumatic brain injury. Arch Phys Med Rehabil. 2006;87:437–442
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (142 KB)
    • CrossRef
27. Ochi F, Esquenazi A, Hirai B, Talaty M. Temporal-spatial feature of gait after traumatic brain injury. J Head Trauma Rehabil. 1999;14:105–115
    • View In Article
    • MEDLINE
    • CrossRef
28. Shores EA, Marosszeky JE, Sandanam J, Batchelor J. Preliminary validation of a clinical scale for measuring the duration of post-traumatic amnesia. Med J Aust. 1986;144:569–572
    • View In Article
    • MEDLINE
29. Esquenazi A. Evaluation and management of spastic gait in patients with traumatic brain injury. J Head Trauma Rehabil. 2004;19:109–118
    • View In Article
    • MEDLINE
    • CrossRef
30. Chua KS, Kong KH. Clinical and functional outcome after alcohol neurolysis of the tibial nerve for ankle-foot spasticity. Brain Inj. 2001;15:733–739
    • View In Article
    • MEDLINE
    • CrossRef
31. Perry J. The use of gait analysis for surgical recommendations in traumatic brain injury. J Head Trauma Rehabil. 1999;14:116–135
    • View In Article
    • MEDLINE
    • CrossRef
32. Griffith ER, Mayer NH. Hypertonicity and movement disorders. In:  Rosenthal M,  Griffith ER,  Bond MR,  Miller JD editor. Rehabilitation of the adult and child with traumatic brain injury. 2nd ed.. Philadelphia: F. A. Davis; 1990;p. 127–147
    • View In Article
33. Bracewell RM, Balasubramaniam R, Wing AM. Interlimb coordination deficits during cyclic movements in cerebellar hemiataxia. Neurology. 2005;64:751–752
    • View In Article
    • CrossRef
34. De Bujanda E, Nadeau S, Bourbonnais D, Dickstein R. Associations between lower limb impairments, locomotor capacities and kinematic variables in the frontal plane during walking in adults with chronic stroke. J Rehabil Med. 2003;35:259–264
    • View In Article
    • MEDLINE
    • CrossRef
35. Liu MQ, Anderson FC, Pandy MG, Delp SL. Muscles that support the body also modulate forward progression during walking. J Biomech. 2006;39:2623–2630
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (324 KB)
    • CrossRef
36. Kerrigan DC, Bang M, Burke DT. An algorithm to assess stiff-legged gait in traumatic brain injury. J Head Trauma Rehabil. 1999;14:136–145
    • View In Article
    • MEDLINE
    • CrossRef
37. Niechwiej-Szwedo E, Inness EL, Howe JA, Jaglal S, McIlroy WE, Verrier MC. Changes in gait variability during different challenges to mobility in patients with traumatic brain injury. Gait Posture. 2007;25:70–77
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (201 KB)
    • CrossRef

• a Vicon 512; Oxford Metrics, 14 Minns Business Park, West Way, Oxford, United Kingdom OX2 0JB.
• b AMTI forceplates; Advanced Mechanical Technology Inc, 176 Waltham St, Watertown, MA, 02472.