Inter-limb Coordination in Balance Control: Implications ... · from each limb in upright stance illustrates the importance of the integrity of the central nervous system (CNS) and

Inter-limb Coordination in Balance Control: Implications for

understanding Balance after Traumatic Brain Injury

by

Olinda Dulce Habib Perez

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Rehabilitation Sciences Institute

University of Toronto

© Copyright by Olinda Dulce Habib Perez 2018

ii

Inter-limb Coordination in Balance Control: Implications for

understanding Balance after Traumatic Brain Injury

Olinda Dulce Habib Perez

Doctor of Philosophy

Rehabilitation Sciences Institute

University of Toronto

2018

Abstract

Coordination of forces generated in the lower extremities is essential for maintaining

balance and recovery from postural instability. Temporal synchronization of force amplitude

from each limb in upright stance illustrates the importance of the integrity of the central

nervous system (CNS) and its contribution to the regulation of the centre of mass within its

base of support in producing appropriate coordination between both limbs. Weight-bearing

symmetry is also important to postural control and the maintenance of balance in response to

upcoming perturbations. Therefore, inter-limb coordination in balance control is critical when

the integrity of the CNS has been compromised as a result of a traumatic brain injury (TBI), and

its compromise can lead to increased risk of falling. Balance impairments after TBI are common

and result from multiple TBI-related sequelae, which includes cognitive deficits and

sensorimotor and coordination impairments. Impaired balance performance is typically seen

after TBI; however, the motor contributions that influence postural control are not well

understood in this population. The overall purpose of this dissertation was to further

iii

understand the role of inter-limb coordination in postural control and to determine the extent

to which inter-limb coordination is altered in individuals following moderate-to-severe TBI. The

decoupling of inter-limb synchronization was particularly important in preparation for balance

perturbations that were less predictable and required stepping responses. Balance asymmetry

and poor static balance control were found in individuals with TBI. There was no change in

inter-limb synchronization and as a result of large within-group variability, inter-limb synchrony

post-TBI did not fall outside healthy control norms. Additionally, anticipatory and reactive

balance control mechanisms were also impaired after injury. Although there were

improvements in some balance measures, balance impairments persist one year after injury.

The findings of this dissertation are the first to quantify balance asymmetry after TBI and

provide further evidence that asymmetry and inter-limb coordination are essential components

of balance control after TBI. Furthermore, these findings suggest that future rehabilitation

intervention may benefit from balance training with a focus on asymmetry to reduce of risk of

falling and that other TBI-related sequelae may also be impeding on the recovery of balance

outcomes.

iv

Acknowledgments

This work could not have been completed without the support of many individuals.

Firstly, I would like to thank my supervisor, Dr. George Mochizuki, for his continuing support

and research expertise throughout this journey. I feel fortunate to have had the opportunity to

learn from and work with Dr. Mochizuki prior to beginning my doctoral research – you were my

mentor before you were my supervisor. Thank you for providing me the opportunity in

becoming an independent researcher and re-igniting my passion for research and, most of all,

for teaching me that real life matters. I would also like to thank my committee members, Dr.

Robin Green and Dr. William McIlroy, for their support, insight, and guidance. And, I would like

to thank Drs. George Mochizuki, Robin Green, and William McIlroy for continuously challenging

my scientific thinking throughout our meetings and advancing my writing abilities through their

feedback.

I am very grateful to the many wonderful professors and scientists that I learned from at

the Rehabilitation Sciences Institute, Toronto Rehabilitation Institute, and Sunnybrook Research

Institute. Throughout my graduate training, I have worked alongside exceptional graduate

students and staff. I am thankful to the many members of the Mobility Team who continually

made coming into the lab every day an enjoyable experience. To the present (and former)

students of the GMo lab, it has been a pleasure to share a desk area with you. I would also like

to thank Drs. Jodi Edwards and Jonathan Singer for their encouragement, mentorship, and

allowing me to pick their brain on many matters. I would like to thank and express my upmost

appreciation to Dr. Dina Brooks, whom I have repeatedly called my s-hero and who went above

and beyond her responsibilities as Graduate Coordinator. Thank you for your mentorship and

believing in me when I was least confident.

I would like to thank my wonderful friends, inside and outside of academia, for their

support and encouragement, and reminding me of life outside of graduate school. Thank you to

my family, and in particular my parents, who have always supported my career aspirations with

their unconditional love and encouragement. To my life partner in crime, Nathan Gara, for

v

always supporting my career ambitions through his daily love and patience. Thank you for

sharing a desk with me throughout this journey.

My upmost gratitude goes to the participants in the Toronto Rehabilitation Recovery

Study, who I hope I have advocated for in my doctoral work. I would also like to thank the

following funding agencies for supporting my research throughout my doctoral degree: the

NSERC CREATE Collaborative Academic Rehabilitation Engineering Award (CARE), Ontario

Graduate Scholarship, Toronto Rehabilitation Institute Ontario Student Opportunity Trust Fund,

Ontario Neurotrauma Foundation – Réseau provincial de recherché en adaptation-readaptation

(ONF-REPAR), and Rehabilitation Sciences Institute at the University of Toronto. This thesis

could not have been completed without the many individuals who volunteered their time and

the multiple sources of funding.

vi

Dedication

To my mother and father, who instilled curiosity in me from a young age, and for their ongoing

support and wisdom throughout this journey; and to Nathan, who has always shared my

curious nature and has celebrated the highs and supported me through the lows.

vii

Table of Contents

ACKNOWLEDGMENTS ...................................................................................................................... IV

DEDICATION .................................................................................................................................... VI

TABLE OF CONTENTS ....................................................................................................................... VII

LIST OF TABLES ................................................................................................................................ XI

LIST OF FIGURES .............................................................................................................................. XII

LIST OF ABBREVIATIONS .................................................................................................................. XV

LIST OF APPENDICES ....................................................................................................................... XVI

CHAPTER 1 : INTRODUCTION ..............................................................................................................1

1.1 REFERENCES ........................................................................................................................................4

CHAPTER 2 : BACKGROUND ................................................................................................................5

2.1 BALANCE CONTROL ..............................................................................................................................6

2.1.1 Anticipatory and Reactive Balance Control ........................................................................... 10

2.1.2 Inter-limb Coordination in Balance Control ........................................................................... 13

2.2 TRAUMATIC BRAIN INJURY .................................................................................................................. 16

2.2.1 Prevalence, Incidence, and Mechanism ................................................................................. 16

2.2.2 Features of Traumatic Brain Injury Influence Control of Balance .......................................... 18

2.2.3 Characteristics of Balance Control after Traumatic Brain Injury ........................................... 21

2.3 RATIONALE ...................................................................................................................................... 24

2.3.1 Research Purpose, Questions, and Hypotheses ..................................................................... 29

2.3.2 Thesis Questions .................................................................................................................... 30

2.4 REFERENCES ..................................................................................................................................... 32

CHAPTER 3 : THE EFFECTS OF PREDICTABILITY ON INTER-LIMB POSTURAL SYNCHRONIZATION PRIOR

TO BOUTS OF POSTURAL INSTABILITY ............................................................................................... 45

3.1 ABSTRACT ........................................................................................................................................ 46

3.2 INTRODUCTION ................................................................................................................................. 47

3.3 METHODS ........................................................................................................................................ 49

3.3.1 Participants ............................................................................................................................ 49

3.3.2 Instrumentation and Protocol ............................................................................................... 50

viii

3.3.3 Data Analysis ......................................................................................................................... 55

3.3.4 Secondary Analyses ............................................................................................................... 55

3.3.5 Statistical Analysis ................................................................................................................. 56

3.4 RESULTS .......................................................................................................................................... 57

3.4.1 Inter-limb Synchrony in Preparation for Instability ............................................................... 57

3.4.2 Quiet Standing ....................................................................................................................... 60

3.4.3 Secondary Analysis ................................................................................................................ 60

3.5 DISCUSSION ..................................................................................................................................... 61

3.5.1 Condition-dependent Differences in Preparatory COP .......................................................... 61

3.5.2 Perturbation Type of Preparatory COP Activity ..................................................................... 63

3.6 CONCLUSIONS .................................................................................................................................. 64

3.7 REFERENCES ..................................................................................................................................... 66

CHAPTER 4 : CHARACTERIZATION OF BALANCE CONTROL AFTER MODERATE-TO-SEVERE TRAUMATIC

BRAIN INJURY: A LONGITUDINAL RECOVERY STUDY .......................................................................... 69

4.1 ABSTRACT ........................................................................................................................................ 70

4.2 INTRODUCTION ................................................................................................................................. 72

4.3 METHODS ........................................................................................................................................ 74

4.3.1 Participants ............................................................................................................................ 74

4.3.2 Study Design .......................................................................................................................... 76

4.3.3 Data Collection and Procedures ............................................................................................ 76

4.3.4 Data Processing and Analysis ................................................................................................ 77

4.3.5 Statistical Analyses ................................................................................................................ 78


4.4 RESULTS .......................................................................................................................................... 80

4.4.1 Demographic Characteristics ................................................................................................. 80

4.4.2 Cognitive Characteristics ....................................................................................................... 82

4.4.3 Clinical Measures of Balance ................................................................................................. 82

4.4.4 Net Measures of Balance ....................................................................................................... 83

4.4.5 Measures of Asymmetry ........................................................................................................ 85


4.5 DISCUSSION ..................................................................................................................................... 90

4.5.1 Balance Improvements Within the First Five Months after TBI ............................................. 91

4.5.2 Asymmetry after TBI is evident .............................................................................................. 93

ix

4.5.3 Study Limitations ................................................................................................................... 95

4.6 CONCLUSIONS .................................................................................................................................. 96

4.7 REFERENCES ..................................................................................................................................... 97

CHAPTER 5 : SPECTRAL ANALYSIS OF CENTRE OF PRESSURE IDENTIFIES ALTERED BALANCE CONTROL IN

INDIVIDUALS WITH MODERATE-SEVERE TRAUMATIC BRAIN INJURY ............................................... 101

5.1 ABSTRACT ...................................................................................................................................... 102

5.2 INTRODUCTION ............................................................................................................................... 103

5.3 METHODS ...................................................................................................................................... 106

5.3.1 Participants .......................................................................................................................... 106

5.3.2 Data Collection and Procedure ............................................................................................ 107

5.3.3 Data Analysis ....................................................................................................................... 107

5.3.4 Statistical Analysis ............................................................................................................... 108

5.4 RESULTS ........................................................................................................................................ 109

5.4.1 Spectral Measures of Balance ............................................................................................. 111

5.4.2 Inter-limb Coordination ....................................................................................................... 113

5.4.3 Spatial Measures of Balance ............................................................................................... 118

5.5 DISCUSSION ................................................................................................................................... 120

5.5.1 Anticipatory and Reactive Balance Control Mechanisms are Impaired after TBI................ 120

5.5.2 Inter-limb Coordination after TBI ........................................................................................ 122

5.5.3 The Role of the Postural Challenge ...................................................................................... 124

5.6 CONCLUSION .................................................................................................................................. 125

5.7 REFERENCES ................................................................................................................................... 126

CHAPTER 6 : DISCUSSION ............................................................................................................... 131

6.1 SUMMARY OF FINDINGS ................................................................................................................... 132

6.2 INTER-LIMB COORDINATION IN BALANCE CONTROL .............................................................................. 135

6.2.1 The Organization and Control of Inter-limb Coordination in Balance Control .................... 137

6.2.2 Inter-limb Synchrony in Anticipatory Balance Control......................................................... 138

6.2.3 Inter-limb Coordination in Balance Control after Traumatic Brain Injury ........................... 140

6.2.4 Inter-limb Coordination and Asymmetry: Does it contribute to balance impairment in

traumatic brain injury? .................................................................................................................... 145

6.3 BALANCE RECOVERY AFTER TRAUMATIC BRAIN INJURY .......................................................................... 148

6.4 LIMITATIONS AND FUTURE DIRECTIONS............................................................................................... 151

x

6.5 IMPLICATIONS FOR REHABILITATION ................................................................................................... 155

6.6 SUMMARY AND CONCLUSIONS .......................................................................................................... 157

6.7 REFERENCES ................................................................................................................................... 159

APPENDICES .................................................................................................................................. 169

xi

List of Tables

Table 3-1: Summary of perturbation types and their corresponding responses for each

condition in the experimental paradigm. ..................................................................................... 51

Table 3-2: Mean and standard deviations for external perturbation characteristics. ................. 53

Table 4-1: Participant demographics of traumatic brain injury (TBI) and healthy control (HC)

participants. .................................................................................................................................. 81

Table 4-2: Relationship between inter-limb synchronization (Rxy(0)) and standing balance

measures (centre of pressure root mean square (COP RMS) and absolute stance load symmetry)

for eyes open. ............................................................................................................................... 90

Table 5-1: Mean and standard deviation (SD) participant demographics of traumatic brain injury

(TBI) and healthy control (HC) participants. .............................................................................. 110

xii

List of Figures

Figure 2-1: The theoretical framework of postural control. Postural control emerges from the

interaction of the individual, task, and environment. [From: A. Shumway-Cook & M. Woollacott,

Motor Control: Translating Research into Clinical Practice 4th Ed. (2012)]. ................................. 6

Figure 2-2: The theoretical framework of inter-limb coordination. Inter-limb coordination

emerges from the interaction of the individual, task, and the environment. [Adapted from: A.

Shumway-Cook & M. Woollacott, Motor Control: Translating Research into Clinical Practice 4th

Ed. (2012)]. .................................................................................................................................... 26

Figure 2-3: Theoretical framework of the impairments after traumatic brain injury (TBI).

Irrespective of TBI, cognition, psychiatric, and sensorimotor factors affects balance control. The

bidirectional arrow in the theoretical frameworks demonstrates that balance control can be

considered interrelated with inter-limb coordination. ................................................................ 28

Figure 2-4: Subcomponent of theoretical framework of the sensorimotor impairments after

traumatic brain injury (TBI) that affect balance control and inter-limb coordination that will be

the focus of this dissertation. ....................................................................................................... 29

Figure 3-1: Mean force profile across trials from one participant for external perturbations in

small (S, dash) and large (L, solid) conditions (top). Experimental set up for external

perturbations (bottom). ................................................................................................................ 54

Figure 3-2: Anteroposterior (AP) cross-correlation (Rxy(0)) group means and standard error (SE)

for quiet standing (QS), and all conditions in internal and external perturbations for (A); AP

Rxy(0) individual data for external (B) and internal (C) perturbations. Mediolateral (ML) Rxy(0)

group means and SE for QS, and all conditions in internal and external perturbations for (D); ML

Rxy(0) individual data for external (E) and internal (E) perturbations. Conditions include blocked

(B) and random (R) small (S) and large (L) perturbations (BS, BL, RS, RL). ................................... 58

Figure 3-3: Means and standard error bars from first 5 (First5) and last 5 (Last5) trials in

external (A) and internal (B) perturbations for all participants (n=12). ....................................... 59

xiii

Figure 4-1: Anteroposterior (AP) (A) and mediolateral (ML) (B) centre of pressure root mean

square (COP RMS) group means and standard error for TBI participants across recovery in eyes

open (EO; black) and eyes closed (EC; dark grey). The solid and dashed lines represent mean

and 95% confidence intervals for COP RMS in healthy controls in each condition, respectively.

....................................................................................................................................................... 84

Figure 4-2: Anteroposterior (AP) (A) and mediolateral (ML) (B) inter-limb synchrony (Rxy(0))

group means and standard error for TBI participants across recovery in eyes open (EO; black)

and eyes closed (EC; dark grey). The solid and dashed lines represent mean and 95% confidence

intervals for Rxy(0) in healthy controls in each condition, respectively. ....................................... 86

Figure 4-3: Mean and standard error of absolute stance load symmetry ratio of the less loaded

limb for TBI participants across recovery in eyes open (EO; black) and eyes closed (EC; dark

grey). The solid and dashed lines represent mean and 95% confidence intervals for the absolute

stance load symmetry ratio in healthy controls in each condition, respectively. ........................ 87

Figure 4-4: Time in unipedal stance in the left (UL; black) and right (UR; grey) leg group mean

and standard error for TBI participants across recovery (A). Anteroposterior (AP) (B) and

mediolateral (ML) (C) centre of pressure root mean square (COP RMS) group means and

standard error for TBI participants across recovery in the left (UL; black) and right (UR; grey) leg.

The solid and dashed lines represent mean and 95% confidence intervals in healthy controls in

each condition, respectively. Note that there are no dashed lines for UR in panel A. ................ 89

Figure 5-1: – Net centre of pressure mean integral of the power spectral density (PSD) in low

(A) and high (B) frequencies in the anteroposterior (AP) direction, and the low (C) and high (D)

frequencies in the mediolateral (ML) direction. Group means and standard error for TBI

participants across recovery are displayed for eyes open (EO; black) and eyes closed (EC; grey).

The solid and dashed lines represent mean and 95% confidence intervals for the net COP mean

integral of the PSD in healthy controls in each condition, respectively. ................................... 112

xiv

Figure 5-2: Mean and standard deviation for the net centre of pressure (COP) power spectral

density (PSD) for healthy controls in the anteroposterior (AP) and mediolateral (ML) direction

for eyes open (EO and eyes closed (EC) conditions (top row). Individual data net COP PSD from

a sample of four TBI participants across recovery [T1 (black), T2 (grey), and T3 (light grey)] in the

AP and ML direction in both EO and EC conditions (row 2 to 5). .............................................. 113

Figure 5-3: Mean and standard error (SE) anteroposterior (AP) (A) and mediolateral ML (B)

inter-limb coherence in low frequencies and AP (C) and ML (D) inter-limb coherence in for high

frequencies. Group means ±SE for TBI participants are displayed for eyes open (EO; black) and

eyes closed (EC; grey) conditions across recovery. The solid and dashed lines represent mean

and 95% confidence intervals for the mean inter-limb coherence in healthy controls in each

condition, respectively. .............................................................................................................. 115

Figure 5-4: Individual limb centre of pressure (COP) mean integral of the power spectral density

(PSD) in low (A) and high (B) frequencies in the anteroposterior (AP) direction, and the low (C)

and high (D) frequencies in the mediolateral (ML) direction. Group means and standard error

for traumatic brain injury participants across recovery are displayed across time. Lower weight-

bearing (LWB; solid) and higher weight-bearing (HWB; open) limb are represented in eyes open

(EO; black) and eyes closed (EC; grey). Group means and 95% confidence intervals (depicted

with error bars) are displayed for healthy control (HC) data. ................................................... 117

Figure 5-5: Anteroposterior (AP) (A) and mediolateral (ML) (B) centre of pressure root mean


open (EO; black) and eyes closed (EC; grey) conditions. The solid and dashed lines represent

mean and 95% confidence intervals for COP RMS in healthy controls in each condition,

respectively. ............................................................................................................................... 119




Ed. (2012)]. ................................................................................................................................. 136

xv

List of Abbreviations

ANOVA – Analysis of Variance

AP - Anteroposterior

APA – Anticipatory Postural Adjustment

BOS – Base of Support

CNS – Central Nervous System

COM – Centre of Mass

COP – Centre of Pressure

EO – Eyes Open

EC – Eyes Closed

EPA – Early Postural Adjustments

HC – Healthy Controls

HWB – High Weight-Bearing

ICF – International Classification of

Function, Disability and Health

LWB – Low Weight-Bearing

ML – Mediolateral

PSD – Power Spectral Density

RMS – Root Mean Square

TBI – Traumatic Brain Injury

SMA – Supplementary Motor Area

List of Appendices

Appendix A – Relationship between inter-limb synchrony and mediolateral postural sway

variability

1

Chapter 1 : Introduction

2

Effective balance control requires a multifaceted interaction of neural and

musculoskeletal systems. The interaction of these two systems are responsible for generation

and coordination of forces and torques when controlling the optimal position of the centre of

mass (COM) within the base of support (BOS) during upright balance (Woollacott & Shumway-

Cook, 1996). To maintain quiet upright bipedal stance, the central nervous system (CNS)

generates muscle contractions in the lower extremities that produce adequate forces and

torque from each limb. Relatively equal force production with temporally synchronized

moments from each limb during bipedal stance demonstrates the importance of the integrity of

the CNS and its contribution to the regulation of the COM within the BOS in producing

appropriate coordination between both limbs in healthy individuals and brain injured

populations. Additionally, the ability to coordinate the upcoming decoupling of weight shifting

prior to anticipatory postural adjustments is important for postural stability. Thus, the

coordination of the lower extremities is not only important for quiet standing, but also for the

preparation and reaction for postural instability.

Though maintenance of upright balance appears to be a simple motor task, upright

balance has demonstrated to be a challenging task for individuals post-TBI (Basford et al., 2003;

Geurts, Ribbers, Knoop, & van Limbeek, 1996; Lehmann et al., 1990; Pickett, Radfar-Baublitz,

McDonald, Walker, & Cifu, 2007; Wöber et al., 1993). Traumatic brain injury (TBI) is a leading

cause of disability, which leads to multiple impairments that directly and indirectly affect

balance control (Colantonio, Croxford, Farooq, Laporte, & Coyte, 2009). Furthermore, there is

growing evidence to suggest that individuals with TBI have balance asymmetries and unilateral

motor weakness that may place these individuals at an increased risk of falling (Choi et al.,

3

2012; Drijkoningen et al., 2015; Newton, 1995). Balance control is essential for mobility and

daily tasks. Advancing understanding of the underlying changes in control, such as sensorimotor

components of balance, would enhance knowledge of how weight-bearing asymmetry and

inter-limb coordination are affected after TBI.

While there exist many factors that may contribute to balance control and TBI, including

cognitive, psychiatric, sensorimotor and behavioural components, the primary goal of this

dissertation is to better understand the role of inter-limb coordination in postural control and

to determine the extent to which TBI affects inter-limb coordination. The background literature

of this dissertation begins with an overview of balance control and the inter-limb coordination

in balance control, followed by an outline of TBI and how TBI-related sequelae may impede

balance control. Finally, a description of all studies that have characterized balance after

moderate-to-severe TBI is presented.

4

1.1 References

Basford, J. R., Chou, L. S., Kaufman, K. R., Brey, R. H., Walker, A., Malec, J. F., . . . Brown, A. W.

(2003). An assessment of gait and balance deficits after traumatic brain injury. Archives

of Physical Medicine and Rehabilitation, 84(3), 343-349. doi:10.1053/apmr.2003.50034

Choi, G. S., Kim, O. L., Kim, S. H., Ahn, S. H., Cho, Y. W., Son, S. M., & Jang, S. H. (2012).

Classification of cause of motor weakness in traumatic brain injury using diffusion tensor

imaging. Archives of Neurology, 69(3), 363-367. doi:10.1001/archneurol.2011.1930

Colantonio, A., Croxford, R., Farooq, S., Laporte, A., & Coyte, P. C. (2009). Trends in

hospitalization associated with traumatic brain injury in a publicly insured population,

1992-2002. Journal of Trauma, Injury, Infection and Critical Care, 66(1), 179-183.

doi:10.1097/TA.0b013e3181715d66

Drijkoningen, D., Caeyenberghs, K., Vander Linden, C., Van Herpe, K., Duysens, J., & Swinnen, S.

P. (2015). Associations between Muscle Strength Asymmetry and Impairments in Gait

and Posture in Young Brain-Injured Patients. Journal of Neurotrauma, 32(17), 1324-

1332. doi:10.1089/neu.2014.3787

Geurts, A. C., Ribbers, G. M., Knoop, J. A., & van Limbeek, J. (1996). Identification of static and

dynamic postural instability following traumatic brain injury. Archives of Physical

Medicine and Rehabilitation, 77(7), 639-644.

Lehmann, J. F., Boswell, S., Price, R., Burleigh, A., deLateur, B. J., Jaffe, K. M., & Hertling, D.

(1990). Quantitative evaluation of sway as an indicator of functional balance in post-

traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 71(12), 955-

962.

Newton, R. A. (1995). Balance abilities in individuals with moderate and severe traumatic brain

injury. Brain Injury, 9(5), 445-451.

Pickett, T. C., Radfar-Baublitz, L. S., McDonald, S. D., Walker, W. C., & Cifu, D. X. (2007).

Objectively assessing balance deficits after TBI: Role of computerized postography.

Journal of Rehabilitation Research and Development, 44(7), 983-990.

Wöber, C., Oder, W., Kollegger, H., Prayer, L., Buaumgartner, C., Wöber-Bingöl, Ç., . . . Deecke,

L. (1993). Posturographic measurment of body sway in survivors of severe closed head

injury. Archives of Physical Medicine and Rehabilitation, 74(11), 1151-1156.

Woollacott, M. H., & Shumway-Cook, A. (1996). Concepts and methods for assessing postural

instability Journal of Aging and Physical Activity, 4(3), 214-233.

5

Chapter 2 : Background

6

2.1 Balance Control

A well-established model of balance (postural) control describes postural control as an

interdependent interaction of the individual with the task and the environment (Shumway-

Cook & Woollacott, 2012). Using this model, the organization of postural control is defined by

factors within the individual, the task, and the environment (Shumway-Cook & Woollacott,

2012) (Figure 2-1).

Figure 2-1: The theoretical framework of postural control. Postural control emerges from the

interaction of the individual, task, and environment. [From: A. Shumway-Cook & M. Woollacott,

Motor Control: Translating Research into Clinical Practice 4th Ed. (2012)]

Postural control is a complex task resulting from the combination of the interaction of

multiple sensorimotor processes and the interrelationship of multiple body segments (Horak,

2006; Shumway-Cook & Woollacott, 2012). Additionally, the reception and integration of

sensory information and the output of the motor system are critical for postural control (Lin &

Postural Control

Environment

E

T

Task

I

Individual

7

Woollacott, 2005; Peterka & Loughlin, 2004; Stelmach & Worringham, 1985). The core function

of postural control is the maintenance of equilibrium, which requires the ability to maintain the

centre of mass (COM) within an individual’s base of support (BOS) (Woollacott & Shumway-

Cook, 1996). During quiet standing, the gravitational forces acting on the body are balanced by

the central nervous system (CNS) producing active forces against gravity (Horak & Macpherson,

2011). This is achieved by matching the resultant external torque with the torque of the muscle

acting around lower extremity joints (Balasubramaniam & Wing, 2002). Additionally,

maintaining balance produces minor oscillations that reflect continuous and intermittent

muscle activity (Balasubramaniam & Wing, 2002). These small oscillations are measured with

force plates through the centre of pressure (COP), which is a single point position of the vertical

ground reaction force within the BOS, and represents the average of all the pressures over the

standing surface area (Winter, 1995). During quiet upright bipedal stance, the lower extremities

generate appropriate forces and torque from each limb, from the muscle activations, to

maintain balance by keeping the COM within its BOS.

Upright bipedal stance has been modeled as an inverted pendulum that pivots about

the ankle joint (Winter, 1995). In bipedal stance, a net COP lies between both feet; however,

each foot also possesses its own distinct COP. Moreover, the net COP in single leg standing lies

within the foot in contact with the surface. Net anteroposterior (AP) COP displacements are

produced by ankle dorsiflexors and plantarflexors, while net mediolateral (ML) COP

displacements result from a combination of ankle evertors and invertors, and hip abductors and

adductors. Ankle evertors and invertors are responsible for the individual foot ML COP

displacements, while hip adductors and abductors account for the dynamic control of loading

8

and unloading and have a greater contribution to the net ML COP (Winter, 1995). The net COP

continuously moves with respect to the COM to offset or prevent falling while maintaining

balance, and does so by appropriately generating forces from each limb.

During upright standing, the net COP has been typically analyzed in two domains that

provide insight to the postural sway and balance strategies and mechanisms. The most

commonly reported COP measures include spatial domain measures, which describe the

variability in the COP and depict the amplitude of postural sway (Duarte & Freitas, 2010). The

frequencies of the COP oscillations have been shown to be important as they identify postural

strategies and mechanisms (Winter, Patla, Prince, Ishac, & Gielo-Perczak, 1998; Zatsiorsky &

Duarte, 1999). The COP signal, like most other signals, can be broken down to several simple

sinusoids of different frequencies by taking the signal in the time domain and transforming it

into the frequency domain (Winter & Patla, 1997). This type of transformation is conducted by

applying a Fourier transform, where the results display the amplitude of the signal at distinct

frequencies (Winter & Patla, 1997). COP oscillations that depict postural sway has been

generally characterized as consisting of mostly low frequencies (Zatsiorsky & Duarte, 1999). In

particular, COP oscillations with frequencies less than 0.4 Hz are characterized as low frequency

components and are considered to be exploratory (Carpenter, Murnaghan, & Inglis, 2010;

Zatsiorsky & Duarte, 1999), slow migration of the COP to a reference point from the CNS

(Latash, Ferreira, Wieczorek, & Duarte, 2003), or error corrections to equilibrium (Kiemel, Oie,

& Jeka, 2006), suitable for feedforward information (Carpenter et al., 2010; Gatev, Thomas,

Kepple, & Hallett, 1999). Feedforward control of movement inherently reflects predictive or

anticipatory models of behaviour (Massion, 1992; Wolpert & Kawato, 1998). In contrast, COP

9

oscillations with frequencies greater than 0.4 Hz are characterized as higher frequency

components and have been categorized as corrective in response to temporary instability

(Singer & Mochizuki, 2015), representing reactive control mechanisms (Schinkel-Ivy, Singer,

Inness, & Mansfield, 2016).

Frequency decomposition of the COP has been conducted in neurologically impaired

populations such as individuals post-stroke (Paillex & So, 2003; Schinkel-Ivy et al., 2016; Singer

& Mochizuki, 2015; Yanohara et al., 2014), individuals with Huntington’s disease (Myklebust et

al., 2009), Down syndrome (Rigoldi, Galli, Mainardi, Crivellini, & Albertini, 2011), multiple

sclerosis (Kanekar, Lee, & Aruin, 2014), and the elderly population (Bauer et al., 2010; Kirchner,

Schubert, Getrost, & Haas, 2013; Prieto, Myklebust, Hoffmann, Lovett, & Myklebust, 1996;

Suarez et al., 2013). Specifically, individuals post-stroke demonstrate increased amplitude in the

power spectra throughout various frequency ranges in low and high frequencies (Paillex & So,

2003; Singer & Mochizuki, 2015). In addition, individuals post-stroke who demonstrate an

inability to recover from postural instability using stepping reactions produce increased power

in the COP in high frequencies (Schinkel-Ivy et al., 2016). Older adults with bilateral vestibular

hypofunction demonstrate greater spectral energy in frequencies between 0.1 to 0.78 Hz when

compared to a control group (Suarez et al., 2013). In contrast, individuals with multiple sclerosis

demonstrate a reduced proportion of frequencies produced in low frequencies when compared

to healthy controls (Kanekar et al., 2014). Individuals with Huntington’s Disease (Myklebust et

al., 2009) and healthy older adults (Prieto et al., 1996) both demonstrate increased frequencies

compared to healthy young and older adults in the frequencies bounding the 95% of the power

in the range of 0.15 – 5 Hz. Power spectra across a range of frequencies are further increased

10

when vision is omitted and the BOS is reduced in older adults (Bauer et al., 2010). Together, this

body of literature demonstrates that, in general, individuals with neurological impairments and

older adults displayed increased power spectra and/or increased frequencies during upright

stance. Thus, frequency decomposition of COP oscillations in upright stance can provide insight

into the impairments in anticipatory and reactive control mechanisms in populations that may

be at increased risk of falling such as individuals with TBI.

2.1.1 Anticipatory and Reactive Balance Control

Anticipatory balance control prepares the sensory and motor systems for postural

demands induced by self-generated movements or predictable externally-generated

perturbations in a feedforward manner (Massion, 1994; Wolpert & Kawato, 1998). Anticipatory

balance control is dependent on the physiological readiness of the CNS, also known as ‘central

set’, and is influenced by prior experiences and current context (Prochazka, 1989). Prior

experiences in addition to upcoming stimuli, enables descending commands of the CNS to

prepare and modify contextually appropriate postural responses (Horak, Diener, & Nashner,

1989; Prochazka, 1989). In anticipating postural perturbations, muscle activation can be

initiated prior to the perturbation onset and diminish subsequent instability (Cordo & Nashner,

1982; Horak, Esselman, Anderson, & Lynch, 1984). One important aspect of anticipatory control

is the ability of the CNS to modulate cortical and muscular activity for postural control

strategies (Mochizuki, Boe, Marlin, & McIlRoy, 2010). The role of the CNS in predicting the

postural perturbations that correspond to volitionally planned movements result to two

feedforward postural adjustments, early postural adjustments (EPAs) and anticipatory postural

11

adjustments (APAs) (Krishnan, Latash, & Aruin, 2012). Although EPAs and APAs contribute to

feedforward control, their timing and roles during anticipatory control are different. Recent

literature has demonstrated that EPAs are visible 400-500 ms prior to self-initiated or expected

perturbations by minimizing the effects of upcoming expected perturbations, preceding APAs

(Klous, Mikulic, & Latash, 2012; Krishnan et al., 2012). Conversely, APAs are immediately

detected 100-150 ms prior to volitional postural perturbations in postural muscles and, in turn,

COP displacements (Aruin, 2002; Aruin, Forrest, & Latash, 1998), and are responsible for

generating the forces and moments against expected perturbations (Klous et al., 2012).

The timing and the amplitude of forces produced during anticipatory control plays an

important role in the ability to coordinate between the two limbs. For example, during gait

initiation, APAs are biomechanically essential for ML stability by initially increasing the load of

the swing leg followed by the unloading forces from the swing leg to take a step (Breniere & Do,

1986; Brunt et al., 1991; Jian, Winter, Ishac, & Gilchrist, 1993). The loading and unloading from

muscular activations generates a decoupling motion of the COP and COM in order to propel the

COM forwards and towards the stance leg (Burleigh, Horak, & Malouin, 1994; Fiolkowski, Brunt,

Bishop, & Woo, 2002; Jian et al., 1993). Although APAs occur invariably in gait initiation

(Breniere & Do, 1986; Brunt et al., 1991; Brunt, Liu, Trimble, Bauer, & Short, 1999; Delval et al.,

2012; Jian et al., 1993; MacKinnon et al., 2007), they are also detected in forward and

backwards single step movements (Degani, Donna-Dos-Santos, & Latash, 2007; Gélat &

Breniere, 2000; Ito, Azuma, & Yamashita, 2003), upward leg raises (Hall, Brauer, Horak, &

Hodges, 2010), rising on the toes (Cowan, Hodges, & Bennell, 2001), and asymmetrical

perturbations (Chen, Lee, & Aruin, 2015; Mochizuki, Ivanova, & Garland, 2004). Predictable

12

bilateral and unilateral perturbations that do not require a change in BOS have demonstrated

different postural muscle activations prior to the self-initiated perturbations. Bilateral arm

raises have activated postural bilateral muscles prior to self-initiated bilateral shoulder flexion

perturbations, while unilateral arm perturbations elicit more unilateral muscle activations

(Mochizuki et al., 2004). These examples illustrate the capacity of the CNS to appropriately

synchronize and desynchronize the amplitude of forces produced by the lower extremities in

preparation for the postural task. Therefore, the coordination between the lower limbs is

critical for anticipatory feedforward postural control.

Coordination between the lower extremities is also important for reactive balance

control. Reactive balance is essential for unpredictable perturbations and considered to be a

cortical or high-level cognitive processing that control balance reactions (Jacobs & Horak, 2007;

Maki & McIlroy, 2007). Balance control strategies that occur in response to external

perturbations include the fixed-support and change-in-support strategies. Fixed-support

strategies produces stabilizing joint moments about the ankle or hip to maintain the body’s

COM within the existing BOS with feet-in-place ankle and hip strategies, respectively (Horak &

Nashner, 1996). As feet-in-place strategies require both limbs to maintain contact with the BOS,

temporally synchronized force amplitude is required for ongoing postural stability. In contrast,

change-in-support strategies evoke a hand or foot displacement to expand the size of the BOS

to recapture the COM. Because this strategy generates larger forces, it provides increased

stability (Maki & McIlroy, 1997) and is considered one’s ‘preferred’ response (Jensen, Brown, &

Woollacott, 2001; Maki & McIlroy, 1997). Unlike the invariability of APAs in volitional stepping,

APAs do not always precede perturbation evoked stepping reactions. The attenuation or

13

omission of APAs in compensatory stepping is influenced by the unpredictability of the

perturbation (McIlroy & Maki, 1993). Change-in-support stepping is rapidly generated in

comparison to that of volitional stepping and result in truncated APAs (McIlroy & Maki, 1999).

As a result of these rapid reactions, weight-bearing symmetry may alter reactive balance

control. Weight-bearing asymmetry has demonstrated to decrease stepping thresholds during

lateral perturbations when perturbations were towards the limb with greater load (de Kam,

Kamphuis, Weerdesteyn, & Geurts, 2016). Individuals with balance asymmetry from a brain

injury, such as stroke, have demonstrated impaired reactive balance control and are at

increased risk of falls (Inness, Mansfield, Bayley, & McIlroy, 2016; Inness, Mansfield, Lakhani, &

McIlroy, 2014; Mansfield, Inness, Lakhani, & McIlroy, 2012). Additionally, individuals post-

stroke demonstrated poor coordination between limbs and this has been reflective in reactive

balance control mechanisms (Mansfield, Danells, Inness, Mochizuki, & McIlroy, 2011;

Mansfield, Mochizuki, Inness, & McIlroy, 2012; Schinkel-Ivy et al., 2016). Thus, the ability to

coordinate between the lower limbs in response to unpredictable perturbations is essential to

prevent the risk of falling.

2.1.2 Inter-limb Coordination in Balance Control

Temporally synchronized force amplitude production from each limb during bipedal

stance illustrates the importance of the integrity of the CNS and its contribution to the

regulation of the COM within its BOS in producing appropriate coordination between both

limbs. In quiet standing, individuals with an intact CNS generate relatively equal vertical ground

reaction forces from each limb by distributing approximately 50% of their body weight between

14

limbs (Winter, Prince, Stergiou, & Powell, 1993). This implies that the coordination between the

lower extremities is critical for the task of upright bipedal standing. For the purpose of this

thesis, inter-limb coordination is defined as the ability to produce relatively symmetrical force

amplitudes from each limb by temporally synchronizing ankle and hip moments while standing.

Like the net COP, individual limb COP can be examined in various domains. The

neurophysiological contributions within the COP signal during bipedal stance can be examined

in spatial, temporal, and frequency domains. Spatial domain measures have been used to

characterize postural asymmetry through various computations of individual limb COP and

forces (Duarte & Freitas, 2010). Temporal domain measures provide insight into the

relationships between two signals and have been applied to the COP of each limb and the

muscle activations of different muscles (Mochizuki, Ivanova, & Garland, 2005; Nelson-Wong,

Howarth, Winter, & Callaghan, 2009; Winter et al., 1993). Specifically, temporal relationships

describe the neurophysiological synchronicity between two time-varying signals. In balance

control, this relationship has been characterized as inter-limb synchronization, a depiction of

inter-limb coordination (Nelson-Wong et al., 2009). Inter-limb synchrony is calculated as the

cross-correlation function between two signals as a function of time by iteratively shifting one

signal forwards and backwards in time over the length of the signal (Nelson-Wong et al., 2009;

Singer, Mansfield, Danells, McIlroy, & Mochizuki, 2013; Winter et al., 1993). The magnitude of

the cross-correlation function ranges from -1.0 to 1.0 and illustrates the strength of the

correlation of the COP time series. Postural control in healthy individuals illustrates high in-

phase and anti-phase inter-limb synchrony relationships in the AP and ML direction,

respectively (Winter et al., 1993) with little to no phase lags between limbs (Mochizuki et al.,

15

2005; Winter et al., 1993). During quiet standing, a reduction in the cross-correlation function

suggests a reduced relationship or coupling between the left and right limb, which in turn

would indicate poor inter-limb coordination. Neurologically impaired populations with weight-

bearing asymmetry have demonstrated reduced inter-limb synchrony (Mansfield et al., 2011;

Mansfield, Mochizuki, et al., 2012; Singer et al., 2013) and the two measures have been linked

in post-stroke individuals (Mansfield et al., 2011). This highlights the importance of weight-

bearing symmetry for inter-limb coordination. In some cases, individuals with weight-bearing

asymmetry display an offset in the relationship between the left and right foot COP, which

results in a phase lag where one COP signal ‘leads’ the other COP signal. Though

electromechanical delay exists between muscle activation and force production (Nelson-Wong

et al., 2009), the interpretation of the phase-lag between the left and right COP signals found in

individuals with weight-bearing asymmetry post-stroke remains unclear (Mansfield et al.,

2011). Accordingly, a decrease in the relationship between the left and right COP resulting in

reduced inter-limb synchrony, despite a phase-lag between the signals, would indicate poor

inter-limb coordination.

More recently, balance asymmetry and inter-limb coordination have been quantified in

the spectral domain by analyzing the frequency relationship of the COP from each limb through

coherence analysis (Myklebust et al., 2009). Analogous to inter-limb synchrony, the coherence

of two time-varying signals range from 0 to 1.0 and can also be considered a representation of

inter-limb coordination indicating the degree of correlation between signals at each frequency.

According to Myklebust et al. (2009), individual limb COP is produced by the same postural

control system by the CNS and should be highly correlated. Reduced inter-limb coherence

16

would reflect disorders in the central postural control system and/or additionally peripherally

introduced inputs or noise. Inter-limb COP coherence in healthy young and elderly individuals

was compared to individuals with Huntington’s Disease, and the study found greater coherence

in the healthy groups in comparison to the neurologically impaired population (Myklebust et

al., 2009). Thus, a reduction in inter-limb coherence, like in inter-limb synchrony, would be

indicative of poor inter-limb coordination in relation to the postural control system that drives

the frequencies that control anticipatory and reactive control mechanisms. The combination of

spatial, temporal, and spectral analysis may provide much insight to postural asymmetry and

inter-limb coordination in postural control.

2.2 Traumatic Brain Injury

2.2.1 Prevalence, Incidence, and Mechanism

Traumatic brain injury (TBI) is defined as a disruption in the normal function of the brain

as a result of blunt or penetrating trauma or from acceleration and/or deceleration forces

(Coronado et al., 2012) and is one of the leading causes of disability in North America (Centers

for Disease Control and Prevention, 2006; Colantonio, Croxford, Farooq, Laporte, & Coyte,

2009). In the United States, approximately 16% of emergency visits each year have a primary or

secondary diagnosis of TBI (Centers for Disease Control and Prevention, 2015). In Canada, it is

estimated that 3710 men and 1240 women per 100,000 who have experienced a TBI are

currently living in long-term care facilities, and the projected incidence rate of TBI over the next

20 years is expected to increase by 28% to 70 new cases per 100,000 per year (Public Health

Agency of Canada, 2014). The prevalence of adults with TBI in Canada is bimodal, with peaks in

17

young and older adults aged 20-34 and older than 75 years, respectively (Public Health Agency

of Canada, 2014). The most commonly reported mechanisms of TBI are falls, being struck by an

object, and motor vehicle incidents (Centers for Disease Control and Prevention, 2015;

Colantonio et al., 2010).

TBI occurs when the head or brain undergoes a transfer of kinetic energy through a

rapid acceleration and/or deceleration or from direct impacts to the skull, resulting in focal

coup and contre-coup injuries as well as diffuse injuries, including traumatic axonal injury

(Gaetz, 2004; Povlishock & Katz, 2005). When the brain experiences moderate-to-severe

acceleration and deceleration forces, the combination of focal and diffuse injuries is common

(Gaetz, 2004). Focal coup injuries are generally associated with acceleration trauma, where

contusions, haemorrhages and/or haematomas appear beneath the point of impact in the

either extradural, subarachnoid, subdural and intracerebral areas (Gaetz, 2004). Contre-coup

injuries are associated with acceleration and deceleration trauma and typically occur on the

opposite side of the impact when the cerebral cortex makes contact with the cranium (Gaetz,

2004; Shaw, 2002). Focal injuries result in neuronal damage at the site of injury and to the

surrounding cerebrovasculature and are considered primary injuries of TBI. Secondary injuries

from focal trauma result from ischemia, extracellular glutamate neurotoxicity, and cytotoxic

edema (Gaetz, 2004). Furthermore, the effects of acceleration and deceleration forces result in

widespread axonal tissue damage or lesions, known as diffuse or traumatic axonal injury of the

white matter (Gaetz, 2004; McGinn & Povlishock, 2015). The spectrum of TBI from focal and

diffuse axonal injury leads to cell injury, dysfunction or neuronal death with neuropathological

changes that affect neurotransmitter regulation (McGinn & Povlishock, 2015) and white matter

18

connectivity (Sharp, Scott, & Leech, 2014; Shumskaya, van Gerven, Norris, Vos, & Kessels,

2017). Consequently, TBI leads to cognitive impairments, psychiatric deficits, sensory

impairments, motor dysfunction, and behavioural changes (Iaccarino, Bhatnagar, & Zafonte,

2015; Walker & Pickett, 2007).

2.2.2 Features of Traumatic Brain Injury Influence Control of Balance

There are number of impairments that occur after TBI, including but not limited to,

cognitive deficits, psychiatric deficits, behavioural changes, motor impairments, and balance

impairments. According to the International Classification of Function, Disability and Health

(ICF), impairments are related to the physiological or structural changes of the body (de Kleijn-

de Vrankrijker, 2003; Laxe et al., 2014). Given the scope of this thesis, the current section will

describe the impairments and dysfunctions after TBI that specifically relate to balance control.

Balance impairments after TBI are commonly reported and persist many years after injury

(Ponsford et al., 2014; Walker & Pickett, 2007). Given the pathophysiology of TBI, balance

impairments in individuals with TBI likely stem from dysfunction in multiple body systems (e.g.

sensory, motor, cognitive, soft tissue) (Campbell & Parry, 2005). Though balance control may

appear to be a simple motor ability, research has recognized that balance control uses higher-

level cortical resources (Shumway-Cook & Woollacott, 2000; Varghese, Beyer, Williams,

Miyasike-daSilva, & McIlroy, 2015), and thus, can be quite challenging after TBI. Balance control

engages cortical and subcortical structures in static balance control (Varghese et al., 2015) and

dynamic balance control (Jacobs & Horak, 2007; Maki & McIlroy, 2007; Varghese et al., 2015).

Neurological insults like TBI negatively affect balance control by altering activation in brain

19

networks, compromising the integrity of white matter connections, and altering the properties

of the CNS that are responsible for muscle activation and sensorimotor feedback (Di Russo,

Incoccia, Formisano, Sabatini, & Zoccolotti, 2005; Ghajar & Ivry, 2008).

A number of studies have reported that individuals with moderate-to-severe TBI

experience a range of cognitive dysfunctions including attention deficits (Owens et al., 2017;

Shah et al., 2017), reduced concentration (Barman, Chatterjee, & Bhide, 2016; Belmont, Agar, &

Azouvi, 2009), executive function deficits (Merkley, Larson, Bigler, Good, & Perlstein, 2013;

Stuss, 2011; Zimmermann et al., 2015), memory deficits (Azouvi, Vallat-Azouvi, & Belmont,

2009; Slovarp, Azuma, & Lapointe, 2012; Vakil, 2005), and reduced speed of processing

(Dymowski, Owens, Ponsford, & Willmott, 2015; Madigan, DeLuca, Diamond, Tramontano, &

Averill, 2000; Willmott, Ponsford, Hocking, & Schonberger, 2009). The consequences of

cognitive impairments after TBI may have an implication on balance control as attention,

executive function, and visuo-spatial memory are important for balance control (Dalton,

Sciadas, & Nantel, 2016; Little & Woollacott, 2014; Muir-Hunter et al., 2014; van Iersel, Kessels,

Bloem, Verbeek, & Olde Rikkert, 2008). Studies have shown that performing cognitive tasks

while maintaining balance increased or decreased postural sway (Shumway-Cook & Woollacott,

2000; Siu & Woollacott, 2007; Vuillerme & Vincent, 2006; Woollacott & Shumway-Cook, 2002),

increased muscle latencies to perturbations (Anan, Fujiwara, Yaguchi, & Kiyota, 2014),

decreased muscle amplitudes (Rankin, Woollacott, Shumway-Cook, & Brown, 2000), and

increased reaction time for postural strategies (Remaud, Boyas, Caron, & Bilodeau, 2012) and

stepping responses (Uemura, Oya, & Uchiyama, 2013). In addition to cognitive dysfunctions,

psychiatric factors, such as depression and anxiety are also common after TBI (Bombardier et

20

al., 2010; Failla, Juengst, Graham, Arenth, & Wagner, 2016; Guillamondegui et al., 2011; Hart et

al., 2016; Jorge & Arciniegas, 2014; Khan et al., 2016; Lewis & Horn, 2017; Osborn, Mathias,

Fairweather-Schmidt, & Anstey, 2016). Depression and increased level of anxiety in other

populations have been shown to negatively impact balance control (Adkin, Frank, Carpenter, &

Peysar, 2000; Carpenter, Frank, & Silcher, 1999; Carpenter, Frank, Silcher, & Peysar, 2001;

Casteran, Putot, Pfitzenmeyer, Thomas, & Manckoundia, 2016; Sturnieks, Delbaere, Brodie, &

Lord, 2016). These cognitive and psychiatric sequelae of TBI have implications in balance

control, and in turn, may place this population at increased fall risk (McKechnie, Pryor, & Fisher,

2015). While these factors are important in fully understanding the relationship between TBI

and balance, they are not the focus of this dissertation.

Sensorimotor impairments are common after TBI (Debert, Herter, Scott, & Dukelow,

2012; Shumskaya et al., 2017) and are among the multi-systemic body impairments that

contribute to increased risk of falling after TBI (McKechnie, Pryor, & Fisher, 2017). When

function of either system is disrupted as a result of a TBI, motor impairments, including

strength imbalances between limbs and coordination deficits (Khan, Baguley, & Cameron, 2003;

Walker & Pickett, 2007) may become apparent. Unilateral motor weakness, balance

asymmetries and poor inter-limb coordination have been documented following TBI (Arce, Katz,

& Sugarman, 2004; Choi et al., 2012; Drijkoningen et al., 2015; Newton, 1995). Muscle strength

asymmetry has been associated with poor balance performance and greater step time

variability during mobility (Drijkoningen et al., 2015). Greater attention has been placed on the

implications of gait asymmetry after TBI (Ochi, Esquenazi, Hirai, & Talaty, 1999; Vasudevan,

21

Glass, & Packel, 2014; Williams, Morris, Schache, & McCrory, 2009); however, the extent of

balance asymmetry after TBI has as not yet been quantified.

2.2.3 Characteristics of Balance Control after Traumatic Brain Injury

Several studies have characterized features of balance control after TBI using a number

of different approaches. One of the first studies quantifying postural control after TBI

demonstrated that path length and radial displacement of the COP in TBI participants was

greater than healthy controls by almost two-fold. The authors reported that individuals with

greater sway in static balance assessments exhibited a tendency to fall (Lehmann et al., 1990).

Additionally, greater COP radial displacements were found post-TBI in balances tasks where

visual information was omitted suggesting that individuals with TBI have an increased reliance

of visual information than healthy controls (Lehmann et al., 1990). Importantly, this study also

identified strong test – retest reliability (0.72 – 0.99) of COP radial displacement and path

length, implying consistency of the measures and patient characteristics. Similarly, Wöber et al.

(1993) analyzed the association between the depth of brain lesions in individuals with TBI using

magnetic resonance imaging and subsequent postural imbalance. Postural imbalance was

calculated as the Chi-Square of clinical markers of severity, including the Glasgow Coma Scale

and Post Traumatic Amnesia, and posturographic measures of sway. The study found positive

correlations between the depth of brain lesions and the severity of subsequent postural

imbalance, indicating that individuals with TBI with deeper lesions were more likely to produce

greater postural sway (Wöber et al., 1993). Also, individuals with a Glasgow Coma Scale score of

less than 7 demonstrated an increased likelihood of postural imbalance (Wöber et al., 1993). In

22

a subsequent study, static and dynamic balance control was analyzed after mild to severe TBI in

various task conditions and compared to healthy controls (Geurts, Ribbers, Knoop, & van

Limbeek, 1996). Additionally, the weight shifting was analyzed with a visual feedback balance

task. The study found that after six months of injury, TBI participants had greater COP RMS in

both the AP and ML directions than healthy controls, in particular with the omission of vision.

TBI produced less weight-bearing balance tasks than healthy controls, demonstrating difficulty

with this task. Thus, these two studies demonstrated that static and dynamic balance control

were impaired six months post-TBI (Geurts et al., 1996).

Other posturographic measures of balance have been reported throughout the

literature. Basford et al. (2003) used sensory organization tests to analyze static balance in a

small sample of individuals with mild to severe TBI. Overall sensory organization tests scores

demonstrated that TBI participants who self-reported balance instability demonstrated lower

sensory organization test scores when compared to healthy controls (Basford et al., 2003), with

lower sensory organization test scores indicative of poorer balance control. Basford et al.

(2003) also demonstrated vestibular impairment in the majority of TBI participants, using

caloric irrigation tests. However, because the overall sensory organized tests scores were

reported, vestibular function during balance control remained unknown (Basford et al., 2003).

The study suggested that subtle complaints of postural instability from individuals with TBI

need further investigation (Basford et al., 2003). Balance control was assessed in individuals

with severe TBI undergoing inpatient rehabilitation using the sensory organization test and

found that in comparison to normative values, individuals with TBI had difficulty with

somatosensory and vestibular probing conditions (Pickett, Radfar-Baublitz, McDonald, Walker,

23

& Cifu, 2007). Thus, despite having good functional independence, other sensory impairments

may result in reduced postural stability after TBI.

Though cross-sectional studies provide an overview of a population at one specific time

point, studies that look at changes over time can provide insight to the recovery of balance

control. Static balance has been analyzed over a short period of time (i.e. two to six weeks)

within an inpatient rehabilitation setting in individuals with moderate-to-severe TBI (Wade,

Canning, Fowler, Felmingham, & Baguley, 1997). This study found that postural sway decreased

between the two assessments, as depicted by sway index, as a result of inpatient rehabilitation

(Wade et al., 1997). Though Wade et al. (1997) demonstrated an improvement in balance after

TBI, it remains unknown whether improvement persists over the course of recovery. Walker

and Picket (2007) studied motor impairments after severe TBI over the course of two years. The

study found that two years following acute rehabilitation, approximately one third of

individuals exhibited neuromotor abnormalities in neurological examinations and tandem gait

(Walker & Pickett, 2007). More recently, Ponsford et al. (2014) reported functional outcomes in

mild to severe TBI at three time points over a span of 10 years. In their larger cohort of 141

individuals with TBI, over 60% of individuals with TBI reported balance problems two years after

injury; however, this dropped to 40% by five years after injury and then rose once again to 55%

10 years post injury. The study found that individuals with moderate-to-severe injuries were

more likely to report balance impairment 10 years post injury in comparison to individuals with

mild TBI (Ponsford et al., 2014). Though studies such as Ponsford et al. (2014) and Walker and

Pickett (2007) examined a large sample size longitudinally, their measure to characterize

postural instability included a neurological examination (i.e. Romberg test), rather than a

24

quantitative measure that provides a more detailed characterization of the specific impairment.

Therefore, there is a need to understand balance control using quantitative measures of

balance across the course of recovery and compare to a healthy control group to understand

where along recovery they begin to resemble normative values.

In summary, few studies have analyzed posturographic measures in individuals with TBI

(Brauer, Broome, Stone, Clewett, & Herzig, 2004; Geurts et al., 1996; Lehmann et al., 1990;

Wade et al., 1997). Greater postural sway has been linked with individuals with increased TBI

severity (Wöber et al., 1993). Evidence has identified balance impairments up to 10 years post-

injury through self-report and neurological examinations (Ponsford et al., 2014), with a couple

of studies identifying impaired dynamic balance (Arce et al., 2004; Geurts et al., 1996).

Therefore, there is further need to quantify postural control across recovery in individuals with

moderate-to-severe TBI and identify the impairments of postural control mechanisms in this

population.

2.3 Rationale

As previously stated, postural control occurs from an interaction of the individual with

the task and the environment (Shumway-Cook & Woollacott, 2012). In recent years,

measurement of individual limb contributions to balance (inter-limb coordination), has been a

useful approach for characterizing balance dyscontrol and neurological asymmetries after

neurological injury (i.e. stroke) in static balance (Mansfield et al., 2011; Mansfield, Mochizuki, et

al., 2012; Singer et al., 2013), and TBI during dynamic balance tasks (Arce et al., 2004).

25

However, inter-limb coordination within other components of postural control, such as

anticipatory and reactive balance control, and after TBI during static balance tasks and has not

been considered. This represents an important gap in the literature because weight-bearing

asymmetry has been linked to inter-limb coordination measures in individuals with neurological

injuries (Mansfield et al., 2011), but has not been examined in individuals post-TBI.

Recent studies have demonstrated unilateral motor weakness (Choi et al., 2012;

Drijkoningen et al., 2015) that persists following TBI. Additionally, qualitative observations of

asymmetrical stance prior to platform perturbations (Newton, 1995) and poor inter-limb force

amplitude coordination during bimanual load lifting have been reported in individuals with TBI

(Arce et al., 2004). Drijkoningen et al. (2015) found a negative relationship between leg muscle

strength asymmetry and decreased balance stability, indicating greater asymmetrical muscle

strength was associated with poorer balance performance. Additionally, increased leg muscle

strength asymmetry was associated with increased step time variability during gait

(Drijkoningen et al., 2015). Greater emphasis has been placed on the implications of gait

asymmetry after TBI (Ochi et al., 1999; Vasudevan et al., 2014; Williams et al., 2009); however,

balance asymmetry after TBI has as yet to be quantified and the implications on balance control

are unknown. With the current evidence that motor and balance asymmetries occur after TBI,

inter-limb coordination during static balance tasks may be an important determinant in

understanding recovery of balance impairments after TBI. Therefore, postural control and inter-

limb coordination are to be considered interrelated within the Shumway-Cook and Woollacott

(2012) conceptual model, and inter-limb coordination can be assumed to have a similar

interaction between the individual, the task and the environment (Figure 2-2).

26




Ed. (2012)]

After TBI, balance impairments are commonly reported and persist many years after

injury (Ponsford et al., 2014; Walker & Pickett, 2007). The literature on postural control has

identified the importance of cortical resources for static and dynamic control (Jacobs & Horak,

2007; Richard et al., 2017; Siu & Woollacott, 2007; Woollacott & Shumway-Cook, 2002). Though

balance impairments after TBI may arise from various mechanisms of injury, the high incidence

of focal and diffuse injuries and edema from TBI (Gaetz, 2004) have the potential to impair the

cortical structures and connective white matter, ultimately impacting postural control after

Inter-limb

Coordination

Environment

E

T

Task

I

Individual

27

injury. Furthermore, previous studies have proposed that predictive or anticipatory control is

impaired after TBI (Ghajar & Ivry, 2008), and this may transfer to balance control.

Given that inter-limb coordination is not well understood in dynamic balance tasks and

these tasks may be impaired after TBI, it is important to establish an understanding of inter-

limb coordination and dynamic balance control in healthy individuals. Thus, for the purpose of

this thesis, healthy individuals and individuals with TBI will be studied. Additionally, because of

the large gap in the literature quantifying postural control post-TBI throughout the course of

recovery using quantitative balance measures, it is imperative to advance the understanding of

these features as they relate to TBI. As previously described in the literature, cognitive,

psychiatric, and sensorimotor components affect balance control (Figure 2-3). Irrespective of

TBI, there is evidence to suggest that cognition, psychiatric, and sensorimotor factors impact

various aspects of balance (postural) control. The focus of this thesis is on the sensorimotor

components that affect balance control after TBI (Figure 2-4). Therefore, this thesis will

establish new knowledge about the control of balance and inter-limb coordination, changes in

postural control across recovery after TBI, and the mechanisms that contribute to altered

postural control following TBI.

28

Figure 2-3: Theoretical framework of the impairments after traumatic brain injury (TBI).

Irrespective of TBI, cognition, psychiatric, and sensorimotor factors affects balance control. The

bidirectional arrow in the theoretical frameworks demonstrates that balance control can be

considered interrelated with inter-limb coordination.

29

Figure 2-4: Subcomponent of theoretical framework of the sensorimotor impairments after

traumatic brain injury (TBI) that affect balance control and inter-limb coordination that will be

the focus of this dissertation.

2.3.1 Research Purpose, Questions, and Hypotheses

The overall purpose of this thesis was to better understand the role of inter-limb

coordination in postural control and to determine the extent to which inter-limb coordination is

altered in individuals following moderate-to-severe TBI. Inter-limb coordination was

characterized with temporal and spectral balance measures from each limb.

30

2.3.2 Thesis Questions

To meet the overall purpose, this thesis was divided into three studies, looking at the role of

inter-limb coordination in postural control and how it changes after a moderate-to-severe TBI.

1) How does the predictability of postural instability affect inter-limb coordination? Inter-

limb coordination, assessed using COP temporal synchronization between limbs, was

analyzed prior to bouts of temporally predictable and internally and externally applied

postural instability. It was hypothesized that (1) inter-limb synchrony would be highest

when perturbation magnitudes were predictable and of low magnitude; (2) inter-limb

synchrony would be higher in internally generated perturbations; and, (3) inter-limb

synchrony would change from the start to the end of repeated exposure to

perturbations when perturbation magnitudes were known.

2) How does balance control and inter-limb coordination change across recovery after

moderate-to-severe TBI? Inter-limb coordination and asymmetry were examined after

moderate-to-severe TBI comparing balance tasks with varying levels of difficulty. Spatial

measures of net COP and temporal measures of individual limb COP were used to

characterize asymmetry and inter-limb coordination after moderate-to-severe TBI

across recovery. It was hypothesized that (1) net balance measures would improve over

the course of recovery; (2) individuals with TBI would demonstrate postural asymmetry

in measures of asymmetry, including inter-limb synchrony, stance load symmetry, and

postural sway and time in unipedal stance; and (3) measures of postural asymmetry

would improve across recovery.

3) How do anticipatory and reactive balance control mechanisms after moderate-to-severe

TBI change during recovery? By applying spectral analysis on the individual limb COP

31

and net COP, this study informed the conceptual model about the frequencies of sway

associated with anticipatory and reactive mechanisms at different stages of recovery. It

was hypothesized that (1) spectral measures of net COP would improve across time in

individuals with TBI; (2) improvements across time in inter-limb coherence in individuals

with TBI would be detectable; and (3) the individual limb applying the most vertical

force (i.e., highest weight-bearing) would demonstrate increased COP power spectral

density and decrease across recovery.

32

2.4 References

Adkin, A. L., Frank, J. S., Carpenter, M. G., & Peysar, G. W. (2000). Postural control is scaled to

level of postural threat. Gait and Posture, 12(2), 87-93.

Anan, K., Fujiwara, K., Yaguchi, C., & Kiyota, N. (2014). Effect of time pressure on attentional

shift and anticipatory postural control during unilateral shoulder abduction reactions in

an oddball-like paradigm. Journal of Psychological Anthropology, 33(1), 17.

Arce, F. I., Katz, N., & Sugarman, H. (2004). The scaling of postural adjustments during bimanual

load-lifting in traumatic brain-injured adults. Human Movement Science, 22(6), 749-768.

doi:10.1016/j.humov.2003.12.002

Aruin, A. S. (2002). The organization of anticipatory postural adjustments. Journal of Automatic

Control, 12(1), 31-37.

Aruin, A. S., Forrest, W. R., & Latash, M. L. (1998). Anticipatory postural adjustments in

conditions of postural instability Electroencephalography and Clinical Neurophysiology,

109(4), 350-359.

Azouvi, P., Vallat-Azouvi, C., & Belmont, A. (2009). Cognitive deficits after traumatic coma.

Progress in Brain Research, 177, 89-110.

Balasubramaniam, R., & Wing, A. M. (2002). The dynamics of standing balance. TRENDS of

Cognitive Science, 6(12), 531-536.

Barman, A., Chatterjee, A., & Bhide, R. (2016). Cognitive Impairment and Rehabilitation

Strategies After Traumatic Brain Injury. Indian Journal of Psychological Medicine, 38(3),

172-181.




Bauer, C., Groger, I., Rupprecht, R., Meichtry, A., Tibesku, C. O., & Gassmann, K. G. (2010).

Reliability analysis of time series force plate data of community dwelling older adults.

Archives of Gerontology and Geriatrics, 51(3), e100-105.

doi:10.1016/j.archger.2010.01.009

Belmont, A., Agar, N., & Azouvi, P. (2009). Subjective fatigue, mental effort, and attention

deficits after severe traumatic brain injury. Neurorehabilitation and Neural Repair, 23(9),

939-944.

Bombardier, C. H., Fann, J. R., Temkin, N. R., Esselman, P. C., Barber, J., & Dikmen, S. S. (2010).

Rates of major depressive disorder and clinical outcomes following traumatic brain

injury. JAMA, 303(19), 1938-1945. doi:10.1001/jama.2010.599

33

Brauer, S. G., Broome, A., Stone, C., Clewett, S., & Herzig, P. (2004). Simplest tasks have greatest

dual task interference with balance in brain injured adults. Human Movement Science,

23(3-4), 489-502.

Breniere, Y., & Do, M. C. (1986). When and how does steady state gait movement induced from

upright posture begin? Journal of Biomechanics, 19(12), 1035-1040.

Brunt, D., Lafferty, M. J., McKeon, A., Goode, B., Mulhausen, C., & Polk, P. (1991). Invariant

characteristics of gait initiation. American Journal of Physical Medicine and

Rehabilitation, 70(4), 206-212.

Brunt, D., Liu, S., Trimble, M., Bauer, J., & Short, M. (1999). Principles underlying the

organization of movement initiation from quiet stance. Gait and Posture, 10(2), 121-

128.

Burleigh, A. L., Horak, F. B., & Malouin, F. (1994). Modification of postural response and step

initiation: Evidence for goal-directed postural interactions. Journal of Neurophysiology,

72(6), 2892-2902.

Campbell, M., & Parry, A. (2005). Balance disorder and traumatic brain injury: preliminary

findings of a multi-factorial observational study. Brain Injury, 19(13), 1095-1104.

doi:10.1080/02699050500188898

Carpenter, M. G., Frank, J. S., & Silcher, C. P. (1999). Surface height effects on postural control:

A hypothesis for a stiffness strategy for stance. Journal of Vestibular Research, 9(4), 277-

286.

Carpenter, M. G., Frank, J. S., Silcher, C. P., & Peysar, G. W. (2001). The influence of postural

threat on the control of upright stance. Experimental Brain Research, 138(2), 210-218.

doi:10.1007/s002210100681

Carpenter, M. G., Murnaghan, C. D., & Inglis, J. T. (2010). Shifting the balance: evidence of an

exploratory role for postural sway. Neuroscience, 171(1), 196-204.

doi:10.1016/j.neuroscience.2010.08.030

Casteran, M., Putot, A., Pfitzenmeyer, F., Thomas, E., & Manckoundia, P. (2016). Analysis of the

impact of a cognitive task on the posture of elderly subjects with depression compared

with healthy elderly subjects. Clinical Neurophysiology, 127(11), 3406-3411.

doi:10.1016/j.clinph.2016.09.007

Centers for Disease Control and Prevention. (2006). Incidence rates of hospitalization related to

traumatic brain injury--12 states, 2002. MMWR: Morbidity and Mortality Weekly Report,

55(8), 201-214.

Centers for Disease Control and Prevention. (2015). Report to Congress on Traumatic Brain

Injury in the United States: Epidemiology and Rehabilitation. Atlanta, GA.

34

Chen, B., Lee, Y. J., & Aruin, A. S. (2015). Anticipatory and compensatory postural adjustments

in conditions of body asymmetry induced by holding an object. Experimental Brain

Research, 233(11), 3087-3096. doi:10.1007/s00221-015-4377-7







doi:10.1097/TA.0b013e3181715d66

Colantonio, A., Saverino, C., Zagorski, B., Swaine, B., Lewko, J., Jaglal, S., & Vernich, L. (2010).

Hospitalizations and Emergency Department Visits for TBI in Ontario. The Canadian

Journal of Neurological Sciences, 37(6), 783-790.

Cordo, P. J., & Nashner, L. M. (1982). Properties of Postural Adjustments Associated with Rapid

Arm Movements. Journal of Neurophysiology, 47(2), 287-302.

Coronado, V. G., McGuire, L. C., Sarmiento, K., Bell, J., Lionbarger, M. R., Jones, C. D., . . . Xu, L.

(2012). Trends in Traumatic Brain Injury in the U.S. and the public health response:

1995-2009. Journal of Safety Research, 43(4), 299-307. doi:10.1016/j.jsr.2012.08.011

Cowan, S. M., Hodges, P. W., & Bennell, K. L. (2001). Anticipatory activity of vastus lateralis and

vastus medialis obliquus occurs simultaneously in voluntary heel and toe raises. Physical

Therapy in Sport 2(2), 71-79. doi:10.1054/ptsp.2001.0056

Dalton, C., Sciadas, R., & Nantel, J. (2016). Executive function is necessary for the regulation of

the stepping activity when stepping in place in older adults. Aging Clinical and

Experimental Research, 28(5), 909-915. doi:10.1007/s40520-015-0499-9

de Kam, D., Kamphuis, J. F., Weerdesteyn, V., & Geurts, A. C. (2016). The effect of weight-

bearing asymmetry on dynamic postural stability in healthy young individuals. Gait and

Posture, 45, 56-61. doi:10.1016/j.gaitpost.2016.01.004

de Kleijn-de Vrankrijker, M. W. (2003). The long way from the International Classification of

Impairments, Disabilities and Handicaps (ICIDH) to the International Classification of

Functioning, Disability and Health (ICF). Disability and Rehabilitation, 25(11-12), 561-

564. doi:10.1080/09638280110110879

Debert, C. T., Herter, T. M., Scott, S. H., & Dukelow, S. (2012). Robotic assessment of

sensorimotor deficits after traumatic brain injury. Journal of Neurologic Physical

Therapy, 36(2), 58-67. doi:10.1097/NPT.0b013e318254bd4f

Degani, A. M., Donna-Dos-Santos, A., & Latash, M. L. (2007). Postural Preparation to Making a

Step:

35

Is There a “Motor Program” for Postural Preparation? Journal of Applied Biomechanics, 23(4),

261-274.

Delval, A., Dujardin, K., Tard, C., Devanne, H., Willart, S., Bourriez, J. L., . . . Defebvre, L. (2012).

Anticipatory postural adjustments during step initiation: elicitation by auditory

stimulation of differing intensities. Neuroscience, 219, 166-174.


Di Russo, F., Incoccia, C., Formisano, R., Sabatini, U., & Zoccolotti, P. (2005). Abnormal motor

preparation in severe traumatic brain injury with good recovery. Journal of

Neurotruama, 22(2), 297-312.




1332. doi:10.1089/neu.2014.3787

Duarte, M., & Freitas, S. M. (2010). Revision of posturography based on force plate for balance

evaluation. Revista Brasileira de Fisioterapia, 14(3), 183-192.

Dymowski, A. R., Owens, J. A., Ponsford, J. L., & Willmott, C. (2015). Speed of processing and

strategic control of attention after traumatic brain injury. Journal of Clinical

Experimental Neuropsychology, 37(10), 1024-1035.

doi:10.1080/13803395.2015.1074663

Failla, M. D., Juengst, S. B., Graham, K. M., Arenth, P. M., & Wagner, A. K. (2016). Effects of

Depression and Antidepressant Use on Cognitive Deficits and Functional Cognition

Following Severe Traumatic Brain Injury. Journal of Head Trauma Rehabilitation, 31(6),

E62-E73. doi:10.1097/HTR.0000000000000214

Fiolkowski, P., Brunt, D., Bishop, M., & Woo, R. (2002). Does postural instability affect the

initiation of human gait? Neuroscience Letters, 323(3), 167-170.

Gaetz, M. (2004). The neurophysiology of brain injury. Clinical Neurophysiology, 115(1), 4-18.

doi:10.1016/s1388-2457(03)00258-x

Gatev, P., Thomas, S., Kepple, T., & Hallett, M. (1999). Feedforward ankle strategy of balance

during quiet stance in adults. Journal of Physiology, 514(3), 915-928.

Gélat, T., & Breniere, Y. (2000). Adaptation of the gait initiation process for stepping on to a

new level using a single step. Experimental Brain Research, 133(4), 538-546.

doi:10.1007/s002210000452




36

Ghajar, J., & Ivry, R. B. (2008). The predictive brain state: Timing deficiency in traumatic brain

injury? Neurorehabilitation and Neural Repair, 22(3), 217-227.

doi:10.1177/1545968308315600

Guillamondegui, O. D., Montgomery, S. A., Phibbs, F. T., McPheeters, M. L., Alexander, P. T.,

Jerome, R. N., . . . KE., H. (2011). Traumatic Brain Injury and Depression.

Hall, L. M., Brauer, S., Horak, F., & Hodges, P. W. (2010). Adaptive changes in anticipatory

postural adjustments with novel and familiar postural supports. Journal of

Neurophysiology, 103(2), 968-976. doi:10.1152/jn.00479.2009

Hart, T., Fann, J. R., Chervoneva, I., Juengst, S. B., Rosenthal, J. A., Krellman, J. W., . . . Kroenke,

K. (2016). Prevalence, Risk Factors, and Correlates of Anxiety at 1 Year After Moderate

to Severe Traumatic Brain Injury. Archives of Physical Medicine and Rehabilitation, 97(5),

701-707. doi:10.1016/j.apmr.2015.08.436

Horak, F. B. (2006). Postural orientation and equilibrium: what do we need to know about

neural control of balance to prevent falls? Age and Ageing, 35 Suppl 2, ii7-ii11.

doi:10.1093/ageing/afl077

Horak, F. B., Diener, H. C., & Nashner, L. M. (1989). Influences of Central Set on Human Postural

Responses. Journal of Neurophysiology, 62(4), 841-853.

Horak, F. B., Esselman, P., Anderson, M. E., & Lynch, M. K. (1984). The effects of movement

velocity, mass displaced, and task certainty on associated postural adjustments made by

normal and hemiplegic individuals. Journal of Neurology, Neurosurgery, and Psychiatry,

47(9), 1020-1028.

Horak, F. B., & Macpherson, J. M. (2011). Postural orientation and equlibrium Comprehensive

Psychology (pp. 255-292).

Horak, F. B., & Nashner, L. M. (1996). Central programming of postural movements: Adaptation

to altered support-surface configurations. Journal of Neurophysiology, 55(6), 1369-1381.

Iaccarino, M. A., Bhatnagar, S., & Zafonte, R. (2015). Rehabilitation after traumatic brain injury.

Handbook of Clinical Neurology, 127, 411-422. doi:10.1016/B978-0-444-52892-6.00026-

X

Inness, E. L., Mansfield, A., Bayley, M., & McIlroy, W. E. (2016). Reactive Stepping After Stroke:

Determinants of Time to Foot Off in the Paretic and Nonparetic Limb. Journal of

Neurologic Physical Therapy, 40(3), 196-202. doi:10.1097/NPT.0000000000000132

Inness, E. L., Mansfield, A., Lakhani, B., & McIlroy, W. E. (2014). Impaired Reactive Stepping

Among Patients Ready for Discharge From Inpatient Stroke Rehabilitation. Physical

Therapy, 94(12), 1755-1764.

37

Ito, T., Azuma, T., & Yamashita, N. (2003). Anticipatory control in the initiation of a single step

under biomechanical constraints in humans. Neuroscience Letters, 352(3), 207-210.

doi:10.1016/S0304-3940(03)01081-4

Jacobs, J. V., & Horak, F. B. (2007). Cortical control of postural responses. Journal of Neural

Transmission, 114(10), 1339-1348. doi:10.1007/s00702-007-0657-0

Jensen, J. L., Brown, L. A., & Woollacott, M. H. (2001). Compensatory stepping: the

biomechanics of a preferred response among older adults. Experimental Aging

Research, 27(4), 361-376. doi:10.1080/03610730109342354

Jian, Y., Winter, D. A., Ishac, M. G., & Gilchrist, L. (1993). Trajectory of the body COG and COP

during initiation and termination of gait. Gait and Posture, 1(1), 9-22.

Jorge, R. E., & Arciniegas, D. B. (2014). Mood disorders after TBI. Psychiatric Clinics of North

America, 37(1), 13-29. doi:10.1016/j.psc.2013.11.005

Kanekar, N., Lee, Y. J., & Aruin, A. S. (2014). Frequency analysis approach to study balance

control in individuals with multiple sclerosis. Journal of Neuroscience Methods, 222, 91-

96. doi:10.1016/j.jneumeth.2013.10.020

Khan, F., Amatya, B., Judson, R., Chung, P., Truesdale, M., Elmalik, A., & Galea, M. P. (2016).

Factors associated with long-term functional and psychological outcomes in persons

with moderate to severe traumatic brain injury. Journal of Rehabilitation Medicine,

48(5), 442-448. doi:10.2340/16501977-2084

Khan, F., Baguley, I. J., & Cameron, I. J. (2003). Rehabilitation after traumatic brain injury.

Medical Journal of Australia, 178(6), 290-295.

Kiemel, T., Oie, K. S., & Jeka, J. J. (2006). Slow dynamics of postural sway are in the feedback

loop. Journal of Neurophysiology, 95(3), 1410-1418. doi:10.1152/jn.01144.2004

Kirchner, M., Schubert, P., Getrost, T., & Haas, C. T. (2013). Effect of altered surfaces on

postural sway characteristics in elderly subjects. Human Movement Science, 32(6), 1467-

1479. doi:10.1016/j.humov.2013.05.005

Klous, M., Mikulic, P., & Latash, M. L. (2012). Early postural adjustments in preparation to

whole-body voluntary sway. Journal of Electromyography and Kinesiology, 22(1), 110-

116. doi:10.1016/j.jelekin.2011.11.005

Krishnan, V., Latash, M. L., & Aruin, A. S. (2012). Early and late components of feed-forward

postural adjustments to predictable perturbations. Clinical Neurophysiology, 123(5),

1016-1026. doi:10.1016/j.clinph.2011.09.014

Latash, M. L., Ferreira, S. S., Wieczorek, S. A., & Duarte, M. (2003). Movement sway: changes in

postural sway during voluntary shifts of the center of pressure. Experimental Brain

Research, 150(3), 314-324. doi:10.1007/s00221-003-1419-3

38

Laxe, S., Zasler, N., Robles, V., Lopez-Blazquez, R., Tormos, J. M., & Bernabeu, M. (2014). ICF

profiling of patients with traumatic brain injury: an international professional survey.

Disability and Rehabilitation, 36(1), 82-88. doi:10.3109/09638288.2013.780641




962.

Lewis, F. D., & Horn, G. J. (2017). Depression following traumatic brain injury: Impact on post-

hospital residential rehabilitation outcomes. NeuroRehabilitation, 40(3), 401-410.

Lin, S. I., & Woollacott, M. (2005). Association between sensorimotor function and functional

and reactive balance control in the elderly. Age and Ageing, 34(4), 358-363.

doi:10.1093/ageing/afi089

Little, C. E., & Woollacott, M. (2014). Effect of attentional interference on balance recovery in

older adults. Experimental Brain Research, 232(7), 2049-2060. doi:10.1007/s00221-014-

3894-0

MacKinnon, C. D., Bissig, D., Chiusano, J., Miller, E., Rudnick, L., Jager, C., . . . Rogers, M. W.

(2007). Preparation of anticipatory postural adjustments prior to stepping. Journal of

Neurophysiology, 97(6), 4368-4379. doi:10.1152/jn.01136.2006

Madigan, N. K., DeLuca, J., Diamond, B. J., Tramontano, G., & Averill, A. (2000). Speed of

information processing in traumatic brain injury: Modality-specific factors. Journal of

Head Trauma Rehabilitation, 15(3), 943-956.

Maki, B. E., & McIlroy, W. E. (1997). The role of limb movements in maintaining upright stance:

The "change-in-support" strategy. Physical Therapy, 77(5), 488-507.

Maki, B. E., & McIlroy, W. E. (2007). Cognitive demands and cortical control of human balance-

recovery reactions. Journal of Neural Transmission, 114(10), 1279-1296.

doi:10.1007/s00702-007-0764-y

Mansfield, A., Danells, C. J., Inness, E. L., Mochizuki, G., & McIlroy, W. E. (2011). Between-limb

synchronization for control of standing balance in individuals with stroke. Clinical

Biomechanics, 26(3), 312-317. doi:10.1016/j.clinbiomech.2010.10.001

Mansfield, A., Inness, E. L., Lakhani, B., & McIlroy, W. E. (2012). Determinants of limb

preference for initiating compensatory stepping poststroke. Archives of Physical

Medicine and Rehabilitation, 93(7), 1179-1184. doi:10.1016/j.apmr.2012.02.006

Mansfield, A., Mochizuki, G., Inness, E. L., & McIlroy, W. E. (2012). Clinical correlates of

between-limb synchronization of standing balance control and falls during inpatient

stroke rehabilitation. Neurorehabilitation and Neural Repair, 26(6), 627-635.

doi:10.1177/1545968311429688

39

Massion, J. (1992). Movement, posture and equilibrium: Interaction and coordination. Progress

in Neurobiology, 38(1), 35-56.

Massion, J. (1994). Postural control system Current Opinion in Neurobiology, 4(6), 877-887.

McGinn, M. J., & Povlishock, J. T. (2015). Cellular and molecular mechanisms of injury and

spontaneous recovery. Handbook of Clinical Neurology, 127, 67-87. doi:10.1016/B978-0-

444-52892-6.00005-2

McIlroy, W. E., & Maki, B. E. (1993). Do anticipatory postural adjustments precede

compensatory stepping reactions evoked by perturbations? Neuroscience Letters, 164(1-

2), 199-202.

McIlroy, W. E., & Maki, B. E. (1999). The control of lateral stability during rapid stepping

reactions evoked by antero-posterior perturbation: does anticipatory control play a

role? Gait and Posture, 9(3), 190-198.

McKechnie, D., Pryor, J., & Fisher, M. J. (2015). Falls and fallers in traumatic brain injury (TBI)

rehabilitation settings: an integrative review. Disability and Rehabilitation, 37(24), 2291-

2299. doi:10.3109/09638288.2014.1002578

McKechnie, D., Pryor, J., & Fisher, M. J. (2017). An examination of patient characteristics that

contribute to falls in the inpatient traumatic brain injury rehabilitation setting. Disability

and Rehabilitation, 39(18), 1864-1871. doi:10.1080/09638288.2016.1212112

Merkley, T. L., Larson, M. J., Bigler, E. D., Good, D. A., & Perlstein, W. M. (2013). Structural and

functional changes of the cingulate gyrus following traumatic brain injury: relation to

attention and executive skills. Journal of the International Neuropsychology Society,

19(8), 899-910. doi:10.1017/S135561771300074X

Mochizuki, G., Boe, S., Marlin, A., & McIlRoy, W. E. (2010). Perturbation-evoked cortical activity

reflects both the context and consequence of postural instability. Neuroscience, 170(2),

599-609. doi:10.1016/j.neuroscience.2010.07.008

Mochizuki, G., Ivanova, T. D., & Garland, S. J. (2004). Postural muscle activity during bilateral

and unilateral arm movements at different speeds. Experimental Brain Research, 155(3),

352-361.

Mochizuki, G., Ivanova, T. D., & Garland, S. J. (2005). Synchronization of motor units in human

soleus muscle during standing postural tasks. Journal of Neurophysiology, 94(1), 62-69.

doi:10.1152/jn.01322.2004

Muir-Hunter, S. W., Clark, J., McLean, S., Pedlow, S., Van Hemmen, A., Montero Odasso, M., &

Overend, T. (2014). Identifying balance and fall risk in community-dwelling older

women: the effect of executive function on postural control. Physiotherapy Canada,

66(2), 179-186. doi:10.3138/ptc.2013-16

40

Myklebust, J. B., Lovett, E. G., Myklebust, B. M., Reynolds, N., Milkowski, L., & Prieto, T. E.

(2009). Two-dimensional coherence for measurement of asymmetry in postural

steadiness. Gait and Posture, 29(1), 1-5. doi:10.1016/j.gaitpost.2008.05.011

Nelson-Wong, E., Howarth, S., Winter, D. A., & Callaghan, J. P. (2009). Application of

autocorrelation and cross-correlation analyses in human movement and rehabilitation

research. Journal of Orthopaedic and Sports Physical Therapy, 39(4), 287-295.

doi:10.2519/jospt.2009.2969



Ochi, F., Esquenazi, A., Hirai, B., & Talaty, M. (1999). Temporal-spatial feature of gait after

traumatic brain injury. Journal of Head Trauma Rehabilitation, 14(2), 105-115.

Osborn, A. J., Mathias, J. L., Fairweather-Schmidt, A. K., & Anstey, K. J. (2016). Anxiety and

comorbid depression following traumatic brain injury in a community-based sample of

young, middle-aged and older adults. Journal of Affective

Disorders(ttp://dx.doi.org/10.1016/j.jad.2016.09.045).

Owens, J. A., Spitz, G., Ponsford, J. L., Dymowski, A. R., Ferris, N., & Willmott, C. (2017). White

matter integrity of the medial forebrain bundle and attention and working memory

deficits following traumatic brain injury. Brain and Behavior, 7(2), e00608.

doi:10.1002/brb3.608

Paillex, R., & So, A. (2003). Posture debout chez des sujets adultes : spécificités de l'hémiplégie.

Annales de Réadaptation et de Médecine Physique, 46(2), 71-78. doi:10.1016/s0168-

6054(03)00002-3

Peterka, R. J., & Loughlin, P. J. (2004). Dynamic Regulation of Sensorimotor Integration in

Human Postural Control. Journal of Neurophysiology, 91(1), 410-423.




Ponsford, J. L., Downing, M. G., Olver, J., Ponsford, M., Acher, R., Carty, M., & Spitz, G. (2014).

Longitudinal follow-up of patients with traumatic brain injury: outcome at two, five, and

ten years post-injury. Journal of Neurotrauma, 31(1), 64-77. doi:10.1089/neu.2013.2997

Povlishock, J. T., & Katz, D. I. (2005). Update of Neuropathology and Neurological Recovery

After Traumatic Brain Injury. Journal of Head Trauma and Rehabilitation, 20(1), 76-94.

Prieto, T. E., Myklebust, J. B., Hoffmann, R. G., Lovett, E. G., & Myklebust, B. M. (1996).

Measures of postural steadiness: differences between healthy young and elderly adults.

IEEE Transactions on Biomedical Engineering, 43(9), 956-966.

41

Prochazka, A. (1989). Sensorimotor gain control: A basic strategy of motor systems? Progress in

Neurobiology, 33(4), 281-307. doi:http://dx.doi.org/10.1016/0301-0082(89)90004-X

Public Health Agency of Canada. (2014). Mapping Connections: An understanding of

neurological conditions in Canada. Minister of Health.

Rankin, J. K., Woollacott, M. J., Shumway-Cook, A., & Brown, L. A. (2000). Cognitive influence on

postural stability: A neuromuscular analysis in young and older adults. Journal of

Gerontology, 55A(3), M112-M119.

Remaud, A., Boyas, S., Caron, G. A., & Bilodeau, M. (2012). Attentional demands associated

with postural control depend on task difficulty and visual condition. Journal of Motor

Behavior, 44(5), 329-340. doi:10.1080/00222895.2012.708680

Richard, A., Van Hamme, A., Drevelle, X., Golmard, J. L., Meunier, S., & Welter, M. L. (2017).

Contribution of the supplementary motor area and the cerebellum to the anticipatory

postural adjustments and execution phases of human gait initiation. Neuroscience.


Rigoldi, C., Galli, M., Mainardi, L., Crivellini, M., & Albertini, G. (2011). Postural control in

children, teenagers and adults with Down syndrome. Research in Developmental

Disabilities, 32(1), 170-175. doi:10.1016/j.ridd.2010.09.007

Schinkel-Ivy, A., Singer, J. C., Inness, E. L., & Mansfield, A. (2016). Do quiet standing centre of

pressure measures within specific frequencies differ based on ability to recover balance

in individuals with stroke? Clinical Neurophysiology, 127(6), 2463-2471.

doi:10.1016/j.clinph.2016.02.021

Shah, S. A., Goldin, Y., Conte, M. M., Goldfine, A. M., Mohamadpour, M., Fidali, B. C., . . . Schiff,

N. D. (2017). Executive attention deficits after traumatic brain injury reflect impaired

recruitment of resources. Neuroimage: Clinical, 14, 233-241.

doi:10.1016/j.nicl.2017.01.010

Sharp, D. J., Scott, G., & Leech, R. (2014). Network dysfunction after traumatic brain injury.

Nature Reviews: Neurology, 10(3), 156-166. doi:10.1038/nrneurol.2014.15

Shaw, N. A. (2002). The neurophysiology of concussion. Progress in Neurobiology, 281-344.

Shumskaya, E., van Gerven, M. A., Norris, D. G., Vos, P. E., & Kessels, R. P. (2017). Abnormal

connectivity in the sensorimotor network predicts attention deficits in traumatic brain

injury. Experimental Brain Research, 235(3), 799-807. doi:10.1007/s00221-016-4841-z

Shumway-Cook, A., & Woollacott, M. H. (2012). Motor Control: Translating Research into

Clinical Practice: Wolters Kluwer Health, Lippincott Williams & Wilkins.

Shumway-Cook, A., & Woollacott, M. J. (2000). Attentional Demands and Postural Control: The

effects of sensory context. Journal of Gerontology, 55A(1), M10-M16.

42

Singer, J. C., Mansfield, A., Danells, C. J., McIlroy, W. E., & Mochizuki, G. (2013). The effect of

post-stroke lower-limb spasticity on the control of standing balance: Inter-limb spatial

and temporal synchronisation of centres of pressure. Clinical Biomechanics, 28(8), 921-

926. doi:10.1016/j.clinbiomech.2013.07.010

Singer, J. C., & Mochizuki, G. (2015). Post-Stroke Lower Limb Spasticity Alters the Interlimb

Temporal Syncrhonization of Centre of Pressure Displacements Across Multiple

Timescales. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 23(5),

786-795.

Siu, K. C., & Woollacott, M. H. (2007). Attentional demands of postural control: the ability to

selectively allocate information-processing resources. Gait and Posture, 25(1), 121-126.

doi:10.1016/j.gaitpost.2006.02.002

Slovarp, L., Azuma, T., & Lapointe, L. (2012). The effect of traumatic brain injury on sustained

attention and working memory. Brain Injury, 26(1), 48-57.

doi:10.3109/02699052.2011.635355

Stelmach, G. E., & Worringham, C. J. (1985). Sensorimotor deficits related to postural stability.

Implications for falling in the elderly. Clinics in Geriatric Medicine, 1(3), 679-694.

Sturnieks, D. L., Delbaere, K., Brodie, M. A., & Lord, S. R. (2016). The influence of age, anxiety

and concern about falling on postural sway when standing at an elevated level. Human

Movement Science, 49, 206-215. doi:10.1016/j.humov.2016.06.014

Stuss, D. T. (2011). Traumatic brain injury: relation to executive dysfunction and the frontal

lobes. Current Opinion in Neurology, 24(6), 584-589.

doi:10.1097/WCO.0b013e32834c7eb9

Suarez, H., Sotta, G., San Roman, C., Arocena, S., Ferreira, E., Geisinger, D., . . . Picerno, J.

(2013). Postural response characterization in elderly patients with bilateral vestibular

hypofunction. Acta Oto-Laryngologica, 133(4), 361-367.

doi:10.3109/00016489.2012.739731

Uemura, K., Oya, T., & Uchiyama, Y. (2013). Effects of speed and accuracy strategy on choice

step execution in response to the flanker interference task. Human Movement Science,

32(6), 1393-1403. doi:10.1016/j.humov.2013.07.007

Vakil, E. (2005). The effect of moderate to severe traumatic brain injury (TBI) on different

aspects of memory: a selective review. Journal of Clinical and Experimental

Neuropsychology, 27(8), 977-1021. doi:10.1080/13803390490919245

van Iersel, M. B., Kessels, R. P. C., Bloem, B. R., Verbeek, A. L. M., & Olde Rikkert, M. G. M.

(2008). Executive Functions Are Associated With Gait and Balance in Community-Living

Elderly People. The Journals of Gerontology: Series A, 63(12), 1344-1349.

doi:10.1093/gerona/63.12.1344

43

Varghese, J. P., Beyer, K. B., Williams, L., Miyasike-daSilva, V., & McIlroy, W. E. (2015). Standing

still: is there a role for the cortex? Neuroscience Letters, 590, 18-23.

doi:10.1016/j.neulet.2015.01.055

Vasudevan, E. V., Glass, R. N., & Packel, A. T. (2014). Effects of traumatic brain injury on

locomotor adaptation. Journal of Neurologic Physical Therapy, 38(3), 172-182.

doi:10.1097/NPT.0000000000000049

Vuillerme, N., & Vincent, H. (2006). How performing a mental arithmetic task modify the

regulation of centre of foot pressure displacements during bipedal quiet standing.

Experimental Brain Research, 169(1), 130-134. doi:10.1007/s00221-005-0124-9

Wade, L. D., Canning, C. G., Fowler, V., Felmingham, K. L., & Baguley, I. J. (1997). Changes in

postural sway and performance of functional tasks during rehabiliation after truamatic

brain injury Archives of Physical Medicine and Rehabilitation, 78(10), 1107-1111.

Walker, W. C., & Pickett, T. C. (2007). Motor impairment after severe traumatic brain injury: A

longitudinal multicenter study. Journal of Rehabilitation Research and Development,

44(7), 975-982. doi:10.1682/jrrd.2006.12.0158

Williams, G., Morris, M. E., Schache, A., & McCrory, P. R. (2009). Incidence of gait abnormalities

after traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 90(4), 587-

593. doi:10.1016/j.apmr.2008.10.013

Willmott, C., Ponsford, J., Hocking, C., & Schonberger, M. (2009). Factors contributing to

attentional impairments after traumatic brain injury. Neuropsychology, 23(4), 424-432.

Winter, D. A. (1995). Human balance and posture control standing and walking. Gait and

Posture, 3(4), 193-214.

Winter, D. A., & Patla, A. E. (1997). Signal processing and linear systems for the movement

sciences. Waterloo, ON: Waterloo Biomechanics.

Winter, D. A., Patla, A. E., Prince, F., Ishac, M., & Gielo-Perczak, K. (1998). Stiffness Control of

Balance in Quiet Standing. Journal of Neurophysiology, 80(3), 1211-1221.

Winter, D. A., Prince, F., Stergiou, P., & Powell, C. (1993). Medial-lateral and anterior-posterior

motor responses associated with centre of pressure changes in quiet standing.

Neuroscience Research Communications, 12(3), 141-148.




Wolpert, D. M., & Kawato, M. (1998). Multiple paired forward and inverse models for motor

control. Neural Networks, 11(7-8), 1317-1329.

44



Woollacott, M. H., & Shumway-Cook, A. (2002). Attention and the control of posture and gait: A

review of an emerging area of research. Gait and Posture, 16(1), 1-14.

Yanohara, R., Teranish, T., Yutaka, T., Taniso, G., Ueno, R., & Sonada, S. (2014). Recovery

process of standing postural control in hemiplegia after stroke. Journal of Physical

Therapy Science, 26(11), 1761-1765.

Zatsiorsky, V. M., & Duarte, M. (1999). Instant equilibrium point and its migration in standing

tasks: Rambling and trembling components of the stabilogram. Motor Control, 3(1), 28-

38.

Zimmermann, N., Pereira, N., Hermes-Pereira, A., Holz, M., Joanette, Y., & Fonseca, R. P. (2015).

Executive functions profiles in traumatic brain injury adults: Implications for

rehabilitation studies. Brain Injury, 29(9), 1071-1081.

doi:10.3109/02699052.2015.1015613

45

Chapter 3 : The effects of predictability on inter-limb postural

synchronization prior to bouts of postural instability

Habib Perez, O., Singer, J. & Mochizuki, G. (2016). The effects of predictability on inter-limb

postural synchronization prior to bouts of postural instability. Gait and Posture, 46, 167-172.

46

3.1 Abstract

Anticipatory balance control optimizes balance reactions to postural perturbations. Predictive

control is dependent on the ability of the central nervous system to modulate gain in

accordance with specific task demands. Inter-limb synchronization is a sensitive measure of

individual limb contributions to balance control and may reflect the coordination of gain

modulation in preparation for instability. The purpose of the study was to determine whether

gain modulation in advance of predictable bouts of instability was reflected in the extent of

inter-limb synchronization. Two adjacent force plates were used to collect centre of pressure

(COP) data from 12 healthy young adults (27.5 ±3.4 years). Participants prepared for internal

and external balance perturbations using a cueing paradigm with three auditory warning tones

followed by an imperative tone. Perturbations were delivered in blocked and randomized

conditions with two perturbation magnitudes (small and large). Inter-limb synchrony was

calculated using the cross-correlation function of the COP excursions from the left and right

foot for three seconds prior to perturbation onset in the anteroposterior (AP) and mediolateral

(ML) direction. Inter-limb synchrony decreased in the AP and ML directions as perturbation

magnitude became more unpredictable. The need to take a step or not knowing whether a step

was required prior to postural instability reduced ML inter-limb synchrony. No differences were

found between internal and external perturbations. Modulation of postural set was evident in

the extent of inter-limb synchrony.

47

3.2 Introduction

Postural control is not an automatic process, but instead requires attentional resources

to optimize control (Kerr, Condon, & McDonald, 1985; Rankin, Woollacott, Shumway-Cook, &

Brown, 2000; Remaud, Boyas, Caron, & Bilodeau, 2012; Shumway-Cook & Woollacott, 2000; Siu

& Woollacott, 2007). Reliance on these resources for generating contextually appropriate

postural responses implies that top-down processes likely regulate balance control and are

engaged when preparing for balance responses in the event of postural instability. Anticipatory

balance control optimizes reactions to perturbations induced by self-generated movements or

for externally-generated perturbations whose characteristics are known.

Anticipatory control is dependent on the extent of central nervous system (CNS)

readiness, also known as ‘central set’ (Prochazka, 1989) and is influenced by prior experience

and current context (Horak, Diener, & Nashner, 1989). Prior experience with predictable stimuli

enables descending commands to prepare and modify postural responses (Horak et al., 1989;

Prochazka, 1989). By anticipating perturbations, postural muscle activity can be initiated prior

to perturbation onset and mitigate subsequent instability (Cordo & Nashner, 1982; Horak,

Esselman, Anderson, & Lynch, 1984). The role of set modulation is supported by studies

characterizing the cortical potentials associated with the factors of central set (i.e. the

predictability and amplitude) of the upcoming balance perturbations (Mochizuki, Boe, Marlin, &

McIlRoy, 2010; Mochizuki, Sibley, Cheung, & McIlroy, 2009; Mochizuki, Sibley, Esposito,

Camilleri, & McIlroy, 2008).

One important feature of anticipatory control is the extent to which gain modulation

affects postural control strategies. Previous work has demonstrated centre of pressure (COP)

48

shifts (i.e. early postural adjustments; EPA) prior to self-initiated forward postural sway (Klous,

Mikulic, & Latash, 2012) that were distinct from anticipatory postural adjustments. EPAs

optimize postural control in preparation for volitional or reactive instability (Klous et al., 2012).

Additionally, bilateral activation of postural muscles with comparable magnitude and onset in

advance of self-initiated bilateral shoulder-flexion perturbations demonstrated a more

unilateral activation pattern for unilateral perturbations (Mochizuki, Ivanova, & Garland, 2004).

These examples illustrate the capacity of the CNS to modulate gain in accordance with task

demands by coupling or decoupling between-limb coordination as necessary given current

contextual requirements or previous experience (i.e. depending on whether the conditions

under which postural instability is experienced are known or familiar) (Jacobs & Horak, 2007).

The extent of inter-limb synchronization of COP displacements reflects the coordination of

muscle activation patterns between limbs (Winter, Prince, Stergiou, & Powell, 1993) and is a

sensitive measure of balance control (Mansfield, Danells, Inness, Mochizuki, & McIlroy, 2011;

Mochizuki, Ivanova, & Garland, 2005; Singer, Mansfield, Danells, McIlroy, & Mochizuki, 2013).

Thus, measuring the extent of between-limb coordination in advance of postural instability may

enhance understanding of the processes by which humans prepare for and optimize balance-

correcting responses.

The purpose of this study was to determine whether changes in inter-limb postural

synchronization prior to bouts of postural instability are indicative of modulation in postural

set. Consistent with studies identifying modifiability in pre-perturbation cortical activity related

to postural set alteration (Mochizuki et al., 2010) and because high inter-limb synchrony may

contribute to optimal balance control in quiet standing (Mansfield et al., 2011; Mochizuki et al.,

49

2005; Singer et al., 2013; Winter et al., 1993), we hypothesized that inter-limb synchrony would

be highest when perturbation magnitudes were predictable and of low magnitude. Conversely,

when the parameters of instability were unpredictable and required the execution of

compensatory stepping responses, we expected a reduction in inter-limb synchronization (i.e.

decoupling of between-limb synchrony). As volitional movements have an additional level of

saliency/predictability due to internal models for the intended movement, we predicted that

individuals would demonstrate better control of movement through higher levels of inter-limb

synchrony when perturbations were internally generated. Lastly, we expected that inter-limb

synchrony would change from the start to the end of repeated exposure to perturbations when

perturbation magnitudes were known, but would not differ when magnitudes were

unpredictable. Preliminary data has been presented in abstract form (Habib Perez & Mochizuki,

2014).

3.3 Methods

3.3.1 Participants

Fifteen healthy adults volunteered for the study. Three participants were excluded –

two reported to be in the first trimester of their pregnancy, while another participant did not

meet our inclusion criteria of normal/corrected visual acuity during testing. Data were collected

from twelve participants (seven males, 27.5±3.4 years, 172.7±9.6cm, 68.8±11.6kg) who

provided written, informed consent prior to the start of study. Participants completed the

Waterloo Footedness Questionnaire (Elais, Bryden, & Bulman-Fleming, 1998) to determine

lower extremity dominance. Dominant limb was defined as the limb used most frequently for

50

manipulation of object and providing support during an activity (Lakhani, Mansfield, Inness, &

McIlroy, 2011). The study was approved by the Research Ethics Board at Sunnybrook Research

Institute.

3.3.2 Instrumentation and Protocol

Two force plates (Bertec Corp., Columbus, OH, USA) were arranged with the Y-axes in

parallel and separated by 5mm. Ground reaction forces and moments from each plate were

sampled at 500Hz using a 16-bit acquisition system (Power 1401 mk2, Spike2 v7.02 software,

Cambridge Electronic Design, Cambridge, UK). External perturbations were applied using a

5.5cm aluminum rod (tipped with a dense rubber cap) attached to a load cell (MLP-300,

Transducer Techniques, Temecula, CA, USA). Load cell signals were sampled at 1000Hz and

down-sampled to align to the collection of the force channels for analysis (LVC-U5, A-Tech

Instruments Ltd., Toronto, Canada).

Each foot was positioned on a separate force plate. Foot placement was standardized

across individuals (medial aspects of each foot 0.185m apart) at the start of each trial. For

baseline measures, one 30-second quiet standing (QS) trial was collected. Participants then

were presented with a series of three auditory warning cues (1-second inter-cue interval) while

maintaining a fixed gaze on a target approximately 3m ahead of them. On the presentation of a

fourth auditory tone (a different tone for each condition), participants experienced a postural

perturbation.

51

Participants completed two sessions at least one week apart, each consisting of internal

or external perturbations. Each session consisted of three conditions: a block of small (BS), a

block of large (BL) and a block of randomly presented small (RS) or large (RL) perturbations

(Table 3-1).

Table 3-1: Summary of perturbation types and their corresponding responses for each

condition in the experimental paradigm.

Perturbation Type External Internal

Conditions Response

Blocked Small BS Feet-in-place No Step

Blocked Large BL Change-in-support Step

Randomized Small RS Feet-in-place No Step

Randomized Large RL Change-in-support Step

Thirty trials of each condition were presented as a block in the S and L conditions and a set of

60 trials was presented in random order for the random condition for a total of 120 trials. Inter-

trial duration was 15 seconds and participants were given a 5-minute break in between each

condition. Internal perturbations were performed first and the order of BS and BL conditions

was randomized among participants. The random condition was always presented last to allow

familiarization with the different auditory tones associated with either the S or L conditions for

both Internal and External perturbations.

52

Internal perturbations – For the L conditions, participants were instructed to take a volitional

step in the anterior direction with their dominant foot as quickly as possible immediately after

hearing the imperative tone. For the S conditions, participants were instructed to not take a

step immediately after the imperative tone (i.e. prepare not to step).

External perturbations – External perturbations were delivered manually using an aluminum

rod, similar to previous work (Adkin, Campbell, Chua, & Carpenter, 2008). Perturbations were

applied directly over the spinous processes (approximate location T5) to reduce any torques

about the trunk. A rectangular foam pad (15.5cm x 20.5cm x 1.5cm) was placed between the

two scapulae to reduce the impact of the rod on the spinous processes. Small perturbations

associated with the S condition resulted in feet-in-place reactions, while large perturbations

associated with the L conditions resulted in stepping responses (see Figure 3-1 and Table 3-2).

Force thresholds for each participant were determined by exposing the participant to various

levels of force magnitudes in random order (i.e. calibration trials). This allowed the

experimenter to determine the force threshold for small and large perturbations, which evoked

a feet-in-place and stepping reactions, respectively. The same experimenter performed all

perturbations.

Prior to each set of blocked trials, participants were informed about the number and

magnitude of the perturbation. For the random trials, participants were informed that they

would experience either a small or large perturbation. In all conditions, participants were

instructed to “use whatever strategy you need to regain your balance”.

53

Table 3-2: Mean and standard deviation for external perturbation characteristics. Load cell

onset (LConset), load cell peak force (LCpeak), the time at which peak occurred (LCpeak(τ)), and

the percentage of the load cell peak force applied at the onset of postural activity (%LC-

COPonset) is listed.

Conditions BS BL RS RL

LConset (ms) 121 ±155 195 ±177 162 ±173 176 ±192

LCpeak (N) 87.9 ±27.8 215.0 ±61.5 108.9 ±29.6 225.4 ±57.3

LCpeak(τ) (ms) 484 ±144 598 ±165 507 ±156 570 ±165

%LC-COPonset (%) 58.0 ±18.2 47.2 ±20.3 39.9 ±15.0 39.5 ±10.2

54

Figure 3-1: Mean force profile across trials from one participant for external perturbations in

small (S, dash) and large (L, solid) conditions (top). Experimental set up for external

perturbations (bottom).

55

3.3.3 Data Analysis

Centre of Pressure – Raw data were low-pass filtered using a zero-lag, fourth-order,

Butterworth filter with a cut-off frequency of 10 Hz. Anteroposteior (AP) and mediolateral (ML)

COP under each limb were calculated. The cross-correlation function was calculated using the

left and right mean-removed AP and ML COP waveforms. Cross-correlation coefficients at zero

phase-lag (Rxy(0)) were calculated at incremental phase-shifts defined by the sampling rate, by

iteratively shifting the right limb COP forwards and backwards in time over the entire length of

the record (Singer et al., 2013), over the 3 second duration prior to perturbation (i.e. during the

3 seconds of preparation for internal or external instability). Similarly, for baseline measures of

standing balance, Rxy(0) were also calculated for quiet standing over a 30 second duration. To

determine the extent of set modulation within a condition, the AP Rxy(0) for the first five trials

(First5) and the last five trials (Last5) were analyzed. These variables were evaluated both in

internal and external perturbation conditions.

3.3.4 Secondary Analyses

For External Perturbations (Table 3-2), load cell magnitude and timing characteristics

were analyzed to confirm consistency in perturbation within each condition (i.e. S or L). Load

cell onset (LConset) was defined as the time with respect to the imperative tone at which the

load cell increased more than 0.6 N within 0.014 s, a rate of 40 N/s. Load cell peak force

(LCpeak) was taken as the maximum value and the time point which this peak occurred was

defined as LCpeak(τ). The percentage of load cell peak applied at the onset of postural activity

(%LC-COPonset) was also measured. Onset of postural activity was defined as the time point

56

where the net AP COP position was 8mm greater than the mean AP COP during the 200ms just

prior to the imperative tone (McIlroy & Maki, 1999).

3.3.5 Statistical Analysis

Data from all trials from each participant were evaluated for normality. A 4 × 2 within-

subject analysis of variance (ANOVA) was conducted to analyze the effects of Condition (BS, BL,

RS, RL) and Perturbation Type (Internal, External) on AP and ML COP measures. Separate,

paired t-tests were used to compare Rxy(0) in the AP direction in the First5 and Last5 trials for

each condition to investigate the role of set modulation within a condition. Additionally, a

paired t-test was conducted between QS and the S condition in Internal Perturbations. We

hypothesized that no differences would be found between QS and S.

To identify whether S and L perturbations were significantly different in magnitude and

temporally consistent in external perturbations, separate one-way repeated measures ANOVA

were conducted on the following load cell outcome measures: LConset, LCpeak, LCpeak(τ), %LC-

COPonset. For all ANOVAs, Greenhouse-Geisser corrected degrees of freedoms were used if the

assumption of sphericity was not met.

57

3.4 Results

3.4.1 Inter-limb Synchrony in Preparation for Instability

A significant main effect of Condition was found for Rxy(0) in the AP direction

(F2.1,23.2=5.02, p=0.014, partial η2 =.31). Bonferroni adjusted post-hoc analysis identified a

tendency towards significant differences between BS (Rxy(0)=0.83) and BL (Rxy(0)=0.78; p=0.057)

and RS (Rxy(0)=0.77; p=0.09). For ML synchrony, a significant main effect of Condition was found

for Rxy(0) (F3,33=9.32, p<0.0005, partial η2 =.46). Bonferroni adjusted post-hoc analysis revealed

that BS (Rxy(0)=-0.26) was significantly more synchronized than all other conditions (Rxy(0) range

-0.05 to -0.07; p≤0.04; Figure 3-2). The results demonstrate that inter-limb synchrony

decreased as perturbation magnitude became more unpredictable. No significant main effects

of Perturbation Type were found.

58

Figure 3-2: Anteroposterior (AP) cross-correlation (Rxy(0)) group means and standard error (SE)

for quiet standing (QS), and all conditions in internal and external perturbations for (A); AP

Rxy(0) individual data for external (B) and internal (C) perturbations. Mediolateral (ML) Rxy(0)

group means and SE for QS, and all conditions in internal and external perturbations for (D); ML

Rxy(0) individual data for external (E) and internal (E) perturbations. Conditions include blocked

(B) and random (R) small (S) and large (L) perturbations (BS, BL, RS, RL). * denotes a significant

difference in condition (p≤0.05)

Paired t-test demonstrated that inter-limb synchrony (Rxy(0)) for conditions in External

perturbations was not significantly different between the First5 and Last5 trials in BS, BL and RS

conditions (p>0.05); however, RL did demonstrate a significant reduction from the First5 to

59

Last5 trials (p=0.01; Figure 3-3A). Internal perturbations conditions showed a similar pattern as

in External perturbations, but no significance was found between the First5 and Last5 trials in

each condition (Figure 3-3B).

Figure 3-3: Means and standard error bars from first 5 (First5) and last 5 (Last5) trials in

external (A) and internal (B) perturbations for all participants (n=12). * denotes significant

paired t-test differences at p < 0.05.

60

3.4.2 Quiet Standing

Participants demonstrated high AP inter-limb synchrony at zero phase lag (Rxy(0)=0.86

±0.03) and did not differ to BS (Internal). ML inter-limb synchrony at zero phase lag in QS

(Rxy(0)=-0.52 ±0.10) was significantly more synchronized than BS (Internal) (Rxy(0)=-0.23 ±0.12;

t(11)= -5.83 p<0.05.

3.4.3 Secondary Analysis

3.4.3.1 Characteristics of External Perturbations

The characteristics of the external perturbations were quantified to determine whether

S and L perturbations differed in magnitude and temporal characteristics (Table 3-2). Peak force

was significantly lower for small perturbations (BS and RS) compared to large perturbations

(F1.3,14.6=129.1, p<0.00005, partial η2 =.92); however, there was also a significant difference

between BS and RS (p<0.05). In spite of these differences, the stepping behaviors were

appropriately suited to the specific task condition. That is, all participants executed a feet-in-

place compensatory response for small perturbations in both the Block and Random conditions.

A significant main effect of Condition for %LC-COPonset (F3,33=5.50, p=0.004, partial η2 =.33)

demonstrated that random conditions required less force, with COP onset requiring significant

less force application in random conditions (RS and RL) than BS (p≤0.02).

61

3.5 Discussion

This study examined whether set modulation was revealed in the extent of inter-limb

postural synchronization prior to bouts of postural instability. As hypothesized, our results

demonstrated relatively higher levels of AP and ML inter-limb synchrony when the magnitude

of the perturbation was most predictable and of low magnitude. Contrary to our hypothesis,

there were no significant differences across perturbation types (Internal vs. External) in inter-

limb synchrony during preparation.

3.5.1 Condition-dependent Differences in Preparatory COP

Information about the magnitude, timing and direction of the forthcoming perturbation

influences preparation for instability (Horak et al., 1989; Maki & McIlroy, 2007). In our study,

timing and direction were maintained across conditions while perturbation magnitude in the

forward direction remained the varying factor. Previous studies have shown responses to

instability which scale to the magnitude of the perturbation, and in addition, scale in magnitude

in parallel with the cortical activity associated with the anticipated differences in magnitude

(Mochizuki et al., 2010). Importantly, in conditions where the amplitude of forthcoming

instability was unknown, the amplitude of cortical potentials scaled in magnitude to the larger

of two perturbation amplitudes. These findings indicate that the CNS scales readiness to the

magnitude of imminent instability or to the larger of possible perturbation amplitudes. The

current study found that the randomization of the perturbation magnitudes resulted in a

decrease in the extent of inter-limb COP synchrony in comparison to when trials were blocked.

In the ML direction preparing for a step or not knowing the magnitude of the perturbation also

62

elicited different preparatory COP excursions. Evidence from the perturbation force thresholds

(%LC-COPonset) demonstrated that randomized conditions required less force for individuals to

begin moving in a forward direction. Combined, these findings suggests that randomization of

perturbations reduces the threshold needed for evoking a response.

The paradigm used in this study provided cueing prior to perturbation onset, which

redirects attention to the corresponding task. Concurrent visual tracking tasks and balance

perturbations reduce the amplitude of post-perturbation cortical evoked potentials and are

associated with increased AP COP displacement and muscle activity (Quant, Adkin, Staines,

Maki, & McIlroy, 2004). Inter-limb synchrony was highest in both AP and ML directions during

BS conditions, and in comparison to QS, AP inter-limb synchrony was not different. Accordingly,

BS conditions may have required minimal additional motor tasks in comparison to other

blocked and random conditions. Thus, the increased inter-limb synchrony in BS, representing

efficient control, may be associated with lower cognitive demands and suggests that the other

conditions required more attentional resources to prepare for instability resulting in decreased

inter-limb synchrony. Alternately, the reduction in inter-limb synchrony from QS to BS and from

BS to the other conditions in the ML direction may reflect a decoupling of the COP during

movement preparation to enable EPAs to effectively contribute to balance reactions (Klous et

al., 2012). The combinations of high inter-limb synchrony in the AP direction combined with the

lower ML inter-limb synchrony may, therefore, represent the most efficient method to prepare

for unpredictable perturbations of different magnitudes. Decoupling of inter-limb synchrony

prior to a perturbation may arise as a result of the different roles of each limb during

compensatory stepping.

63

3.5.2 Perturbation Type of Preparatory COP Activity

The results demonstrated that inter-limb synchrony did not differ between Internal and

External perturbations. Self-initiated movements use different neural networks during

preparation when tasks are associated with timing and choice of movements (Hoffstaedter,

Grefkes, Zilles, & Eickhoff, 2013). This may have contributed to different strategies

implemented and greater between-subject variability in inter-limb synchrony within Internal

perturbations in comparison to External perturbations (Figure 3-2C/F). Between-subject

variability may have also resulted from differences in arousal levels (Maki & Whitelaw, 1993;

Sibley, Mochizuki, Frank, & McIlroy, 2010) associated with positioning of the experimenter

behind the participant.

Similarities in the magnitude of inter-limb synchrony during preparation of postural

instability between perturbation types suggest that regardless of the mechanism that drives the

execution of movement it may be the information present in cues themselves (i.e. timing, task

requirements) that is common and salient. If the physiological importance of set is to use past

experience and current context to optimize movement, the motor task being performed may

be irrelevant. During preparation, self-generated or externally generated perturbations elicited

similar physiological characteristics (Mochizuki et al., 2009). In a different motor task, whole

body-balance disturbances paired with auditory cues elicited faster reaction times compared to

auditory cueing alone. This was maintained when whole body-balance disturbances were

omitted (Lakhani, Miyasike-Dasilva, Vette, & McIlroy, 2013). This work demonstrates that the

current context of balance instability and auditory cuing led to an adaptation of the task

64

requirements. Similarly, inter-limb synchrony may be associated with scaling of set rather than

distinct preparation for self-initiated movements.

The only difference between perturbation types was observed in the First5 and Last5

comparison in which trial-by-trial modulation was probed. During External perturbations, inter-

limb synchrony significantly decreased in the condition where perturbation magnitude was

unknown. Differences in AP inter-limb synchrony between the First5 and the Last5 trials were

observed in RL condition. No such differences were observed for Internal perturbations.

Consistent with the idea that reduced synchrony with Large perturbations allows for efficient

decoupling of bilateral control, it is possible that in the initial RL trials of External perturbations,

individuals were unaware of the extent to which decoupling should have occurred. By the time

they got to the Last5 trials, participants learned to use decoupling of inter-limb synchrony to

optimize balance responses in situations where perturbation magnitude was unknown. In

contrast, overt motor planning for the stepping response for internally generated perturbations

may have already used decoupling of the limbs within the first few trials.

3.6 Conclusions

In examining whether set modulation is evident in changes in inter-limb synchronization

prior to bouts of postural instability, it was determined that inter-limb synchrony decreased as

perturbation magnitude became more unpredictable. Moreover, larger or unpredictable

perturbations reduced preparatory ML inter-limb synchrony. These findings identify

65

characteristic adaptations to control bilateral activity in preparation for instability and provide

important insight into preparatory postural control.

66

3.7 References

Adkin, A. L., Campbell, A. D., Chua, R., & Carpenter, M. G. (2008). The influence of postural

threat on the cortical response to unpredictable and predictable postural perturbations.

Neuroscience Letters, 435(2), 120-125. doi:10.1016/j.neulet.2008.02.018

Cordo, P. J., & Nashner, L. M. (1982). Properties of Postural Adjustments Associated with Rapid

Arm Movements. Journal of Neurophysiology, 47(2), 287-302.

Elais, L. J., Bryden, M. P., & Bulman-Fleming, M. B. (1998). Footedness is a better predictor than

is handedness of emotional lateralization. Neuropsychologia, 36(1), 37-43.

Habib Perez, O., & Mochizuki, G. (2014). Inter-limb synchrony is increased prior to bouts of

volitional stepping. Archives of Physical Medicine and Rehabilitation, 95(10), e91.

doi:http://dx.doi.org/10.1016/j.apmr.2014.07.303

Hoffstaedter, F., Grefkes, C., Zilles, K., & Eickhoff, S. B. (2013). The "what" and "when" of self-

initiated movements. Cerebral Cortex, 23(3), 520-530. doi:10.1093/cercor/bhr391



Horak, F. B., Esselman, P., Anderson, M. E., & Lynch, M. K. (1984). The effects of movement

velocity, mass displaced, and task certainty on associated postural adjustments made by

normal and hemiplegic individuals. Journal of Neurology, Neurosurgery, and Psychiatry,

47(9), 1020-1028.



Kerr, B., Condon, S. M., & McDonald, L. A. (1985). Cognitive spatial processing and the

regulation of posture. Journal of Experimental Psychology: Human Perception and

Performance, 11(5), 617-622. doi:10.1037/0096-1523.11.5.617



116. doi:10.1016/j.jelekin.2011.11.005

Lakhani, B., Mansfield, A., Inness, E. L., & McIlroy, W. E. (2011). Compensatory stepping

responses in individuals with stroke: a pilot study. Physiotherapy Theory & Practice,

27(4), 299-309. doi:http://dx.doi.org/10.3109/09593985.2010.501848

Lakhani, B., Miyasike-Dasilva, V., Vette, A. H., & McIlroy, W. E. (2013). 'Priming' the brain to

generate rapid upper-limb reactions. Experimental Brain Research, 230(3), 261-270.

doi:10.1007/s00221-013-3650-x

67

Maki, B. E., & McIlroy, W. E. (2007). Cognitive demands and cortical control of human balance-

recovery reactions. Journal of Neural Transmission, 114(10), 1279-1296.

doi:10.1007/s00702-007-0764-y

Maki, B. E., & Whitelaw, R. S. (1993). Influence of expectation and arousal on centre-of-

pressure responses to transient postural perturbations. Journal of Vestibular Research,

3(1), 25-29.










Mochizuki, G., Ivanova, T. D., & Garland, S. J. (2004). Postural muscle activity during bilateral

and unilateral arm movements at different speeds. Experimental Brain Research, 155(3),

352-361.



doi:10.1152/jn.01322.2004

Mochizuki, G., Sibley, K. M., Cheung, H. J., & McIlroy, W. E. (2009). Cortical activity prior to

predictable postural instability: Is there a difference between self-initiated and

externally-initiated perturbations? Brain Research, 1279, 29-36.

doi:10.1016/j.brainres.2009.04.050

Mochizuki, G., Sibley, K. M., Esposito, J. G., Camilleri, J. M., & McIlroy, W. E. (2008). Cortical

responses associated with the preparation and reaction to full-body perturbations to

upright stability. Clinical Neurophysiology, 119(7), 1626-1637.

doi:10.1016/j.clinph.2008.03.020



Quant, S., Adkin, A. L., Staines, W. R., Maki, B. E., & McIlroy, W. E. (2004). The effect of a

concurrent cognitive task on coritcal potentials evoked by unpredictable balance

perturbations. BMC Neuroscience, 5(18), 1-12. doi:doi:10.1186/1471-2202-5-18

68

Rankin, J. K., Woollacott, M. J., Shumway-Cook, A., & Brown, L. A. (2000). Cognitive influence on

postural stability: A neuromuscular analysis in young and older adults. Journal of

Gerontology, 55A(3), M112-M119.

Remaud, A., Boyas, S., Caron, G. A., & Bilodeau, M. (2012). Attentional demands associated

with postural control depend on task difficulty and visual condition. Journal of Motor

Behavior, 44(5), 329-340. doi:10.1080/00222895.2012.708680

Shumway-Cook, A., & Woollacott, M. J. (2000). Attentional Demands and Postural Control: The

effects of sensory context. Journal of Gerontology, 55A(1), M10-M16.

Sibley, K. M., Mochizuki, G., Frank, J. S., & McIlroy, W. E. (2010). The relationship between

physiological arousal and cortical and autonomic responses to postural instability.






Siu, K. C., & Woollacott, M. H. (2007). Attentional demands of postural control: the ability to

selectively allocate information-processing resources. Gait and Posture, 25(1), 121-126.

doi:10.1016/j.gaitpost.2006.02.002




69

Chapter 4 : Characterization of balance control after

moderate-to-severe traumatic brain injury: A longitudinal

recovery study

Habib Perez, O., Green, R. E. & Mochizuki, G. Characterization of balance control after

moderate-to-severe traumatic brain injury: A longitudinal recovery study. Under Review in

Physical Therapy

70

4.1 Abstract

Background: Balance impairments after traumatic brain injury (TBI) are common and persist

after injury. Postural asymmetries in balance have been reported, but not quantified across

recovery.

Objective: To characterize balance recovery after moderate-to-severe TBI with a focus on

postural asymmetry.

Methods: A secondary analysis of longitudinal balance data from 45 participants with

moderate-to-severe TBI was conducted. Balance was assessed with force plates at

approximately two, five and twelve months post-injury in two bipedal stances and two unipedal

stances for TBI participants. Single-visit data from age-matched healthy control was collected

for visual comparison. Spatial and temporal centre of pressure (COP) measures were calculated

from force plates in the anteroposterior (AP) and mediolateral (ML) directions.

Results: Despite improvements in net ML COP postural sway from two-to-five months post

injury, there were no changes across recovery in AP postural sway. Postural sway in TBI was

greater than normative values at all time points in both directions. Inter-limb synchrony did not

change across recovery in either direction. TBI weight-bearing asymmetry was lower than

normative values at all time points and did not change across recovery. The characteristics of

unipedal stance differed between limbs.

Conclusions: The absence of recovery in ML COP postural sway, inter-limb synchrony, and

weight-bearing symmetry indicate that ML control may contribute to balance impairments after

71

TBI. These impairments may extend to dynamic balance tasks and may also place individuals

with TBI at higher risk of falls.

72

4.2 Introduction

Traumatic brain injury (TBI) is among the leading causes of disability in North America

(Colantonio, Croxford, Farooq, Laporte, & Coyte, 2009). Balance control, defined as the ability

to maintain one’s center of mass within the base of support, is often impaired after TBI (Walker

& Pickett, 2007). These impairments persist years after injury, particularly in individuals with

more severe TBIs (Ponsford et al., 2014). Previous studies of balance control after moderate-to-

severe TBI have characterized balance impairments at different stages of recovery or have

employed cross-sectional comparisons of balance measures to those of healthy controls or to

published normative values. These studies have found impairments in balance at the acute and

chronic stages post injury (Basford et al., 2003; Geurts, Ribbers, Knoop, & van Limbeek, 1996;

Lehmann et al., 1990; Pickett, Radfar-Baublitz, McDonald, Walker, & Cifu, 2007; Wade, Canning,

Fowler, Felmingham, & Baguley, 1997; Wöber et al., 1993).

Longitudinal examinations of balance impairment after injury have been sparse. One

study showed that improvements in postural sway are observed within the first months of

injury (Wade et al., 1997). To our knowledge, in the only longitudinal balance study beyond the

early recovery phase, postural stability was examined over a two-year recovery window using

the Romberg test (Walker & Pickett, 2007) rather than quantitative analysis of posturographic

measures. While the authors (Walker & Pickett, 2007) reported an improvement in the number

of individuals who completed the Romberg test post-injury, no longitudinal studies employing

posturographic outcomes to quantitatively measure balance impairment have been undertaken

beyond the first months of injury (Wade et al., 1997). Thus, it remains unclear whether

73

continued balance recovery might be observed in the later stages of recovery with more

sensitive outcome indices.

Typical posturographic analyses of balance report net measures of control (i.e.,

contributions of both limbs) as their primary outcome (Brauer, Broome, Stone, Clewett, &

Herzig, 2004; Geurts et al., 1996; Wade et al., 1997). Net measures of balance identify postural

sway metrics in the anteroposterior (AP) and mediolateral (ML) directions, reflecting ankle and

hip motion control, respectively (Winter, 1995). Although net postural sway quantifies overall

balance control, it lacks specificity with regard to the individual limb contributions to control

processes (Winter, Prince, Stergiou, & Powell, 1993). This is especially important considering

evidence that unilateral motor weakness and asymmetrical stance persist following TBI (Choi et

al., 2012; Drijkoningen et al., 2015; Newton, 1995). The extent of asymmetry specific to balance

control has not yet been quantified. Because TBI may contribute to asymmetrical balance,

which is linked to elevated fall risk in other neurologically-injured populations (Keenan, Perry, &

Jordan, 1984; Mansfield, Mochizuki, Inness, & McIlroy, 2012), it is important to characterize

recovery trajectories in the context of asymmetry.

Balance asymmetry has been investigated in neurologically impaired populations, such

as stroke, where unilateral weakness/impairment is common (Keenan et al., 1984). Between-

limb centre of pressure (COP) synchronization (the spatial and temporal relationships between

the COP of both feet) is a sensitive measure of balance control during static standing in healthy

adults (Mochizuki, Ivanova, & Garland, 2005; Winter et al., 1993) and in adults after stroke

(Mansfield, Danells, Inness, Mochizuki, & McIlroy, 2011; Mansfield et al., 2012; Singer,

Mansfield, Danells, McIlroy, & Mochizuki, 2013; Singer & Mochizuki, 2015). In stroke patients,

74

synchronization is reduced and delayed during standing balance tasks (Mansfield et al., 2011).

In addition, single leg standing identifies postural asymmetry that may be an indicator of

unilateral weakness or impairment. Because both bipedal and unipedal stances are used in

functional balance assessments in other neurologic populations across recovery to identify

underlying balance deficits (Inness et al., 2011; Mancini & Horak, 2010), they may also identify

postural asymmetries after TBI.

The purpose of this study was to characterize the recovery trajectory of balance control

following TBI with a specific focus on measures of postural asymmetry. It was hypothesized that

(1) net balance measures would improve over the course of recovery; (2) individuals with TBI

would demonstrate postural asymmetry in measures of asymmetry, including inter-limb

synchrony, stance load symmetry, and postural sway and time in unipedal stance; and (3)

measures of postural asymmetry would improve across recovery.

4.3 Methods

4.3.1 Participants

The experimental group (TBI participants) was drawn from the database of The Toronto

Rehab Traumatic Brain Injury Recovery Study (Christensen et al., 2008; Till, Colella, Verwegen, &

Green, 2008), a prospective longitudinal study of cognitive and motor recovery, for which

motor assessments were undertaken at approximately 2, 5 and 12 months post-injury. The

study was carried out on a large, urban acquired brain injury program. Inclusion criteria for the

Recovery Study included moderate-to-severe TBI as indicated by a Glasgow Coma Scale (GCS) of

75

12 or less and/or post traumatic amnesia (PTA) of 1 day or more and/or positive neuroimaging

findings; out of PTA by 3 months post-injury; aged between 17 and 75; able to use at least one

upper extremity; and, functional command of English. Exclusion criteria for the larger study

included: past history of TBI and history of psychotic or neurological illness. An additional

inclusion criterion for the current study was the completion of a motor assessment at each time

point. Over the course of the Recovery Study, different technologies were used to assess

balance. Initially, assessments were conducted on a single force plate. Later in the Recovery

Study, assessments were conducted on dual (adjacent) force plates in order to obtain data from

each individual limb. Consequently, an exclusion criterion for the current study was the use of

mixed force plate techniques (i.e. single and dual force plates) across different time points for

an individual participant. Another exclusion criterion was the presence of lower extremity

orthopedic injury.

Ninety-seven participants with clinically-confirmed TBI completed instrumented balance

assessments, 51 of whom completed all three balance assessments using either a single or dual

force plate approach. Six of these individuals were excluded for lower extremity orthopedic

injuries. Therefore, 45 TBI participants were included in the analysis, with 14 and 31

participants completing balance assessments post-injury on single and dual force plates,

respectively. An age-matched group of 22 healthy control (HC) participants were recruited from

the university and local community and assessed on one occasion. All participants (or their

substitute decision maker) provided informed consent for the original study. The current study

was approved by the Research Ethics Board at the Toronto Rehabilitation Institute.

76

4.3.2 Study Design

The study employed a secondary analysis of prospectively collected data. Within-

subjects analysis enabled characterization of recovery over time. The primary balance-related

outcome measures included root mean square (RMS) of the net (both feet combined) COP in

the AP and ML directions, cross-correlation of the COP in the AP and ML directions (all in

bipedal stance), and unipedal stance time and RMS in unipedal stance.

4.3.3 Data Collection and Procedures

Posturographic (force plate) and clinical measures of balance (The Community Balance

and Mobility Scale [CB&M]) (Howe, Inness, & Wright, 2011) were assessed at approximately 2-

months (T1), 5-months (T2), and 12-months (T3) post injury. Data were collected using either a

single force plate (AccuSway, Advanced Mechanical Technology, Inc. (AMTI)) or dual force plate

(AMTI, Watertown Massachusetts) arrangement.

For the single force plate assessments, participants placed both feet on one force plate.

For the dual force plates, the plates were positioned with the Y-axes in parallel and separated

by 1 mm, with participants placing one foot on each force plate. For both the single and dual

force plate arrangements, participants stood in two bipedal and two unipedal conditions:

standardized position (McIlroy & Maki, 1997) with eyes open (EO) and eyes closed (EC), and

unipedal stance with the right leg (UR) and the left leg (UL). For all bipedal conditions,

participants were instructed to stand quietly for 45 seconds looking straight ahead. For

77

unipedal stances, participants were instructed to maintain their balance without touching the

ground as best as possible for 30 seconds. Ground-reaction forces and moments from each

plate were sampled at 50 Hz for all TBI participant data. The sampling rate for all TBI

assessments was kept consistent regardless of whether the single or dual force plate

arrangement was used. HC data were collected at one time point for all balance tasks for visual

comparison.

Comprehensive neuropsychological assessments were conducted, including measures of

attention and executive function (Trails Making Tests [TMT] A and B, respectively) (Lezak,

1995), Symbol Digit Modalities Test (SDMT) oral (Smith, 1982), and estimated pre-morbid IQ, as

measured by the Wechsler Test of Adult Reading (WTAR) (Wechsler, 2001). In the current

study, WTAR was included for patient characterization. The TMT and SDMT, two highly

sensitive TBI measures, were employed to demonstrate the presence of cognitive impairment

and to provide evidence of cognitive recovery (Perianez et al., 2007; Reitan & Wolfson, 1985).

4.3.4 Data Processing and Analysis

Ground-reaction forces and moments were low-pass filtered using a 4th ordered dual-

pass Butterworth filter with a 10 Hz cut off frequency prior to processing. In bipedal stance, the

net COP was calculated. The root mean square (RMS) of the AP and ML COP time series was

calculated for all participants. For bipedal stance, inter-limb synchrony was measured by

calculating the cross-correlation function between the left and right mean-removed AP and ML

COP waveforms. Inter-limb synchrony was defined by the cross-correlation coefficients at zero

78

phase-lag (Rxy(0)), which were calculated at incremental phase-shits defined by the sampling

rate, by iteratively shifting the right limb COP forwards and backwards in time over the entire

length of the record (Singer et al., 2013). The magnitude of the coefficient (ranging from -1.0 to

1.0) illustrates the strength of correlation of the COP time series.

Weight-bearing symmetry was calculated using two measures. Stance load symmetry

was defined as a ratio of the vertical force applied onto one force plate, relative to the sum of

the vertical forces applied onto both force plates. This was determined for both the left and

right side. Absolute stance load symmetry was defined as the stance load symmetry ratio of

the less-loaded limb (regardless of whether it was the left or right side) (Mansfield et al., 2011).

Unipedal stance time and RMS were used to characterize asymmetry in weakness and

postural instability, respectively. Time in unipedal stance was calculated as the time from foot

lift off to the first drop down contact if the participant could not sustain the full 30 seconds.

RMS during unipedal stance was only calculated for the duration over which participants

maintained a unipedal standing position. All balance data were processed using MATLAB

R2014a (MathWorks, Natick, USA).

4.3.5 Statistical Analyses

Statistical analyses were performed using SPSS v23.0 (IBM SPSS Statistics, Armonk, USA).

Descriptive and exploratory analyses of changes in attention and executive function over time

in the TBI group were conducted using a one-way analysis of variance (ANOVA). To test the

hypothesis that net measures of balance (net AP and ML COP RMS) would improve across

79

recovery for TBI participants, a series of 3×2 two-way ANOVA were conducted to assess the

main effects of Time (T1, T2, T3) and Condition (EO, EC). A one-way ANOVA was used to probe

the effect of Time on a clinical and complementary measure of balance (the CB&M) in the TBI

group. To test the hypothesis that postural symmetry would improve across recovery, a series

of 3×2 two-way ANOVA was conducted to analyze the effects of Time and Condition on features

of the Rxy(0) in the AP and ML direction. Because the Rxy(0) values are bounded by values of -1

and 1, a Fisher transform was applied. Hypothesis testing for improvement in postural

symmetry over time also included paired t-tests to determine whether stance load symmetry

was different between the left and right foot at each time point in TBI participants and a 3×2

two-way ANOVA was conducted to analyze the effects of Time and Condition on absolute

stance load symmetry. The final probe of the hypothesis of improved symmetry over time, as

measured by unipedal standing duration and AP and ML RMS, were a series of 3×2 within-

subjects ANOVAs with Time (T1, T2, T3) and Leg (Left, Right) as factors.

Data were log-transformed if the values were not normally distributed. In cases where

the assumption of sphericity was not met, Greenhouse-Geisser values were reported. Statistical

significance was set at p<0.05.


Additional exploratory analyses were conducted to test relationships between: inter-

limb synchrony and postural sway and inter-limb synchrony and absolute stance load symmetry

80

in individuals with TBI. Spearman correlations were conducted between Rxy(0) and COP RMS

and absolute stance load symmetry (Mansfield et al., 2011).

4.4 Results

4.4.1 Demographic Characteristics

All demographic information for TBI participants and HC are listed in Table 4-1. TBI

participants were assessed on average at 64.7 (±28.1), 156.2 (±36.5), and 401.5 (±60.6) days

post injury at T1, T2 and T3, respectively. There was no significant difference in age between

TBI and HC participants (t(51.0) = 1.7, p>0.05, d = 0.42) and there were no significant

differences in mass between the HC group and TBI group at T1 (t(65) = 1.3, p>0.05, d = 0.34). A

significant increase in mass within the TBI group across Time (F1.3,61.5 = 42.6, p<0.001, partialη2 =

.49) from T1 to T2 (p<0.001) and T2 to T3 (p<0.001) was observed. The HC group completed

more years of education in comparison to TBI participants (t(64) = -3.0, p=0.004, d = 0.77).

81

Table 4-1 – Participant demographics of traumatic brain injury (TBI) and healthy control (HC)

participants. Demographics of TBI participants for single and dual force plate collections are

also listed. TBI severity was characterized by the lowest post-traumatic amnesia (PTA) scale and

the number of TBI participants in each category is described below. The Community Balance

and Mobility (CB&M) scale is listed for the TBI group. Cognitive function includes years of

education, Wechsler Test of Adult Reading (WTAR), Trails Making Tests (TMT) A and B, and

Symbol Digit Modalities Test (SDMT).

TBI HC

Force plate

arrangement

Single & Dual

(all participants)

Single Dual Dual

Sample size 45 (36 M/9 F) 14 (12 M/2 F) 31 (24 M/7 F) 22 (10 M/12 F)

Age 43.2 (±17.3) 36.6 (±13.8)

Mass (kg)

T1 74.1 (±13.5) 69.6 (±12.8)

T2 78.7 (±14.3) * --

T3 82.0 (±15.9) *# --

Severity (PTA)

Moderate 1 0 1 --

Severe 9 3 6 --

Very Severe 23 7 16 --

Extremely Severe 7 3 4 --

CB&M Scale

T1 60.8 (±20.3)

T2 71.3 (±18.0)*

T3 74.2 (±16.7)*

Cognitive Function

Years of education 14.5 (±3.2)☨ 17 (±3.4)

WTAR 114.6 (±9.1) --

TMT A (t-score)

T1 41.5 (±10.7) 35.3 (±8.1) 44.2 (±10.7) --

T2 47.4 (±13.3)* 37.9 (±8.5) 51.7 (±12.9) --

T3 48.8 (±14.3)* 39.2 (±9.3) 53.0 (±14.1) --

TMT B (t-score)

T1 47.3 (±14.5) 42.6 (±11.0) 49.6 (±15.5) --

T2 49.7 (±13.7) 41.9 (±8.1) 53.4 (±14.3) --

T3 50.9 (±12.5) 44.2 (±9.1) 53.9 (±12.8) --

SDMT oral

T1 47.8 (±16.2) 40.5 (±11.9) 51.0 (±16.9) --

T2 52.5 (±17.3) * 43.2 (±13.8) 56.6 (±17.3) --

T3 57.1 (±20.0) *# 46.8 (±17.2) 62.7 (±19.7) --

82

*denotes significant differences within TBI participants at p<0.05 from T1 value. #denotes

significant differences within TBI participants at p<0.05 from T2 value. ☨ denotes significant

differences between HC and TBI at p<0.05. Note that PTA severity was obtained from 40 of the

45 TBI participants. TMT A, TMT B and SDMT oral was obtained from 42, 41 and 42 TBI

participants respectively. WTAR was obtained from 28 TBI participants at 12 months post

injury.

4.4.2 Cognitive Characteristics

Data for the WTAR, TMT and SDMT oral are presented in Table 4-1. There was a

significant effect of Time for the TMT A (F2,82 = 14.2, p<0.001, partialη2 = .26) and the SDMT oral

(F2,82 = 30.0, p<0.001, partialη2 = .42), with improvements across recovery for both measures.

Bonferroni adjusted post-hoc analysis identified that TMT A significantly increased from T1 and

T2 (p=0.001) and T1 and T3 (p<0.001). Bonferroni adjusted post-hoc analysis demonstrated that

SDMT oral significantly increased from T1 and T2 (p<0.001) and from T2 and T3 (p=0.001). No

significant change across Time was observed on the TMT B measure.

4.4.3 Clinical Measures of Balance

Analysis of the CB&M scale identified a significant effect of Time (F1.6,62.0 = 31.8, p<0.05,

partial η2 = .46). Bonferroni post-hoc tests identified that the improvements in clinical balance

measures were statistically significant only between T1 and T2 (p<0.001) and between T1 and

T3 (p<0.001). There was no statistically significant difference between T2 and T3 (p>0.05).

83

4.4.4 Net Measures of Balance

4.4.4.1 Bipedal COP RMS

The 3×2 ANOVA demonstrated that there were no changes across recovery in AP RMS;

however, ML RMS decreased across time. There was a significant within-subjects reduction in

the measure of ML RMS across Time (F1.6,66.7=4.32, p=0.024, partial η2 = .10) for TBI participants.

Bonferroni adjusted post-hoc analysis identified a significant reduction for the ML RMS from T1

to T2 (p=0.015). There was a significant Condition effect in AP RMS, with TBI participants

producing greater RMS in EC compared to EO (F1,42=55.0, p<0.001, partial η2 = .57). Figure 4-1

demonstrates the results for TBI participants and normative ranges of HC. AP and ML RMS for

TBI participants fell outside the HC ranges at all time points.

84

Figure 4-1 – Anteroposterior (AP) (A) and mediolateral (ML) (B) centre of pressure root mean


open (EO; black) and eyes closed (EC; dark grey). The solid and dashed lines represent mean

and 95% confidence intervals for COP RMS in healthy controls in each condition, respectively.

Asterisks denotes significant difference across Time (T1 and T2) for TBI participants at p<0.05.

85

4.4.5 Measures of Asymmetry

4.4.5.1 Inter-limb COP Temporal Synchrony

Inter-limb synchrony did not change across recovery in either the AP or ML direction;

there were no significant effects of Time for AP Rxy(0) and ML Rxy(0) (Figure 4-2A and B, p>0.05).

A significant Condition effect was found in AP Rxy(0) (F1,27=14.9, p=0.001, partial η2=.36), and

approaching significance for ML Rxy(0) (p=0.058). AP Rxy(0) in EC was significantly higher than in

EO. AP inter-limb synchrony in TBI participants was within normal ranges of HC; however, mean

ML inter-limb synchrony in TBI participants was outside the HC ranges in EO.

86

Figure 4-2 – Anteroposterior (AP) (A) and mediolateral (ML) (B) inter-limb synchrony (Rxy(0))

group means and standard error for TBI participants across recovery in eyes open (EO; black)

and eyes closed (EC; dark grey). The solid and dashed lines represent mean and 95% confidence

intervals for Rxy(0) in healthy controls in each condition, respectively.

87

4.4.5.2 Weight-bearing Symmetry

Paired t-tests within the TBI participants demonstrated that there were no stance load

asymmetries between the left and right limbs (p>0.05) at each time point. Absolute stance load

symmetry did not significantly change across recovery for TBI participant (p>0.05) and there

were no significant differences between EO and EC (p>0.05). However, TBI participants at all

time points fell outside HC ranges in both EO and EC across their recovery (Figure 4-3).

Figure 4-3 – Mean and standard error of absolute stance load symmetry ratio of the less loaded

limb for TBI participants across recovery in eyes open (EO; black) and eyes closed (EC; dark

grey). The solid and dashed lines represent mean and 95% confidence intervals for the absolute

stance load symmetry ratio in healthy controls in each condition, respectively.

88

4.4.5.3 Unilateral Stance

Of the 31 participants whose data were collected on dual force plates, two participants

were unable to complete unipedal stance at T1 and were excluded from the within-subjects

ANOVA analyses. Unilateral stance duration demonstrated asymmetries in TBI participants

(Figure 4A). A significant main effect of Time (F2,56=10.1, p<0.001, partial η2 = .27) and Leg

(F1,28=6.3, p=0.018, partial η2 = .18) was observed. The left leg maintained a significantly longer

duration in unilateral stance in comparison to the right leg (p=0.018). Bonferroni adjusted post-

hoc analysis demonstrated that TBI participants significantly improved unilateral stance

duration between the first two time points (p<0.001); however, time in unipedal stance showed

a trend towards deteriorating at T3 (p=0.067). Unipedal stance duration in TBI participants fell

outside HC ranges throughout their recovery. Asymmetries in TBI participants were not

identified in unilateral stance RMS. The 3×2 within-subjects ANOVA revealed no significant

main effects of Time or between Leg in AP and ML RMS in TBI participants (p>0.05; Figure 4-4B

and 4-4C).

89

Figure 4-4 – Time in unipedal stance in the left (UL; black) and right (UR; grey) leg group mean

and standard error for TBI participants across recovery (A). Anteroposterior (AP) (B) and

mediolateral (ML) (C) centre of pressure root mean square (COP RMS) group means and

standard error for TBI participants across recovery in the left (UL; black) and right (UR; grey) leg.

The solid and dashed lines represent mean and 95% confidence intervals in healthy controls in

each condition, respectively. Note that there are no dashed lines for UR in panel A. Asterisk in

(A) denotes significant difference between T1 and T2 in TBI participants


Exploratory analyses of Rxy(0) were conducted with COP RMS and absolute stance load

symmetry to test relationships between these measures. Reduced AP Rxy(0) was significantly

associated with greater ML RMS across all three time points, with the strongest relationship at

T1 (rs = -0.55 p = 0.002; see Table 4-2). Increased ML Rxy(0) was significantly correlated to lower

90

ML RMS at T1 and T2. There were no statistically significant relationships between AP COP RMS

with either AP or ML Rxy(0). Absolute stance load symmetry did not demonstrate any statically

significant relationships with AP or ML Rxy(0). Scatter plots for significant relationships are

shown in Appendix A.

Table 4-2 – Relationship between inter-limb synchronization Rxy(0) and standing balance

measures (centre of pressure root mean square (COP RMS) and absolute stance load symmetry)

for eyes open. Spearman correlation coefficients and the associated p-values are shown in

parentheses. Statistically significant correlations and p-values are highlighted in bold font.

AP Rxy(0) ML Rxy(0)

AP COP RMS

T1 0.082 (0.67) - 0.053 (0.79)

T2 0.27 (0.14) - 0.17 (0.36)

T3 0.14 (0.46) - 0.19 (0.31)

ML COP RMS

T1 - 0.55 (0.002) 0.49 (0.007)

T2 - 0.40 (0.027) 0.49 (0.005)

T3 - 0.40 (0.026) 0.28 (0.13)

Absolute stance load symmetry

T1 0.20 (0.30) - 0.22 (0.25)

T2 0.10 (0.58) - 0.22 (0.25)

T3 0.28 (0.13) - 0.009 (0.96)

4.5 Discussion

This study characterized balance control across recovery after TBI with a focus on

quantifying postural asymmetry. As hypothesized, improvements were observed from two to

five months of injury in the ML direction; however, in contrast to the study hypotheses, net

91

measures of balance in the AP direction did not improve across recovery in individuals with TBI.

Individuals with TBI demonstrated postural asymmetry in comparison to normative data ranges

of absolute stance load asymmetry, reduced ML inter-limb synchrony, and lower unipedal

stance time. In contrast to the study hypothesis, postural asymmetry measures did not improve

across recovery. Overall, the findings indicated that some balance impairments do improve

over the course of recovery after moderate-to-severe TBI from 2 to 5 months post-injury,

though not beyond.

4.5.1 Balance Improvements Within the First Five Months after TBI

Overall ML balance control, as measured by ML RMS, improved from T1 to T2 post-

injury. Additionally, functional balance and mobility (CB&M) improved from T1 to T2, but

plateaued from T2 to T3. Consistent with previous studies (Wade et al., 1997; Walker & Pickett,

2007), the greatest improvements in balance control occurred within the first five months after

TBI. Net ML postural sway is controlled by hip abductors and adductors, and reflects the ability

to load and unload the vertical force from one limb to another (Winter et al., 1993). The

reduction in ML sway from two months to five months after injury suggests that the CNS may

prioritize the reduction ML RMS to increase stability and reduce risk of falls (Maki, Holliday, &

Topper, 1994). Tightened AP postural sway resulting from increased ML sway has been

reported after the onset of Parkinson’s disease (Mitchell, Collins, De Luca, Burrows, & Lipsitz,

1995), and increased ML sway has been associated with a reduction in occipital and cerebellar

brain volume in individuals with right cerebral artery infarctions (Manor et al., 2010). The

overlap in balance impairments between different neurologically impaired populations

92

indicates that reduced balance control in the ML direction may be a common consequence of

neurological injury. The current study found greater AP and ML RMS values than Geurts et al.

(1996), but similar values to that of Brauer et al. (2004) and may be attributable to

comparability of TBI severity used in the studies.

Duration in unipedal stance recovered in individuals after TBI. Time in unipedal stance

increased by approximately 10 seconds from T1 to T2, suggesting improvements in strength

and postural steadiness in both limbs (Jonsson, Seiger, & Hirschfeld, 2004). Unipedal stance is a

common task that is included in various clinical balance assessments and is associated with

increased fall risk (Hurvitz, Richardson, Werner, Ruhl, & Dixon, 2000; Springer, Marin, Cyhan,

Roberts, & Gill, 2007). Previous studies looking at unipedal stance duration after moderate-

severe TBI report reduced duration in patients with chronic TBI (Williams & Morris, 2011) and

similar unipedal stance time to that of T2 of the current study (McFadyen, Swaine, Dumas, &

Durand, 2003). Unipedal stance time is associated with mobility (Williams & Morris, 2011), and

thus may be an important assessment during the course of recovery.

In contrast to our hypotheses, measures of balance control focusing on postural

asymmetry in bipedal stance did not show changes across recovery. Though inter-limb

synchrony in the AP and ML direction did not improve post-injury, significant relationships were

found between inter-limb synchrony and ML RMS. The secondary analyses of AP and ML inter-

limb synchrony and ML COP RMS suggest that inter-limb synchrony was linked to ML postural

sway. Similar findings have been noted in individuals post stroke (Mansfield et al., 2011).

Absolute stance load symmetry also did not improve across recovery. Unlike the previous study

(Mansfield et al., 2011), the present study did not find a relationship between inter-limb

93

synchrony and weight-bearing asymmetry. These variables are sensitive to postural asymmetry

in neurological populations with hemiparesis (Mansfield et al., 2012) and spasticity (Singer et

al., 2013); however, research has shown similar results with initial improvements in the first six

months, followed by plateau at 12 months and a subsequent decline by 24 months post-stroke

(Singer, Nishihara, & Mochizuki, 2016). The commonalities between the current and previous

studies suggest that improvements within the first five months of recovery in overall balance

and unipedal stance time corresponds and may be primarily attributed to inpatient and

outpatient rehabilitation (Wade et al., 1997) and/or spontaneous recovery (Nudo, 2013).

4.5.2 Asymmetry after TBI is evident

This study found that weight-bearing asymmetry exists after TBI, with the least

vertically-loaded foot bearing only 44% of total body weight. In comparison to that of HC (48%),

absolute stance load symmetry was lower in individuals with TBI at all time points, indicating a

subtle but substantial postural asymmetry. Importantly, absolute stance load symmetry is lower

in individuals post-stroke ranging from 40-45% (Mansfield et al., 2011; Mansfield et al., 2012;

Singer et al., 2013; Singer et al., 2016). Weight-bearing asymmetry may reduce postural stability

for dynamic balance control (de Kam, Kamphuis, Weerdesteyn, & Geurts, 2016). Though

previous studies have observed postural asymmetry after TBI prior to postural perturbations

(Newton, 1995), in static balance through a greater lateral COP position (Lehmann et al., 1990),

and during bimanual lifting (Arce, Katz, & Sugarman, 2004), the extent of asymmetry after TBI

has not been quantified. Postural asymmetry after TBI underscores the balance deficits that

exist in this population.

94

Cross-correlation analysis (Rxy(0)) revealed poor-to-good inter-limb synchrony. ML inter-

limb synchrony in individuals with TBI fell outside of the HC ranges, similar to observations in

individuals post-stroke (Mansfield et al., 2011; Mansfield et al., 2012; Singer et al., 2013). Arce

et al. (2004) found poor inter-limb vertical force amplitude coupling in the heels and fore-feet

during bimanual load lifting after TBI. The authors suggested that because of an absence of

simultaneous weight shifting from the heel to forefoot and back in both limbs in individuals

with TBI, postural gain control in vertical force is impaired after TBI (Arce et al., 2004). Although

the mean AP inter-limb synchrony fell outside the HC 95% confidence interval, the range of AP

inter-limb synchrony (-0.024 – 0.99) from the TBI sample group was greater, demonstrating

that some TBI participants fall outside the normal range of HC (0.54 – 0.98). Poor inter-limb

synchrony may be related to motor impairment and increased falls, as shown in post-stroke

adults (Mansfield et al., 2012).

The majority of the analyses characterizing postural asymmetry in bipedal conditions did

not reveal unilateral weaknesses or impairments. In contrast, unipedal stance time uniquely

revealed a significant leg effect in TBI participants, indicative of a weaker or impaired right limb

over the left limb. Additionally, increased unipedal stance RMS in the right leg suggests that the

weaker and impaired right limb in TBI participants produces greater postural sway. Initial

dynamic force control within the first five seconds of unipedal stance increases force variability

(Jonsson et al., 2004), and may have contributed to the reduction in unipedal stance time in TBI

participants. We suggest that future studies analyze the dynamic control from bipedal to

unipedal stance to understand postural steadiness after TBI.

95

4.5.3 Study Limitations

Although the present study identified postural asymmetry and an absence of

improvements across recovery after TBI, there are limitations. Sample size was significantly

reduced as a result of the inclusion/exclusion criteria and future studies will benefit from a

greater sample size. Changes in body weight may have influenced the quantitative balance

measures (Hue et al., 2007). However, since premorbid weight was unknown it is unclear

whether weight gain was a return to premorbid weight or related to inactivity post-injury. In

parallel with Christensen et al. (2008), on a group level, both the TMT A and SDMT oral

demonstrated improvements across recovery, indicative of recovery of attention and executive

function, respectively. In contrast, the TMT B did not show a significant change across time,

indicating that task switching did not improve on a group level. These findings may represent a

dissociation between cognitive and balance measures. Alternatively, the absence of

improvement in both cognitive and balance measures reflects heterogeneity in the TBI cohort.

The within group variability also reflects the challenges in examining brain-behaviour

relationships in moderate-to-severe TBI, given the frequent nature of diffuse brain injury.

Different physiological processes that govern a specific behaviour may or may not be

differentially affected by the injury. Thus, while it is possible that balance measures are

influenced by cognitive impairment, analysis of interactions between these factors was beyond

the scope of this study.

96

4.6 Conclusions

The current study is the first to use posturographic measures to examine balance

control and postural asymmetry longitudinally out to one year post-TBI. The observation of

poor ML net COP RMS and ML inter-limb synchrony, absolute stance load asymmetry, and

increased AP COP RMS indicate that poor AP stability and ML control may contribute to the

majority of balance impairments after TBI.

97

4.7 References



doi:10.1016/j.humov.2003.12.002




Brauer, S. G., Broome, A., Stone, C., Clewett, S., & Herzig, P. (2004). Simplest tasks have greatest

dual task interference with balance in brain injured adults. Human Movement Science,

23(3-4), 489-502.




Christensen, B. K., Colella, B., Inness, E., Hebert, D., Monette, G., Bayley, M., & Green, R. E.

(2008). Recovery of cognitive function after traumatic brain injury: a multilevel modeling

analysis of Canadian outcomes. Archives of Physical Medicine and Rehabilitation, 89(12

Suppl), S3-15.




doi:10.1097/TA.0b013e3181715d66







1332. doi:10.1089/neu.2014.3787




Howe, J., Inness, E. L., & Wright, V. (2011). The Community Balance & Mobility Scale. The Center

for Outcome Measurement in Brain Injury. Retrieved from

http://www.tbims.org/combi/cbm ( accessed September 1, 2016 ).

98

Hue, O., Simoneau, M., Marcotte, J., Berrigan, F., Dore, J., Marceau, P., . . . Teasdale, N. (2007).

Body weight is a strong predictor of postural stability. Gait and Posture, 26(1), 32-38.

doi:10.1016/j.gaitpost.2006.07.005

Hurvitz, E. A., Richardson, J. K., Werner, R. A., Ruhl, A. M., & Dixon, M. R. (2000). Unipedal

Stance Testing as an Indicator of Fall Risk Among Older Outpatients. Archives of Physical

Medicine and Rehabilitation, 81(5), 587-591. doi:10.1053/mr.2ooO.6293

Inness, E. L., Howe, J. A., Niechwiej-Szwedo, E., Jaglal, S. B., McIlroy, W. E., & Verrier, M. C.

(2011). Measuring Balance and Mobility after Traumatic Brain Injury: Validation of the

Community Balance and Mobility Scale (CB&M). Physiotherapy Canada, 63(2), 199-208.

doi:10.3138/ptc.2009-45

Jonsson, E., Seiger, A., & Hirschfeld, H. (2004). One-leg stance in healthy young and elderly

adults: a measure of postural steadiness? Clinical Biomechanics, 19(7), 688-694.

doi:10.1016/j.clinbiomech.2004.04.002

Keenan, M. A., Perry, J., & Jordan, C. (1984). Factors affecting balance and ambulation following

stroke. Clinical Orthopaedics and Related Research, 182, 165-171.




962.

Lezak, M. D. (1995). Neuropsychological Assessment–3rd edition. New York: Oxford University

Press.

Maki, B. E., Holliday, P. J., & Topper, A. K. (1994). A prospective study of postural balance and

risk of falling in an ambulatory and independent elderly population. Journal of

Gerontology, 49(2), M72-84.

Mancini, M., & Horak, F. B. (2010). The relevance of clinical balance assesments tools to

differentiate balance deficits. European Journal of Physical and Rehabilitation Medicine,

46(2), 239-248.

Manor, B., Hu, K., Zhao, P., Selim, M., Alsop, D., Novak, P., . . . Novak, V. (2010). Altered control

of postural sway following cerebral infarction: a cross-sectional analysis. Neurology,

74(6), 458-464. doi:10.1212/WNL.0b013e3181cef647






99


doi:10.1177/1545968311429688

McFadyen, B. J., Swaine, B., Dumas, D., & Durand, A. (2003). Residual effects of a traumatic

brain injury on locomotor capacity. A first study of spatiotemporal patterns during

unobstructed and obstructed walking. Journal of Head Trauma Rehabilitation, 18(6),

512-525.

McIlroy, W. E., & Maki, B. E. (1997). Preferred placement of the feet during quiet stance:

development of a standardized foot placement for balance testing. Clinical

Biomechanics, 12(1), 66-70.

Mitchell, S. L., Collins, J. J., De Luca, C. J., Burrows, A., & Lipsitz, L. A. (1995). Open-loop and

closed-loop postural control mechanisms in Parkinson's disease: increased mediolateral

activity during quiet standing. Neuroscience Letters, 197(2), 133-136.



doi:10.1152/jn.01322.2004



Nudo, R. J. (2013). Recovery after brain injury: mechanisms and principles. Frontiers in Human

Neuroscience, 7, 887. doi:10.3389/fnhum.2013.00887

Perianez, J. A., Rios-Lago, M., Rodriguez-Sanchez, J. M., Adrover-Roig, D., Sanchez-Cubillo, I.,

Crespo-Facorro, B., . . . Barcelo, F. (2007). Trail Making Test in traumatic brain injury,

schizophrenia, and normal ageing: sample comparisons and normative data. Archives of

Clinical Neuropsychology, 22(4), 433-447. doi:10.1016/j.acn.2007.01.022







Reitan, R. M., & Wolfson, D. (1985). The Halstead–Reitan Neuropsycholgical Test Battery:

Therapy and clinical interpretation. Tucson, AZ: Neuropsychological Press.





100




786-795.

Singer, J. C., Nishihara, K., & Mochizuki, G. (2016). Does Poststroke Lower-Limb Spasticity

Influence the Recovery of Standing Balance Control? A 2-Year Multilevel Growth Model.

Neurorehabilitation and Neural Repair, 30(7), 626-634. doi:10.1177/1545968315613862

Smith, A. (1982). The Symbol Digit Modalities Test (SDMT) manual. Los Angeles: Western

Publication Services.

Springer, B. A., Marin, R., Cyhan, T., Roberts, H., & Gill, N. W. (2007). Normative values for the

unipedal stance test with eyes open and closed. Journal of Geriatric Physical Therapy,

30(1), 8-15.

Till, C., Colella, B., Verwegen, J., & Green, R. E. (2008). Postrecovery cognitive decline in adults

with traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 89(12

Suppl), S25 - 34.






44(7), 975-982. doi:10.1682/jrrd.2006.12.0158

Wechsler, D. (2001). Wechsler Test of Adult Reading (WTAR). San Antonio: Psychological Corp.

Williams, G., & Morris, M. E. (2011). Tests of static balance do not predict mobility performance

following traumatic brain injury. Physiotherapy Canada, 63(1), 58-64.

doi:10.3138/ptc.2009-53


Posture, 3(4), 193-214.







101

Chapter 5 : Spectral analysis of centre of pressure identifies

altered balance control in individuals with moderate-severe

traumatic brain injury

Habib Perez, O., G. Green, R. E. & Mochizuki, G. Spectral analysis of centre of pressure identifies

altered balance control in individuals with moderate-severe traumatic brain injury. Under

review with Disability and Rehabilitation

102

5.1 Abstract

Purpose: To identify impairments and recovery of anticipatory and reactive balance control

after TBI through spectral analyses of static balance tasks and to characterize the contributions

of each limb to balance control.

Methods: Balance was assessed with force platforms at 2, 5 and 12 months post-injury in eyes

open and closed. Secondary analysis of longitudinal balance data from moderate-to-severe TBI

(n = 31) was conducted. Single-visit data from age-matched controls (n = 22) were collected for

visual comparison. Net and individual limb center of pressure (COP) measures and inter-limb

COP coherence were calculated in low (≤ 0.4 Hz) and high (≥ 0.4 Hz) frequencies in the

anteroposterior and mediolateral directions.

Results: Removing vision increased net COP spectral power in low and high frequencies.

Individuals with TBI demonstrated recovery in high frequencies in net COP in the mediolateral

direction over time. Inter-limb coherence in the anteroposterior and mediolateral directions

increased (recovered) over time in high frequencies. Weight-bearing asymmetry was visible in

high frequencies in the anteroposterior and mediolateral direction post-TBI.

Conclusions: Increased amplitude of low and high frequencies suggests that individuals with TBI

have impaired anticipatory and reactive balance mechanisms, which may be driven by weight-

bearing asymmetries and place individuals post-TBI at increased risk of falls.

103

5.2 Introduction

Balance impairments after traumatic brain injury (TBI) are common (Walker & Pickett,

2007). In this population, balance impairments lead to increased instability, which elevates fall

risk (Horak, 2006). One method of measuring balance impairment is to assess the systems and

mechanisms that affect balance control; however, only a few studies have analyzed balance

after TBI using quantitative measures that provide characterization of specific impairments

(Basford et al., 2003; Geurts, Ribbers, Knoop, & van Limbeek, 1996; Habib Perez, Green, &

Mochizuki, 2015; Lehmann et al., 1990; Pickett, Radfar-Baublitz, McDonald, Walker, & Cifu,

2007; Wade, Canning, Fowler, Felmingham, & Baguley, 1997). These studies have identified

improvements in balance after injury and have determined that simple balance tasks (i.e.,

standing quietly with eyes open) may not be sufficiently sensitive to reliably detect balance

impairments. Furthermore, while there is evidence indicating that dynamic balance control,

which includes anticipatory and reactive strategies for control, is impaired after injury (Arce,

Katz, & Sugarman, 2004; Ghajar & Ivry, 2008; Newton, 1995), evidence is sparse. Because

anticipatory and reactive mechanisms facilitate preparation and reaction to postural instability,

it is important to advance understanding of changes that occur in the components of balance

that increase risk of falls.

Most studies evaluating balance after TBI focus on clinical balance assessments and

spatial measures of balance (Arce et al., 2004; Basford et al., 2003; Geurts et al., 1996; Habib

Perez et al., 2015; Habib Perez, Green, & Mochizuki, Under Review; Lehmann et al., 1990;

Newton, 1995; Pickett et al., 2007; Wade et al., 1997; Walker & Pickett, 2007). However,

spectral analyses of balance measures have additional utility in understanding the underlying

104

mechanisms of balance control and may help to better characterize and predict recovery in the

TBI patients. Postural sway is generally characterized as consisting of low and high frequencies

(Zatsiorsky & Duarte, 1999). Low frequency components (≤ 0.4 Hz) are considered to be

exploratory (Carpenter, Murnaghan, & Inglis, 2010; Zatsiorsky & Duarte, 1999), slow migrations

of the COP to a reference point from the central nervous system (Latash, Ferreira, Wieczorek, &

Duarte, 2003), or error corrections to equilibrium (Kiemel, Oie, & Jeka, 2006) useful for

feedforward (in contrast to feedback) control (Gatev, Thomas, Kepple, & Hallett, 1999).

Feedforward control of movement is linked to predictive models of behaviour (Massion, 1992;

Wolpert & Kawato, 1998) and plays a role in anticipatory control mechanisms. Higher frequency

components (≥ 0.4 Hz) are considered to reflect corrective responses to temporary instability

(Singer & Mochizuki, 2015), representing reactive control mechanisms (Schinkel-Ivy, Singer,

Inness, & Mansfield, 2016). Frequency decomposition of the centre of pressure (COP) have

characterized balance in individuals post-stroke (Paillex & So, 2003; Schinkel-Ivy et al., 2016;

Singer & Mochizuki, 2015; Yanohara et al., 2014) and other neurologic populations (Kanekar,

Lee, & Aruin, 2014) and may provide insight into the impairments in anticipatory and reactive

control mechanisms that may place individuals with TBI at an increased risk of falls.

Typically, balance assessments involve quantifying the overall (net) contributions of

both lower limbs to control. While appropriate in conditions where no asymmetries exist,

individual foot contributions may provide additional information about the underlying control

features in asymmetric conditions (Winter, Prince, Stergiou, & Powell, 1993). Importantly,

previous studies identify persistent unilateral motor weakness and anecdotally report

asymmetrical stance after TBI (Choi et al., 2012; Newton, 1995). In addition, recent work from

105

our laboratory demonstrates that balance asymmetry and mediolateral inter-limb synchrony in

TBI closely resembles that of individuals with stroke (Mansfield, Danells, Inness, Mochizuki, &

McIlroy, 2011; Mansfield, Mochizuki, Inness, & McIlroy, 2012; Singer, Mansfield, Danells,

McIlroy, & Mochizuki, 2013), and may be one factor contributing to increased balance

instability after TBI (Habib Perez et al., Under Review). Balance asymmetry after TBI reflects

underlying unilateral motor weakness, and may suggest that the limb with the higher weight-

bearing is driving balance control mechanisms. In the spectral domain, balance asymmetry in

individuals with neurological impairments has been identified by quantifying the extent of COP

coherence, a measurement of correlation of frequency spectra between two time-varying

signals, and is a representation of inter-limb coordination (Myklebust et al., 2009). Examining

the correlation of the COP frequencies of each limb can identify disorders of the postural

system relating to anticipatory and reactive control mechanisms and may be useful in

understanding inter-limb coordination in this population.

The purpose of this study was to investigate whether anticipatory and reactive balance

impairments exist after TBI by carrying out spectral analyses (power spectral density and

coherence) of static balance tasks. To our knowledge, no study has examined how individual

foot contributions affect anticipatory and reactive control mechanisms in quiet standing across

the recovery of TBI. It was hypothesized that spectral measures of net COP would improve

across time in individuals with TBI. Omission of vision was used to increase postural challenge,

thus it was hypothesized that performing tasks with the eyes closed would increase the

amplitude of spectral measures of balance. Despite previous findings where inter-limb

synchrony did not change across time (Habib Perez et al., Under Review), as a result of

106

increased sensitivity of spectral measures, it was hypothesized that improvements across time

in inter-limb coherence in individuals with TBI would be detectable. Furthermore, it was

hypothesized that the individual limb applying the most vertical force (i.e., highest weight-

bearing) would demonstrate increased COP power spectral density and decrease across

recovery. Complementary spatial measures of control were analyzed to test the hypothesis that

these would also improve over recovery in individuals with TBI and to determine whether

spatial measures of control across time were affected by postural challenge.

5.3 Methods

5.3.1 Participants

This study involved a secondary analysis of balance data from a larger study with

individuals with clinically-confirmed TBI (The Toronto Rehab Traumatic Brain Injury Recovery

Study). Details of the larger study and its inclusion and exclusion criteria have been reported

previously (Habib Perez et al., Under Review). Additional eligibility criteria for the current study

were completion of three instrumented balance assessments on dual force plates, and the

absence of lower extremity orthopedic injuries.

Seventy-three participants with TBI completed instrumented balance assessments on

dual force plates. Of the 73 TBI participants, 37 participants were ineligible because of missing

data within an assessment, and five were ineligible due to lower extremity orthopedic injuries.

Therefore, 31 participants with TBI were included in the analysis. A group of aged-matched

healthy control (HC) participants were recruited for visual comparison and assessed on one

107

occasion. Demographic characteristics of the sample are reported in Table 5-1. All participants

(or a substitute decision maker) provided informed consent. The study was approved by the

Research Ethics Board at the Toronto Rehabilitation Institute.

5.3.2 Data Collection and Procedure

Clinical measures of balance (Community Balance and Mobility (CB&M) scale)(Howe,

Inness, & Wright, 2011) and force plate balance data were undertaken with participants with

TBI at approximately 2-months (T1), 5-months (T2), and 12-months (T3) post injury. Data were

collected using dual force plates (AMTI, Watertown Massachusetts), positioned with the y-axes

in parallel, and separated with a minimal distance (1 mm). Participants stood with each foot on

a separate force plate in two stance conditions: eyes open (EO) and eyes closed (EC). Foot

position was standardized (McIlroy & Maki, 1997). Participants were instructed to stand quietly

for 50 seconds looking straight ahead. Ground-reaction forces and moments from each plate

were sampled at 50 Hz for all TBI participant data. HC participants were collected after all TBI

data was collected and used for visual comparison to TBI participants.

5.3.3 Data Analysis

Ground-reaction forces and moments were low-pass filtered using a 4th ordered dual-

pass Butterworth filter with a 10 Hz cut off frequency prior to processing. The net AP and ML

COP and the respective time series were calculated for all participants. For the spectral analysis,

individual limb and net displacements of the COP were processed by fast Fourier transform. The

108

power spectrum density (PSD) estimation was obtained in Matlab using Welch method with

non-overlapping Hanning windows of 5-second data segments. The power spectrum was

divided into two frequency bands: low frequency (0-0.4 Hz) and high frequency (0.4-3 Hz). The

integral for the low and high frequency bands was computed using the trapezoidal method.

Individual limb COP PSD was calculated and grouped based on the leg with the lowest weight-

bearing (LWB) and highest weight-bearing (HWB) limb for comparison in the AP and ML

direction. Net AP and ML COP PSD were also calculated. Inter-limb coherence estimates were

obtained from non-overlapping Hanning windows of 5-second data segments, with a frequency

resolution of 0.04 Hz. Coherence analysis included frequencies in the 0-5 Hz range, with

frequencies below 0.4 Hz and above 0.4 Hz representing low and high frequency bands,

respectively. In the AP and ML direction, the coherence for each frequency bin in the low and

high frequency bands were averaged accordingly if they exceeded the statistical confidence

limit (see below). Inter-limb coherence is expressed as a number between 0 and 1, with 1

indicating a perfect correlation and 0 indicating an absence of a correlation. For each

participant, the average coherence in low and high frequencies were transformed using Fisher

transformation (Mima & Hallett, 1999). Root mean square (RMS) of the net AP and ML COP was

used as the spatial measure of balance control.

5.3.4 Statistical Analysis

Statistical analyses were performed using IBM SPSS 23.0. To test the hypotheses that

spectral and spatial measures would improve across recovery for TBI participants and vision

affects balance stability and, a series of 2 × 3 analyses of variance (ANOVA), with Condition and

109

Time as factors, were conducted for AP and ML PSD, AP and ML inter-limb coherence, and net

AP and ML COP RMS. Coherence estimates for the low and high frequencies bands were

considered statistically significant when they exceeded the confidence limit of 0.283 (n =10) at

α = 0.05 (Rosenberg, Amjad, Breeze, Brinllinger, & Halliday, 1989). To test the hypothesis that

the higher weight-bearing limb would demonstrate increased COP power and decrease across

recovery, a 2 × 2 × 3 ANOVA was conducted with Foot (LWB, HWB), Condition and Time as

factors. RMS and PSD measures were log transformed to meet assumptions of normality and

inter-limb coherence was Fisher transformed. In cases where sphericity was not met,

Greenhouse-Geisser values were reported. Statistical significance was set at p<0.05.

5.4 Results

Thirty-one participants with TBI completed balance assessments across three time

points post-injury and data from 22 HC participants were collected prospectively. TBI

participants were assessed on average at 59.8 (±29.0), 150 (±28.9), and 412 (±69.3) days post

injury at T1, T2 and T3. There was no significant difference in age between groups (t(49.8)= 1.3,

p>0.05, d = 0.42; Table 5-1) and no significant difference in mass between the HC group and TBI

group at T1 (t(49)=1.3, p>0.05, d = 0.36). There was a significant mass increase of approximately

8 kg in the TBI participants across the three time points (F1.4,38.9 = 24.1, p<0.001, partial η2 = .46),

with Bonferroni post-hoc tests identifying significant differences from T1 to T2 (p<0.001) and T1

to T3 (p<0.001). Clinical balance measures (CB&M) improved over time (F1.4,38.6 = 26.4, p<0.001,

partial η2 = .49), specifically between T1 and T2 (p<0.001) and T1 and T3 (p<0.001). There was

110

no statistically significant difference between T2 and T3 (p>0.05). The HC group completed

more years of education in comparison to TBI participants (t(50) = -2.6, p=0.01, d =0.72).

Table 5-1 – Mean and standard deviation (SD) participant demographics of traumatic brain

injury (TBI) and healthy control (HC) participants. TBI severity was characterized by the lowest

post-traumatic amnesia (PTA) scale and the number of TBI participants in each category is

described below.

TBI HC

Sample size 31 (24 M/7 F) 22 (10 M/12 F)

Age 41.9 (16.9) 36.6 (13.8)

Mass (kg)

T1 73.9 (10.7) 69.6 (12.8)

T2 79.18 (12.3)* --

T3 81.9 (13.9)* --

Severity (PTA)

Moderate 1 --

Severe 6 --

Very Severe 16 --

Extremely Severe 4 --

CB&M Scale

T1 65.0 (18.1) --

T2 78.0 (15.1) * --

T3 78.7 (14.1) * --

Years of education 14.60 (3.27) 17.0 (3.4)

*Denotes significant differences within TBI participants at p<0.05 from T1 value. Note that PTA

severity was obtained from 27 of the 31 TBI participants.

111

5.4.1 Spectral Measures of Balance

5.4.1.1 Net COP Power Spectral Density

Omitting vision increased net COP PSD in the AP direction at low frequencies and in the

AP and ML direction at high frequencies. Additionally, changes across recovery were found for

high frequencies in the ML direction. The 2 × 3 ANOVA demonstrated significant effects of

Condition for low frequencies in the AP (F1,27=20.3, p<0.05, partial η2 = .429) direction (Figure 5-

1A). Significant effects of Condition were also identified for high frequencies in the AP

(F1,27=88.9, p<0.05, partial η2 = .77) and ML (F1,27=31.5, p<0.05, partial η2 = .54) direction, with

greater net COP power in EC than EO. A significant effect of Time was observed for ML net COP

PSD in high frequencies (F2,54=7.1, p<0.05, partial η2 = .21). Bonferroni-adjusted post-hoc tests

showed ML net COP PSD in high frequencies was significantly greater in T1 when compared to

T2 (p=0.03) and T3 (p=0.01; Figure 5-1D). No other significant effects of Time were found. In

low frequencies, the mean net COP PSD of TBI participants fell outside the top 95% CI of HC. In

high frequencies, this only occurs in EC. Figure 5-2 demonstrates the mean PSD for the net COP

for HC and a sample of four TBI participants across recovery.

112

Figure 5-1 – Net centre of pressure mean integral of the power spectral density (PSD) in low (A)

and high (B) frequencies in the anteroposterior (AP) direction, and the low (C) and high (D)

frequencies in the mediolateral (ML) direction. Group means and standard error for TBI

participants across recovery are displayed for eyes open (EO; black) and eyes closed (EC; grey).

The solid and dashed lines represent mean and 95% confidence intervals for the net COP mean

integral of the PSD in healthy controls in each condition, respectively.

113

Figure 5-2 – Mean and standard deviation for the net centre of pressure (COP) power spectral

density (PSD) for healthy controls in the anteroposterior (AP) and mediolateral (ML) direction

for eyes open (EO) and eyes closed (EC) conditions (top row). Individual data net COP PSD from

a sample of four TBI participants across recovery [T1 (black), T2 (grey), and T3 (light grey)] in the

AP and ML direction in both EO and EC conditions (row 2 to 5).

5.4.2 Inter-limb Coordination

5.4.2.1 Inter-limb Coherence

Omitting vision increased inter-limb coherence in low and high frequencies in the AP

direction and in high frequencies in the ML direction. Additionally, inter-limb coherence

114

increased across recovery in high frequencies in both the AP and ML direction (see Figure 5-3).

A main effect of Condition was found in the 2 × 3 ANOVA for AP coherence for the low

(F1,27=9.0, p<0.05, partial η2 = .25) and high (F1,27=31.2, p<0.05, partial η2 = .54) frequencies, and

for high frequencies in the ML direction (F1,25=31.9, p<0.05, partial η2 = .56). In addition, high

frequencies demonstrated significant effects of Time for AP coherence (F2,54=5.8, p<0.05, partial

η2 = .17) and ML coherence (F2,50=3.6, p<0.05, partial η2 = .13). Bonferroni-adjusted post-hoc

tests identified that inter-limb coherence in high frequencies increased between T1 and T3 in

the AP (p=0.009) and ML (p=0.03) direction. When visually comparing AP and ML inter-limb

coherence to HC, TBI participants fell within the 95% CI of the HC data.

115

Figure 5-3 – Mean and standard error (SE) anteroposterior (AP) (A) and mediolateral ML (B)

inter-limb coherence in low frequencies and AP (C) and ML (D) inter-limb coherence in for high

frequencies. Group means ±SE for TBI participants are displayed for eyes open (EO; black) and

eyes closed (EC; grey) conditions across recovery. The solid and dashed lines represent mean

and 95% confidence intervals for the mean inter-limb coherence in healthy controls in each

condition, respectively. Pound (#) in C and D denotes significant difference between T1 and T3

in TBI participants.

116

5.4.2.2 Individual Limb COP

Amplitude of the PSD in the individual limb increased in low and high frequencies with

the omission of vision in the AP and ML direction (see Figure 5-4). The high frequency range in

the ML direction demonstrated a trend towards increased PSD amplitude for the foot with

higher weight-bearing (HWB). Specifically, the 2 × 2 × 3 ANOVA demonstrated a significant main

effect of Condition with EC producing greater power in low frequencies in the AP direction

(F1,27=21.0, p<0.05, partial η2 = .44), and greater power in high frequencies in the AP (F1,27=87.9,

p<0.05, partial η2 = .77) and ML (F1,27=66.1, p<0.05, partial η2 = .71) direction. Low frequency

PSD in the ML direction showed a trend (p=0.08) towards a higher amplitude with the EC.

Differences between LWB and HWB approached statistical significance in high frequencies in

the ML direction (p=0.075, see Figure 5-4D). A Time × Condition interaction approached

statistical significance for high frequencies in the ML direction (p=0.066). With the exception of

a few incidences, PSD in low and high frequencies in both the AP and ML direction in TBI

participants primarily fell outside the 95% CI of HC.

117

Figure 5-4 – Individual limb centre of pressure (COP) mean integral of the power spectral

density (PSD) in low (A) and high (B) frequencies in the anteroposterior (AP) direction, and the

low (C) and high (D) frequencies in the mediolateral (ML) direction. Group means and standard

error for TBI participants across recovery are displayed across time. Lower weight-bearing

(LWB; solid) and higher weight-bearing (HWB; open) limb are represented in eyes open (EO;

black) and eyes closed (EC; grey). Group means and 95% confidence intervals (depicted with

error bars) are displayed for healthy control (HC) data.

118

5.4.3 Spatial Measures of Balance

Omitting vision increased AP RMS. The 2 × 3 ANOVA for AP RMS demonstrated

significant effects of Condition (F1,27=50.3, p<0.05, partial η2 = .65), with EC producing a greater

AP RMS in comparison to EO. No significant main effect of Time or Condition x Time interaction

was identified for AP RMS. No significant main effect of Condition or Condition x Time

interaction was found for ML RMS. When comparing RMS measures of balance to HC, AP and

ML RMS in TBI participants fell outside the 95% CI of HC participants in EO and EC across time.

(see Figures 5-5A and B).

119

Figure 5-5 – Anteroposterior (AP) (A) and mediolateral (ML) (B) centre of pressure root mean


open (EO; black) and eyes closed (EC; grey) conditions. The solid and dashed lines represent

mean and 95% confidence intervals for COP RMS in healthy controls in each condition,

respectively.

120

5.5 Discussion

This study examined whether anticipatory and reactive control mechanisms are

impaired after TBI and characterized contributions of each limb to these control mechanisms.

Spectral and spatial analyses of the COP during quiet standing with eyes open and closed

measured changes in balance at three time points during recovery of TBI. Consistent with the

hypotheses, removing vision increased net COP measures of balance, inter-limb coherence, and

individual foot COP spectral measures. Recovery changes after TBI were identified in high

frequencies in the ML direction. Differences were visible in the PSD in high frequencies in the

ML direction between the higher and lower weight-bearing limb. Additionally, inter-limb

coherence in the high frequency range increased in both the AP and ML direction over the

course of recovery; however, mean values fell within the boundaries of healthy control

normative data.

5.5.1 Anticipatory and Reactive Balance Control Mechanisms are Impaired after

TBI

This study identified greater low frequency power of net COP in the AP direction and

greater high frequency power in both the AP and ML direction for TBI participants. The only

improvements observed over the course of recovery were in the high frequencies of the net ML

COP. This measure of recovery was primarily driven by performance in the EO condition, as TBI

participants produced greater power in the EC condition at T2 and T3 compared to HC. This

suggests that overall control of anticipatory and reactive balance mechanisms may be impaired

after TBI and do not improve unless visual information is present.

121

The effects of removing vision on balance control have been previously studied

(Shumway-Cook & Horak, 1986; Wade & Jones, 1997). This manipulation increases sway,

especially in the AP direction because of the direction in which visual or optic flow information

is processed (Bertenthal, Rose, & Bai, 1997; Dijkstra, Schoner, Giese, & Gielen, 1994; Moraes,

de Freitas, Razuk, & Barela, 2016). Distinct frequency intervals have been linked to sensory

systems used by postural control systems, suggesting that the visual, vestibular, and

somatosensory/proprioceptive system are associated with low (<0.3 Hz), middle (0.3–1 Hz) and

high (> 1 Hz) frequencies of sway, respectively (Bizid et al., 2009; Kanekar et al., 2014; Nagy et

al., 2007; Nagy et al., 2004). Identifying anticipatory and reactive balance control mechanisms

using low and high frequencies in the present study does not completely align to previous

reports. Balance control, both anticipatory and reactive, requires the use of the multiple

sensory systems (Mohapatra & Aruin, 2013; Mohapatra, Krishnan, & Aruin, 2012; Winter, 1995)

and it may be inaccurate to suggest that individual frequency intervals are solely used by the

associated sensory system (Palmieri, Ingersoll, Stone, & Krause, 2002). Some studies suggest

that an impaired vestibular system contributes to balance deficits post-TBI (Basford et al., 2003;

Walker & Pickett, 2007); however, disruption of vestibular system function has been shown to

affect postural sway frequencies that span the 0.25–5 Hz range (Fitzpatrick, Burke, & Gandevia,

1996; Forbes, Siegmund, Schouten, & Blouin, 2014). Given that the majority of observed

differences between HC and TBI participants were found in EC, it is possible that both the

somatosensory and vestibular system were impaired in our sample and individuals post-injury

were thus further reliant on visual information in both low and high frequencies. It may also be

possible that brain regions that integrate multiple sensory modalities (Brang, Taich, Hillyard,

122

Grabowecky, & Ramachandran, 2013; Ghazanfar & Schroeder, 2006) were injured, thus

impacting the ability to accurately process incoming sensory information.

The present findings advance understanding of the characteristics of anticipatory and

reactive balance control after TBI. A previous study demonstrated age-appropriate sway

amplitudes and slightly higher latencies in unpredictable feet-in-place reactions after TBI

(Newton, 1995), while poor postural gain control has been found in self-initiated bimanual

load-lifting post TBI (Arce et al., 2004). It has been postulated that impaired anticipatory neural

circuits lead to delayed reaction times and increased variability in movement control (Di Russo,

Incoccia, Formisano, Sabatini, & Zoccolotti, 2005; Ghajar & Ivry, 2008; Segalowitz, Dywan, &

Unsal, 1997). Preparation for balance instability and the latencies and amplitude of postural

reactions are imperative for dynamic balance control (Horak, Diener, & Nashner, 1989; Maki &

McIlroy, 1997; McIlroy & Maki, 1999; Mochizuki, Boe, Marlin, & McIlRoy, 2010). The current

results corroborate findings of previous studies and suggest that anticipatory and reactive

balance mechanisms were impaired. This may place these individuals at an increased risk of

falls. Further studies are needed to understand the components of dynamic balance control to

address balance specific impairments post-TBI.

5.5.2 Inter-limb Coordination after TBI

Inter-limb coordination was analyzed using coherence analysis to identify frequencies of

the COP under each foot that were correlated (Myklebust et al., 2009). Although inter-limb

coherence improved across time points, group means fell within the 95% CI of HC. Inter-limb

123

coherence in the AP and ML direction increased at high frequencies, suggesting that the

relationship between limbs in frequencies that correspond to reactive balance control increases

out to one year after injury. Previous work from our laboratory has demonstrated that inter-

limb coordination requires decoupling in preparation for unpredictable perturbations in healthy

adults (Habib Perez, Singer, & Mochizuki, 2016). Although inter-limb coherence in TBI

participants did not differ from HC in either low or high frequencies, the increased coherence at

high frequencies suggests an inappropriately high level of coupling between limbs, with respect

to reactive balance control mechanisms. Poor inter-limb vertical force amplitude coupling

during bimanual load lifting has been found after TBI (Arce et al., 2004). Moreover, poor inter-

limb synchrony in the ML direction has been found post-TBI, resembling the levels of synchrony

observed in stroke patients (Habib Perez et al., Under Review). Inter-limb coherence is a

relatively novel analysis for postural asymmetry and has shown sensitivity for identifying

differences between healthy adults and other neurologically-impaired populations. For

example, individuals with Huntington’s Disease produced reduced inter-limb coherence

compared to healthy young and older adults (Myklebust et al., 2009). The absence of visual

differences found in the present study between TBI and HC may be attributable to differences

in the way in which ‘significant’ coherence was identified. The current study applied a statistical

confidence limit to coherence, similar to what has been used in other physiological studies

(Rosenberg et al., 1989), while Myklebust et al. (2009) did not. The use of confidence limits

eliminated possible low coherence values in TBI participants.

Inter-limb coordination in the current study was also analyzed using weight-bearing

asymmetry categorization. Asymmetry in high frequencies approached statistical significance in

124

the ML direction, suggesting that weight-bearing asymmetry may affect reactive balance

mechanisms. Weight-bearing asymmetry reduces postural stability for dynamic balance control

(de Kam, Kamphuis, Weerdesteyn, & Geurts, 2016). Interestingly, the limb with the higher level

of weight-bearing in the present study displayed means much greater than the 95% CI of HC in

both EO and EC at high frequencies. Individuals post-stroke that demonstrate a reduced ability

to recover from balance also display greater COP amplitude of high frequencies in standing

balance (Schinkel-Ivy et al., 2016). The present finding suggests that the imbalance in weight-

bearing symmetry may be driving the amplitude of power of the higher weight-bearing limb in

frequencies that are important for reactive balance control. This may subsequently place

individuals with TBI at a higher risk of falls and renders them less likely to recover from balance

perturbations.

5.5.3 The Role of the Postural Challenge

The current study used an EC condition to increase the postural challenge. In line with

our hypothesis, spectral measures demonstrated significant main effects of Condition with

higher COP power in EC within the TBI group. Additionally, spatial measures of balance (root

mean square of the COP) in the current study were higher in EC in comparison to the EO

condition, but only in the AP direction. Thus, the differences were most notable in all measures

when vision was omitted.

Other studies have demonstrated that performing concurrent cognitive-balance tasks

reveal balance impairments after TBI (McCulloch, Buxton, Hackney, & Lowers, 2010). This

125

suggests that a postural challenge may be more sensitive at detecting balance impairments in

individuals with TBI. As an example, adding a mental arithmetic task increases overall frequency

of the COP in the AP direction (Vuillerme & Vincent, 2006). The findings of the current study

revealed that spectral measures demonstrated that TBI participants fall outside the 95% CI of

healthy controls more frequently. Thus, increasing postural challenge through eyes closed or via

a cognitive task may provide insight into the balance impairments after TBI.

5.6 Conclusion

This study found that individuals with moderate-to-severe TBI produced higher power in

the net COP when compared to normative data of healthy controls, demonstrating that

anticipatory and reactive balance mechanisms may be impaired after TBI. Additionally, high

frequencies, which represent reactive control mechanisms, were largely affected in inter-limb

coordination measures. Lastly, increasing the postural challenge frequently identified visual

differences between TBI and HC. Inferring anticipatory and reactive balance control

mechanisms through spectral analysis of COP in static balance tasks advances knowledge of

balance control after TBI.

126

5.7 References



doi:10.1016/j.humov.2003.12.002




Bertenthal, B. I., Rose, J. L., & Bai, D. L. (1997). Perception-action coupling in the development

of visual control of posture. Journal of Experimental Psychology: Human Perception and

Performance, 23(6), 1631-1643.

Bizid, R., Jully, J. L., Gonzalez, G., Francois, Y., Dupui, P., & Paillard, T. (2009). Effects of fatigue

induced by neuromuscular electrical stimulation on postural control. Journal of Science

and Medicine in Sport, 12(1), 60-66. doi:10.1016/j.jsams.2007.10.006

Brang, D., Taich, Z. J., Hillyard, S. A., Grabowecky, M., & Ramachandran, V. S. (2013). Parietal

connectivity mediates multisensory facilitation. Neuroimage, 78, 396-401.

doi:10.1016/j.neuroimage.2013.04.047

Carpenter, M. G., Murnaghan, C. D., & Inglis, J. T. (2010). Shifting the balance: evidence of an

exploratory role for postural sway. Neuroscience, 171(1), 196-204.








Di Russo, F., Incoccia, C., Formisano, R., Sabatini, U., & Zoccolotti, P. (2005). Abnormal motor

preparation in severe traumatic brain injury with good recovery. Journal of

Neurotruama, 22(2), 297-312.

Dijkstra, T. M., Schoner, G., Giese, M. A., & Gielen, C. C. (1994). Frequency dependence of the

action-perception cycle for postural control in a moving visual environment: relative

phase dynamics. Biological Cybernetics, 71(6), 489-501.

Fitzpatrick, R., Burke, D., & Gandevia, S. C. (1996). Loop gain of reflexes controlling human

standing measured with the use of postural and vestibular disturbances. Journal of

Neurophysiology, 76(6), 3994-4008.

127

Forbes, P. A., Siegmund, G. P., Schouten, A. C., & Blouin, J. S. (2014). Task, muscle and

frequency dependent vestibular control of posture. Frontiers in Integrative

Neuroscience, 8(94). doi:10.3389/fnint.2014.00094

Gatev, P., Thomas, S., Kepple, T., & Hallett, M. (1999). Feedforward ankle strategy of balance

during quiet stance in adults. Journal of Physiology, 514(3), 915-928.






doi:10.1177/1545968308315600

Ghazanfar, A. A., & Schroeder, C. E. (2006). Is neocortex essentially multisensory? Trends in

Cognitive Sciences, 10(6), 278-285. doi:10.1016/j.tics.2006.04.008

Habib Perez, O., Green, R. E., & Mochizuki, G. (2015, June 28 - July 2). Characterization of

stability control in early recovery from traumatic brain injury. Paper presented at the

International Society of Posture and Gait Research World Congress, Seville, Spain.

Habib Perez, O., Green, R. E., & Mochizuki, G. (Under Review). Characterization of balance

control after moderate-severe traumatic brain injury: A longitudinal recovery study.

Habib Perez, O., Singer, J. C., & Mochizuki, G. (2016). The effects of predictability on inter-limb

postural synchronization prior to bouts of postural instability. Gait and Posture, 46, 167-

172.

Horak, F. B. (2006). Postural orientation and equilibrium: what do we need to know about

neural control of balance to prevent falls? Age and Ageing, 35 Suppl 2, ii7-ii11.

doi:10.1093/ageing/afl077



Howe, J., Inness, E. L., & Wright, V. (2011). The Community Balance & Mobility Scale. The Center

for Outcome Measurement in Brain Injury. Retrieved from

http://www.tbims.org/combi/cbm ( accessed September 1, 2016 ).

Kanekar, N., Lee, Y. J., & Aruin, A. S. (2014). Frequency analysis approach to study balance

control in individuals with multiple sclerosis. Journal of Neuroscience Methods, 222, 91-

96. doi:10.1016/j.jneumeth.2013.10.020

Kiemel, T., Oie, K. S., & Jeka, J. J. (2006). Slow dynamics of postural sway are in the feedback

loop. Journal of Neurophysiology, 95(3), 1410-1418. doi:10.1152/jn.01144.2004

128

Latash, M. L., Ferreira, S. S., Wieczorek, S. A., & Duarte, M. (2003). Movement sway: changes in

postural sway during voluntary shifts of the center of pressure. Experimental Brain

Research, 150(3), 314-324. doi:10.1007/s00221-003-1419-3




962.









doi:10.1177/1545968311429688

Massion, J. (1992). Movement, posture and equilibrium: Interaction and coordination. Progress

in Neurobiology, 38(1), 35-56.

McCulloch, K. L., Buxton, E., Hackney, J., & Lowers, S. (2010). Balance, attention, and dual-task

performance during walking after brain injury: associations with falls history. Journal of

Head Trauma and Rehabilitation, 25(3), 155-163.

McIlroy, W. E., & Maki, B. E. (1997). Preferred placement of the feet during quiet stance:

development of a standardized foot placement for balance testing. Clinical

Biomechanics, 12(1), 66-70.




Mima, T., & Hallett, M. (1999). Corticomuscular coherence: a review. Journal of Clinical





Mohapatra, S., & Aruin, A. S. (2013). Static and dynamic visual cues in feed-forward postural

control. Experimental Brain Research, 224(1), 25-34. doi:10.1007/s00221-012-3286-2

129

Mohapatra, S., Krishnan, V., & Aruin, A. S. (2012). The effect of decreased visual acuity on

control of posture. Clinical Neurophysiology, 123(1), 173-182.

doi:10.1016/j.clinph.2011.06.008

Moraes, R., de Freitas, P. B., Razuk, M., & Barela, J. A. (2016). Quality of Visual Cue Affects

Visual Reweighting in Quiet Standing. PloS One, 11(3), e0150158.

doi:10.1371/journal.pone.0150158




Nagy, E., Feher-Kiss, A., Barnai, M., Domjan-Preszner, A., Angyan, L., & Horvath, G. (2007).

Postural control in elderly subjects participating in balance training. European Journal of

Applied Physiology, 100(1), 97-104. doi:10.1007/s00421-007-0407-x

Nagy, E., Toth, K., Janositz, G., Kovacs, G., Feher-Kiss, A., Angyan, L., & Horvath, G. (2004).

Postural control in athletes participating in an ironman triathlon. European Journal of

Applied Physiology, 92(4-5), 407-413. doi:10.1007/s00421-004-1157-7



Paillex, R., & So, A. (2003). Posture debout chez des sujets adultes : spécificités de l'hémiplégie.

Annales de Réadaptation et de Médecine Physique, 46(2), 71-78. doi:10.1016/s0168-

6054(03)00002-3

Palmieri, R. M., Ingersoll, C. D., Stone, M. B., & Krause, B. A. (2002). Center-of-Pressure

Parameters Used in the Assessment of Postural Control. Journal of Sport Rehabilitation,

11(1), 51-66.




Rosenberg, J. R., Amjad, A. M., Breeze, P., Brinllinger, D. R., & Halliday, D. M. (1989). The fourier

approach to the identification of functional coupling between neuronal spike trains.

Progress in Biophysics and Molecular Biology, 53(1), 1-31.

Schinkel-Ivy, A., Singer, J. C., Inness, E. L., & Mansfield, A. (2016). Do quiet standing centre of

pressure measures within specific frequencies differ based on ability to recover balance

in individuals with stroke? Clinical Neurophysiology, 127(6), 2463-2471.

doi:10.1016/j.clinph.2016.02.021

Segalowitz, S. J., Dywan, J., & Unsal, A. (1997). Attentional factors in response time variability

after traumatic brain injury: An ERP study. Journal of the International

Neuropsychological Society, 3(2), 95-107.

130

Shumway-Cook, A., & Horak, F. B. (1986). Assesssing the influence of sensory interaction on

balance. Physical Therapy, 66(10), 1548-1550.








786-795.

Vuillerme, N., & Vincent, H. (2006). How performing a mental arithmetic task modify the

regulation of centre of foot pressure displacements during bipedal quiet standing.





Wade, M. G., & Jones, G. (1997). The role of vision and spatial orientation in the maintenance

of posture. Physical Therapy, 77(6), 619-628.



44(7), 975-982. doi:10.1682/jrrd.2006.12.0158


Posture, 3(4), 193-214.




Wolpert, D. M., & Kawato, M. (1998). Multiple paired forward and inverse models for motor

control. Neural Networks, 11(7-8), 1317-1329.

Yanohara, R., Teranish, T., Yutaka, T., Taniso, G., Ueno, R., & Sonada, S. (2014). Recovery

process of standing postural control in hemiplegia after stroke. Journal of Physical

Therapy Science, 26(11), 1761-1765.

Zatsiorsky, V. M., & Duarte, M. (1999). Instant equilibrium point and its migration in standing

tasks: Rambling and trembling components of the stabilogram. Motor Control, 3(1), 28-

38.

131

Chapter 6 : Discussion

132

The overall purpose of this dissertation was to better understand the role of inter-limb

coordination in postural control and to determine the extent to which inter-limb coordination is

altered in individuals following moderate-to-severe TBI. Across three studies, the following

objectives were examined: (1) to determine whether changes in inter-limb postural

synchronization prior to bouts of postural instability are indicative of modulation in postural

set; (2) to characterize the recovery trajectory of balance control following TBI with a specific

focus on measures of postural asymmetry; and, (3) to investigate whether anticipatory and

reactive balance impairments exist after TBI by carrying out spectral analyses (power spectral

density and coherence) of static balance tasks. This dissertation provides evidence that the

level of inter-limb COP synchrony is dependent on the predictability of perturbation magnitude

prior to postural instability, suggesting that inter-limb synchrony is important during

preparation for postural instability. Additionally, this dissertation demonstrated that although

there are slight improvements in the overall balance five months post injury in individuals with

moderate-to-severe TBI, poor levels of balance control remain and weight-bearing asymmetry

was consistent one year post-injury. Levels of inter-limb synchrony did not change, but inter-

limb coherence increased across time. Weight-bearing asymmetry in TBI was noticeable in

reactive balance measures.

6.1 Summary of findings

The preparation for postural instability across different conditions of perturbation

predictability demonstrated that inter-limb synchrony of the COP in the AP and ML directions

was higher when the magnitude of the perturbation was smaller and more predictable. Larger

133

perturbations or perturbations where perturbation magnitude was unpredictable reduced

preparatory ML COP inter-limb synchrony in comparison to when conditions were performed in

blocked (rather than randomized order) and predictable. Most intriguing was how perturbation

force thresholds required less force for individuals to begin moving in a forward direction when

perturbations were randomized and most unpredictable. Inter-limb synchrony of the COP was

not affected by whether the perturbations were internally or externally generated. However,

inter-limb synchrony in the AP direction significantly decreased across blocks of trials (i.e. from

the first five trials to the last five trials) when the perturbation magnitude was larger and

unknown in external perturbations. These results demonstrate that to optimize balance

responses when a stepping response may be required, the CNS prepares for unknown

magnitude of perturbations of instability through the decoupling of inter-limb synchrony.

Although the results from the first study demonstrated functional utility of inter-limb

coupling and decoupling as part of anticipatory postural control, balance control and inter-limb

coordination after moderate-to-severe TBI has not been thoroughly studied. Thus, this was

further investigated in the second study. At five months post-injury, a reduction of the net COP

of RMS in the ML direction was observed in TBI participants as compared to the 2-month time

point. This improvement, however, was not observed in the AP direction for net COP. There

were no changes across the course of recovery in inter-limb synchrony in the AP and ML

direction. However, reduced AP and ML COP inter-limb synchrony was associated with greater

ML RMS across recovery. Individuals post-TBI did not improve their weight-bearing asymmetry

throughout all three time points and vertical force loading ratio was comparable to that of

stroke patients (Mansfield, Danells, Inness, Mochizuki, & McIlroy, 2011). Despite poor and

134

unchanged weight-bearing asymmetry in bipedal stances, individuals post-TBI stood in unipedal

stance for 10 s longer within the first few months of recovery in both limbs. Interestingly, it was

unipedal stance that demonstrated a limb difference between the left and right side with

greater unipedal stance duration in the left limb. The visual comparison of balance outcomes to

the healthy control group demonstrated that many of the balance and inter-limb coordination

outcome measures in individuals with TBI at 12 months post-injury did not reach normative

values. Though static balance impairments persist a year after injury, it remains unknown how

this transfers to dynamic balance control.

The use of spectral analyses to the COP measures in static balance provided an

opportunity to examine whether anticipatory and reactive balance control mechanisms were

impaired after TBI and to better understand how these control mechanisms were influenced by

individual limb contributions. ML net COP power spectral density in high frequencies, indicative

of reactive balance control mechanisms, decreased across the span of 12 months post-TBI. High

frequencies were affected by individual limb contributions, as inter-limb coherence increased

across recovery in high frequencies in both the AP and ML direction. The amount of weight-

bearing from each limb demonstrated a trend towards a difference in the amplitude of power

spectral density in the high frequencies. Omitting vision increased net COP power spectral

density in the AP direction at low frequencies, indicative of anticipatory balance control

mechanisms, and in the AP and ML direction at high frequencies. The omission of vision also

increased inter-limb coherence and individual foot COP spectral measures, specifically in high

frequencies. Similar to the previous study, despite initial improvements post-injury, the spectral

and spatial balance measures of the net COP remained above the normative healthy control

135

values. Thus, anticipatory and reactive balance control mechanisms appear to be diminished

after TBI with greater differences to normative healthy control values in reactive balance

control mechanisms.

Through a series of experiments and various methods of analyses of COP data, this

dissertation demonstrates the importance of inter-limb coordination in balance control in

preparing for instability, provides quantitative evidence of diminished balance control after TBI,

and demonstrates that the balance impairments after TBI may be attributable to reduced inter-

limb coordination, as exemplified by reduced weight-bearing symmetry and poor inter-limb

synchrony and inter-limb coherence. Additionally, there is evidence to suggest that both

anticipatory and reactive control mechanisms of balance control may be negatively impacted.

6.2 Inter-limb Coordination in Balance Control

As previously stated, postural control emerges from the interaction of the individual, the

task, and the environment (Shumway-Cook & Woollacott, 2012). Postural control reflects the

ability to maintain equilibrium and requires maintaining the COM within a steady BOS and also

while the BOS is altered (Maki & McIlroy, 1997). During upright quiet standing, healthy

individuals generate equal forces and torque from each limb in order to maintain balance by

keeping the COM within the BOS (Woollacott & Shumway-Cook, 1996). Measuring individual

limb contributions in static balance, which reflects inter-limb coordination, has been

informative to further understanding balance control in healthy individuals and individuals with

balance impairments (Mansfield et al., 2011; Mansfield, Mochizuki, Inness, & McIlroy, 2012;

136

Mochizuki, Ivanova, & Garland, 2005; Singer, Mansfield, Danells, McIlroy, & Mochizuki, 2013;

Winter, Prince, Stergiou, & Powell, 1993). Because individual limb contributions are critical for

postural control, inter-limb coordination and postural control are considered interrelated and

have comparable interactions between the individual, the task, and the environment (Figure 6-

1, repeat of Figure 2-2).




Ed. (2012)]

Effective inter-limb coordination in balance control requires individual limbs to produce

relatively symmetrical force output from each limb by temporally synchronizing ankle and hip

moments during upright balance. However, modulating inter-limb coordination to the context

Inter-limb

Coordination

Environment

E

T

Task

I

Individual

137

of the balance task and the environment is also necessary. In addition to overall balance control

measures (i.e. net COP), this dissertation used spatial, temporal, and frequency outcome

measures to advance the knowledge of inter-limb coordination within balance control that

includes weight-bearing asymmetry, inter-limb synchronization, and inter-limb coherence,

respectively. Subsequent sections will discuss inter-limb coordination in balance control in

preparation for balance instability in young healthy adults and inter-limb coordination and

balance control after moderate-to-severe TBI.

6.2.1 The Organization and Control of Inter-limb Coordination in Balance Control

An important consideration in characterizing the impact of TBI on inter-limb

coordination is: how is inter-limb coordination organized in the CNS? The coordination of

homologous and non-homologous limb movements is primarily driven by the activation of

supplementary motor area (SMA), premotor and sensorimotor cortices, cingulate cortex, and

cerebellum as components of a distributed motor network (Debaere et al., 2001). Increased

brain activation in these regions is observed, in particular the SMA, when movement

coordination is less stable during anti-phase or non-isodirectional movements (Debaere et al.,

2001) and during bilateral lower extremity movements (Noble, Eng, & Boyd, 2014). Relatively

equal bilateral force output from the lower extremities is thought to be driven by a central

command and distally modulated by separate controls in each limb for the task (Baldissera &

Tesio, 2017; Mochizuki, Semmler, Ivanova, & Garland, 2006). Additionally, the absence of

temporal delays in muscle activity between limbs and high correlations in the timing and

frequency of oscillations of the muscular activity in the ankle dorsiflexors and plantarflexors

138

while standing (Mochizuki et al., 2006) provide some evidence that between-limb coordination

is important for balance control. In TBI, local or diffuse injury (Gaetz, 2004; Povlishock & Katz,

2005) to the SMA, cingulate motor cortex, and cerebellum or other regions that provide

descending control to parallel spinal circuits involved in co-modulation of muscle activity can

alter the strength of task-dependent co-modulation of bilateral postural muscles (i.e. inter-limb

control) during standing. Bilateral synchronization of single motor units between limbs is

altered post-stroke (Farmer, Bremner, Halliday, Rosenberg, & Stephens, 1993) and

synchronization between motor units of ankle plantarflexor and dorsiflexor muscles in

individuals with stroke demonstrate prolonged motor unit synchronization in comparison to a

control group (Datta, Farmer, & Stephens, 1991), suggesting that bilateral motor unit activation

from descending pathways are impaired after injury and may be more reliant on coordination

of spinal, rather than descending inputs. Thus, altered motor unit synchronization which stems

from cortico-muscular synchronization relationships (Farmer, 1998) may result in poor postural

control. The resulting temporal dissociation or reduction in correlation between limbs when the

task requires high associations provides evidence for the importance of integrity of descending

pathways on inter-limb coordination in balance.

6.2.2 Inter-limb Synchrony in Anticipatory Balance Control

In Study 1 of this dissertation, inter-limb synchronization prior to postural instability in

the AP direction from either self-generated or predictable externally-generated perturbations

were scaled to the magnitude of the perturbation. Specifically, AP perturbations reduced ML

inter-limb synchronization with greater magnitude than AP inter-limb synchronization (Habib

139

Perez, Singer, & Mochizuki, 2016). Winter et al. (1993) demonstrated that the unloading and

loading mechanism was critical for control of net COP in the ML direction, controlled by hip

abductors and adductors; however, individual limb COP fluctuations in the ML direction

resulted from ankle invertors and evertors. Conversely, unloading and loading mechanisms

were negligible for net COP and individual limb COP in the AP direction. The significant

reduction in the ML inter-limb synchrony from quiet standing to the other conditions prior to

postural perturbations suggests that ankle invertors and evertors play a large role in the

preparation for instability for early and anticipatory postural adjustments (Klous, Mikulic, &

Latash, 2012; Krishnan, Latash, & Aruin, 2012). Accordingly, the tightly coupled AP inter-limb

synchrony throughout all conditions highlight that the direction of perturbation played a role in

activating the ankle plantarflexors as individuals prepared for postural instability in the forward

direction. Self-generated movements or predictable, externally-generated perturbations

engage in feedforward balance control as a result of the physiological readiness of the CNS,

central set (Prochazka, 1989). Central set is dependent on prior experience and current context,

which enables descending commands of the CNS to prepare and modify contextually

appropriate balance responses and have been shown to be scaled to the magnitude of the

perturbation (Horak, Diener, & Nashner, 1989; Mochizuki, Boe, Marlin, & McIlRoy, 2010;

Prochazka, 1989). The findings of the first study illustrate that inter-limb synchrony was actively

controlled as a result of the physiological readiness for perturbations in the AP direction, and

ankle invertors and evertors actively decoupled ML inter-limb synchrony in order to maintain

overall stability in preparation for instability. On the other hand, reduced ML inter-limb

synchrony in quiet standing in the young healthy group in comparison to ML synchrony

reported in previous studies (Mochizuki et al., 2005; Winter et al., 1993) may suggest that ML

140

inter-limb synchrony (i.e. anti-phase relationship of COP under each foot) (Debaere et al.,

2001), may be more difficult to maintain and in turn may display increased variability than AP

inter-limb synchrony. This decoupling in the ML direction may serve as an essential component

of an anticipatory postural adjustment, which would be observed in stepping tasks, but not in

quiet standing. Thus, the high instability or variability of this type of coordination may be

condition-dependent and linked to the requirements of the task.

6.2.3 Inter-limb Coordination in Balance Control after Traumatic Brain Injury

As a part of the overall purpose, this dissertation sought to determine the extent to

which inter-limb coordination was altered in individuals following moderate-to-severe TBI.

Balance control after TBI is commonly impaired and may result from strength imbalances

between limbs and coordination deficits that are prevalent after TBI (Khan et al., 2016; Walker

& Pickett, 2007) as a result of the physiological and structural changes to the body (i.e. brain)

after injury (Laxe et al., 2014). Previous studies have reported asymmetrical upright standing in

static balance (Lehmann et al., 1990) and prior to externally-generated platform perturbations

(Newton, 1995), and poor inter-limb force amplitude coordination in self-generated

perturbations (Arce, Katz, & Sugarman, 2004). Collectively, previous studies underscore that

asymmetry and inter-limb coordination are components of the balance problems that are

present after TBI.

Study 2 of this dissertation demonstrated weight-bearing asymmetry in individuals with

moderate-to-severe TBI with the least vertically loaded limb carrying 44 to 46% of total body

141

weight. Normative weight-bearing symmetry ranges between 48 to 52% between the limbs,

and although there was a 2 to 4% difference between normative values and individuals with

TBI, these differences place individuals with TBI within weight-bearing asymmetry ranges

observed in individuals with stroke (Mansfield et al., 2011; Mansfield, Mochizuki, et al., 2012;

Singer et al., 2013; Singer & Mochizuki, 2015). In Study 3 of this dissertation, weight-bearing

asymmetry was used to characterize inter-limb coordination for anticipatory and reactive

balance control mechanisms in the AP and ML direction after TBI. A trend towards differences

in the amplitude of power produced between the higher weight-bearing limb and lower weight-

bearing limb were visible in the ML direction in high frequencies, which are indicative of

reactive balance control mechanisms. In other neurologically impaired populations, such as

stroke, weight-bearing asymmetry led to impaired reactive stepping responses (Inness,

Mansfield, Bayley, & McIlroy, 2016; Inness, Mansfield, Lakhani, & McIlroy, 2014; Mansfield,

Inness, Lakhani, & McIlroy, 2012). Additionally, weight-bearing asymmetry has been shown to

decrease postural stability for dynamic balance control in lateral perturbations towards the

more loaded limb (de Kam, Kamphuis, Weerdesteyn, & Geurts, 2016). These findings suggest

that the imbalance in weight-bearing symmetry may be driving the amplitude of power in the

higher weight-bearing limb in frequencies that are important for reactive balance control. As a

result, this may place individuals with TBI at an increased risk of falls and are less likely to

recover from balance perturbations.

Inter-limb coordination after TBI was also evaluated through temporal and spectral

measures in Studies 2 and 3 of this dissertation: inter-limb COP synchronization and inter-limb

COP coherence, respectively. Individuals with moderate-to-severe TBI revealed moderate to

142

strong inter-limb COP synchronization and inter-limb COP coherence in low and high

frequencies. Although individuals with TBI produced inter-limb synchrony and inter-limb

coherence within normative healthy control values, the within-group variability of inter-limb

synchrony was higher in individuals with TBI in both the AP and ML directions. Within-group

variability in inter-limb COP coherence among individuals with TBI was not as large as inter-limb

COP synchronization. Higher within-group variability among individuals with TBI has been

reported in reaction time tasks (Hetherington, Stuss, & Finlayson, 1996; Stuss, Pogue, Buckle, &

Bondar, 1994) and gait spatial and temporal parameters (Niechwiej-Szwedo et al., 2007). Intra-

individual variability has been shown to be an important feature in individuals with TBI in

reaction time tasks (Ghajar & Ivry, 2008; Hetherington et al., 1996). Without repeated

measures from healthy controls, a comparison of intra-individual variability between groups

was not feasible. Postural sway variability, however, can be inferred using the net COP RMS. In

Study 2, inter-limb synchrony in both the AP and ML directions was associated with ML RMS,

suggesting that greater postural sway variability in mechanisms that control loading and

unloading was linked with reduced inter-limb synchronization. Similar findings were also found

in individuals post-stroke (Mansfield et al., 2011), who also demonstrate similar heterogeneity

in these measures and slightly poorer inter-limb synchrony. The relationship between inter-

limb COP coherence and RMS and inter-limb synchronization was not explored. Inter-limb COP

coherence is a relatively novel analysis for inter-limb coordination and has shown to be a

sensitive measure in individuals with Huntington’s Disease (Myklebust et al., 2009). Individuals

with Huntington’s Disease produced reduced inter-limb coherence compared to healthy young

and older adults. This finding may have resulted from the application of a statistical confidence

143

limit to the inter-limb COP coherence (Mima & Hallett, 1999; Rosenberg, Amjad, Breeze,

Brinllinger, & Halliday, 1989), which eliminated lower coherence values.

Though mean inter-limb COP coherence in individuals with TBI did not deviate outside

the 95% confidence interval of normative healthy control values at most time points after

injury, the increased coherence at high frequencies above the 95% confidence band at 12

months post-injury may suggest inappropriately high level of coupling between limbs with

respect to reactive balance control mechanisms in both the AP and ML directions. Findings from

Study 1 of this dissertation in younger healthy adults revealed that inter-limb COP

synchronization requires decoupling in preparation for unpredictable postural perturbations

(Habib Perez et al., 2016) and increased coupling in reactive balance mechanism may place

individuals at increased risk of falls. If the same mechanism of postural control from Winter et

al. (1993) are applied to inter-limb COP coherence, one would expect that inter-limb coherence

in the AP and ML directions results from ankle plantarflexors and dorsiflexors and ankle

invertors and evertors, respectively. Myklebust et al. (2009) described inter-limb coherence as

the similarity in frequencies in both limbs of the postural control system. Highly correlated

inter-limb coherence in frequencies below 0.4 Hz and above 0.4 Hz would suggest equal that

anticipatory and reactive control mechanism produced by the central postural control system

are equal in both limbs. Typically, postural control muscle activation patterns that correspond

to anticipatory and reactive control mechanism are reflected in fixed and stepping strategies

(Horak & Nashner, 1996; McIlroy & Maki, 1993), with few studies on the role of ankle

invertors/evertors in anticipatory and reactive control mechanisms (Henry, Fung, & Horak,

1998; Van Deun, Stappaerts, Levin, Janssens, & Staes, 2011). Side-alternating vibration training

144

has been shown to increase muscle strength in ankle evertors (i.e. peroneus longus) and reduce

postural sway both directions, but with greater emphasis in the ML direction (Spiliopoulou,

Amiridis, Tsigganos, & Hatzitaki, 2013). As such, Spiliopoulou et al. (2013) suggest that ankle

evertors have a primary role in ML postural control; however, ankle plantarflexors have

demonstrated a role in lateral stability in ankle invertors (Vieira, Minetto, Hodson-Tole, &

Botter, 2013). Asymmetric ankle evertor strength and uncertain contributions from AP postural

control muscles may account for the variability in inter-limb coherence in individuals with TBI

that fall outside the normal range in comparison to healthy controls and the possible

impairments in the inter-limb coherence that are linked to reactive balance control. Alternately,

the increase in inter-limb COP coherence in high frequencies may reflect brain recovery and/or

structural reorganization post-TBI (Sidaros et al., 2008) illustrating greater similarities in the

output from the cortical regions of the central postural control system that are responsible for

ankle and hip postural responses in both the AP and ML direction (Hay & Wachowiak, 2017;

Jacobs & Horak, 2007). Thus, the increase in inter-limb COP coherence outside of the healthy

control range at 12-months post injury may reflect improvements but with greater variability in

this population. Given the mixed findings in the literature and the novelty of inter-limb COP

coherence from one other study (Myklebust et al., 2009), there is uncertainty whether the

increase in inter-limb COP coherence in high frequencies reflects improvements or impairments

in postural responses post-TBI.

One question that remains is whether the differences in balance outcomes observed

between neurologically intact individuals and individuals with TBI are also indicators of balance

impairment. Using the World Health Organization ICF definition, impairment is defined as

145

problems in the physiological function or anatomical functions of the body that demonstrate

significant deviation or loss (World Health Organization, 2001). In practice, the absence of

impairment or a recovery process can be defined as a return of balance measures to premorbid

levels. Since premorbid balance measures were not available, Studies 2 and 3 used data from a

healthy control group to infer balance ability prior to injury. In some instances, there were clear

differences between TBI measures and confidence boundaries of the healthy control group;

however, this was not always the case. In addition, unlike many clinical measures, which have

established measurement properties that contribute to identification of minimally clinical

important differences (Wright, Hannon, Hegedus, & Kavchak, 2012), quantitative force platform

measures have identified measures that relate to increased falls risks in aging and

neurologically impaired populations (Maki, Holliday, & Topper, 1994; Mansfield, Mochizuki, et

al., 2012; Mansfield et al., 2015), though these measure do not have established minimal

clinical important differences. Thus, while it is possible to speculate that individuals with TBI

have impaired balance because of similarities in values to other neurological populations where

balance impairments are more clearly defined, it may be premature to say that balance ability

is impaired rather than just reduced.

6.2.4 Inter-limb Coordination and Asymmetry: Does it contribute to balance

impairment in traumatic brain injury?

As previously stated, the rationale for this dissertation was based on the position that

the balance impairments after TBI (Arce et al., 2004; Lehmann et al., 1990; Newton, 1995) may

be attributable to strength imbalances/asymmetries between limbs and coordination deficits

146

(Drijkoningen, Caeyenberghs, Vander Linden, et al., 2015; Khan et al., 2016; Walker & Pickett,

2007). The question remains: does inter-limb coordination reflect asymmetry and does it

contribute to balance impairment in TBI? From the perspective that poor inter-limb COP

synchrony was associated with greater postural sway and that individuals with TBI

demonstrated persistent weight-bearing asymmetry across time, the answer to the question is

‘yes’. However, heterogeneity within the TBI group limits generalizability. In spite of this, the

importance and relevance of symmetry in balance control is clear. Under the assumption that

asymmetry and dyscoordination negatively impact balance control by: imposing temporal

delays in reactive control (Lakhani, Mansfield, Inness, & McIlroy, 2011; Mansfield et al., 2011),

inducing inefficient biasing of weight-bearing to the weaker or non-paretic limb (Geurts, de

Haart, van Nes, & Duysens, 2005), or reducing ability to optimize force generation to suit task

demands (Horak, Henry, & Shumway-Cook, 1997), then increasing symmetry and coordination

(within the context of the task) should improve balance and reduce fall risk. Though the risk

factors for falls are multi-factorial in TBI (McKechnie, Pryor, & Fisher, 2015, 2017; Tinetti,

Speechley, & Ginter, 1998) and can be unrelated to coordination, some factors, such as lower

extremity disability, which include reduced strength, sensation, and balance, are risk factors for

falls and can be influenced by symmetry and coordination.

At a physiological level, temporal synchronization of muscle activity is observed in

muscles that span the midline including masseter, diaphragm, and rectus abdominis (Carr,

Harrison, & Stephens, 1994) and to a lesser extent in bilateral soleus muscle during postural

tasks (Mochizuki et al., 2006) in healthy adults. While coordination across segments within a

limb (i.e. intra-limb coordination) is functionally relevant (Kelso, Buchanan, & Wallace, 1991),

147

inter-limb coordination is arguably more impactful in bipedal stance and in individuals with

unilateral neurologic injury (Genthon et al., 2008; Wang, Forbes, Croft, Van der Loos, & Blouin,

2015; Wang & Newell, 2012). Importantly, bilateral synchronization of muscle activity is altered

following stroke (Farmer et al., 1993). Thus, it is possible that TBI could similarly disrupt the

processes that facilitate coordinated muscle activity and/or motor behaviour, as motoric

asymmetry in both the upper and lower limbs have been observed in individuals with TBI (Choi

et al., 2012; Hurt, Rice, McIntosh, & Thaut, 1998).

In a general sense, coordination implies organized muscle activity and segmental

movements while performing a task (Enoka & Pearson, 2013). To highlight this point,

individuals with hyperkinetic movement disorders (e.g. ataxia, chorea, dystonia, Huntington’s

Disease) experience disorganized motor control or involuntary movements through neurologic

disease or injury that limits task performance (Bakker et al., 2006; Bastian, 1997; Gibo, Bastian,

& Okamura, 2013). Unlike the relative absence of coordination found in movement disorders,

inter-limb coordination as presented in this dissertation refers to the extent of optimization or

refinement in the control of movement. A reduction in inter-limb coordination does not result

in an inability to stand upright; instead, reduced inter-coordination during upright bipedal quiet

standing cause less efficient balance control (i.e. more impaired), which could increase fall risk.

Therefore, the overall impaired coordination (asymmetry) after TBI that has been highlighted in

previous studies (Campbell & Parry, 2005; Khan, Baguley, & Cameron, 2003; Walker & Pickett,

2007), was noticeable in various inter-limb coordination balance control measures presented in

this dissertation.

148

6.3 Balance Recovery after Traumatic Brain Injury

Previous studies examining the longitudinal changes in balance control after TBI have

investigated posturographic changes within the first few weeks post-injury (Wade, Canning,

Fowler, Felmingham, & Baguley, 1997), while other studies analyze overall balance measures

many years post-injury (Ponsford et al., 2014; Walker & Pickett, 2007). Two of the studies that

comprise this dissertation were able to evaluate posturographic balance control measures

across three times points after injury at two, five, and twelve months post-injury. Changes in

balance control measures were observed in net COP ML postural sway, duration in unipedal

stance, and inter-limb COP coherence. Improvements in balance control were observed from

two to five months of injury in ML postural sway, as demonstrated by a reduction of ML RMS

within these two time points. Consistent with previous studies (Wade et al., 1997; Walker &

Pickett, 2007), the greatest improvements in balance control occurred within the first six

months after TBI in previous studies. Mirroring the reduction in net COP ML RMS, a reduction

of the amplitude of the PSD in the high frequencies in ML direction were found from two to five

months post-injury. While there were no further improvements by 12 months post-injury in net

COP ML RMS, the amplitude in the PSD of the net COP in the high frequencies in the ML

direction continued to improve. The additional reduction in the PSD in high frequencies of the

net ML COP suggests that there were improvements in reactive balance control mechanism up

to 12 months post injury. Net ML postural sway is controlled by hip abductors and adductors,

and reflects the ability to load and unload the vertical force from one limb to another (Winter,

1995; Winter et al., 1993). As there were no changes in AP postural sway, the reduction in ML

sway five months after injury suggests that the organization of the CNS may prioritize the

149

reduction ML RMS to increase stability and reduce risk of falls (Maki et al., 1994; Manor et al.,

2010; Mitchell, Collins, De Luca, Burrows, & Lipsitz, 1995). Additionally, the changes that were

detected in the spectral measures that were not apparent in the spatial measures of balance

suggest that spatial measures may lack sensitivity in identifying improvements or changes in

balance control in individuals with moderate-to-severe TBI.

Another variable that demonstrated improvements across recovery after TBI is duration

in unipedal stance. In the second study, the average time in unipedal stance increased by

approximately 10 seconds from two to five months post-injury, suggesting improvements in

strength and postural steadiness in both limbs (Jonsson, Seiger, & Hirschfeld, 2004). Unipedal

stance is a common task that is included in various clinical balance assessments and has been

associated with increased fall risk (Hurvitz, Richardson, Werner, Ruhl, & Dixon, 2000; Springer,

Marin, Cyhan, Roberts, & Gill, 2007). Prior studies examining unipedal stance duration after

moderate-to-severe TBI have reported reduced duration in patients with chronic TBI (Williams

& Morris, 2011) and similar unipedal stance time at five months post-injury, which is in

alignment with the months post-injury found in previous studies (McFadyen, Swaine, Dumas, &

Durand, 2003). Unipedal stance time has been associated with mobility (Williams & Morris,

2011), and thus unipedal stance may be an important assessment during the course of recovery

when other net measures of balance may not detect such improvements or when

posturographic measures are not feasible for data collection.

One measure of inter-limb coordination demonstrated changes post-injury, both in the

AP and ML direction. Inter-limb COP coherence increased between two and twelve months

post-injury in high frequencies. Though an increase in inter-limb COP coherence may be

150

illustrative of improvements in inter-limb coordination, the increased coherence at high

frequencies in both the AP and ML direction may also suggest an inappropriately high level of

coupling between limbs, specifically, with respect to reactive balance control mechanisms.

Inter-limb coherence is a relatively novel posturographic measure of balance control, and

though the spectral analysis enabled inferences of reactive balance control, there is much need

to investigate the inter-limb coordination prior to reactive balance responses.

It should be noted that some measures of balance control did not identify

improvements over time in Study 2 and 3 of this dissertation. Net COP AP RMS and power

spectral density, inter-limb COP synchronization and inter-limb COP coherence in low

frequencies in both the AP and ML direction, and absolute stance load symmetry did not

improve across time. The absence of improvements in AP RMS, but reductions in ML RMS, in

individuals with TBI suggests that the organization of the CNS after injury may prioritize the

reduction of ML RMS to increase stability and reduce the risk of falls (Maki et al., 1994; Manor

et al., 2010; Mitchell et al., 1995). Though inter-limb COP synchronization did not improve

across time, secondary analyses of the relationship between COP RMS and inter-limb COP

synchrony suggest that reduced inter-limb synchrony was linked to greater ML postural sway.

Although absolute stance load symmetry did not change over time in Study 2, using absolute

stance load symmetry in study 3 provided further information that the higher weight-bearing

limb may be driving the amplitude of the COP oscillations for mechanisms that are important

for reactive balance control. Given the absence of improvement in several balance measures, it

is possible that cognitive and psychiatric factors affect the ability to recover their motor

impairments. Cognitive function after TBI in various cognitive domains has shown

151

improvements from the 2-month to 5-month window, while only manual motor (i.e. grip

strength) and visuospatial domains continue to improve from the 5-month to 12-month

window with an upward trajectory (Christensen et al., 2008). There is evidence to suggest

cognitive and motor functions compete for resources throughout the course of recovery,

particularly in the latter half of the year post injury. (Green et al., 2006). The lack of

improvements in balance measures support the notion that motor impairments that affect

balance persist after injury and may result from competing resources to cognitive functions. It is

unclear as to why prioritization in cognitive function over motor function may have occurred in

the Study 2 of this dissertation; however, given that cognitive and motor dysfunction continues

post-injury and both resources are required for postural control, it is likely that individuals with

TBI are at higher risk of falls.

6.4 Limitations and Future Directions

Despite identifying temporal inter-limb measures in the preparation for postural

instability and also spatial, temporal, spectral measures of inter-limb coordination after

moderate-to-severe TBI, there are several limitations from these studies that will be addressed

to improve future investigations. Firstly, the experimenter in Study 1 standing immediately

behind the participant during externally-generated perturbations may have produced different

levels of arousal in participants and this was not measured. Increased levels of arousal may

have contributed to the large between-subject variability observed in inter-limb synchrony. One

method to address any effects of the experimenter (i.e. experimenter bias) in externally-

generated perturbations is to perturb individuals using a moving platform that will induce

152

similar feet-in-place and change-in-support responses. Future studies may also benefit from

observing electrodermal activity during externally-generated perturbations, like in previous

studies (Maki & Whitelaw, 1993; Sibley, Mochizuki, Frank, & McIlroy, 2010), and how that may

relate to inter-limb synchrony. Understanding levels of arousal during the preparation for

postural instability would discern the postural strategies used in the coupling and decoupling of

AP and ML inter-limb synchronization, respectively.

Studies 2 and 3 evaluated balance control across time in individuals with moderate-to-

severe TBI. The sample size was significantly reduced as a result of the inclusion/exclusion

criteria to include participants who completed three time points. Though this was the first

study to conduct a longitudinal analysis with quantitative posturographic measures of balance

control in individuals with TBI, future studies may benefit from hierarchical linear modeling

(Gueorguiva & Krystal, 2004) to determine the recovery projection of balance control and the

within-subject variability in participants who missed a testing session. The use of a hierarchical

linear model would also address the variability in days post injury when balance was assessed

and the injury severity of individuals with TBI on the balance outcome measures.

Injury severity is typically calculated from Glasgow Coma Scale or the duration of post

traumatic amnesia; however, the complexity of TBI with respect to the type of brain injury (i.e.

diffuse and/or focal), the region of injury, or depth of injury may inform which type of injury

may account for the balance outcomes throughout recovery. Exploring whether there are

relationships with brain injury location and balance control or inter-limb coordination measures

will be beneficial in future studies, as it may identify the relationship between specific sites of

injury and specific balance impairments. Although, it should be noted that establishing brain-

153

behaviour relationships in moderate-severe TBI is challenging. It is uncertain whether

individuals with poor balance control or inter-coordination may have increased injury in the

cortical regions that are required for coordination (Debaere et al., 2001). Furthermore,

although inter-limb COP synchronization and coherence did not differ from healthy control

normative values, the between-subject variability among individuals with TBI was greater than

the healthy controls. Future studies may benefit from identifying participants that lie within and

outside a confidence band of normative values and probe further with intra-individual

variability to establish which participants are at most risk of balance instability and potential

falls risk.

In the longitudinal analysis of balance control in TBI participants, there was significant

body weight gain across the 12 months post-injury. Body weight may have influenced the

quantitative balance measures, as an increase in body weight may decrease postural stability

(Hue et al., 2007). Future studies may profit from obtaining approximate premorbid weight

values, as it was uncertain whether the 8 to 9 kg weight gain was a return to premorbid weight

or related to inactivity following injury, and whether the overall increase in COP RMS resulted

from increased weight. Moreover, given that individuals with TBI were recruited through an

inpatient facility, the recovery of balance during the first five months may have been influenced

by inpatient and outpatient rehabilitation. Future longitudinal studies may benefit from

tracking the types and frequency of rehabilitation provided.

Static balance control is only one aspect of balance control and without testing

individuals in dynamic balance environments the study is limited to understand the falls risks

after TBI. Although this was beyond the scope of this dissertation, it is recommended that

154

future studies investigate anticipatory and reactive balance control with quantitative

posturographic measures across the span of recovery after TBI. Individuals with TBI are more

susceptible to higher levels of anxiety post injury (Osborn, Mathias, Fairweather-Schmidt, &

Anstey, 2016) and increased levels of anxiety has been shown to alter balance control strategies

(Adkin, Frank, Carpenter, & Peysar, 2000; Carpenter, Frank, & Silcher, 1999; Carpenter, Frank,

Silcher, & Peysar, 2001; Sturnieks, Delbaere, Brodie, & Lord, 2016). Additionally, increased

levels of anxiety have found to alter muscular and cortical neurophysiological signals in

response to postural perturbations (Adkin, Campbell, Chua, & Carpenter, 2008; Sibley et al.,

2010). Thus, electrodermal activity, which can measure levels of stress or anxiety, during

perturbation testing is highly suggested in future balance studies in individuals with TBI.

In all three studies, where young and age matched healthy controls were assessed,

balance-specific medical history questionnaires were used in the inclusion criteria to exclude

anyone with musculoskeletal injuries or neurological injuries that may affect the findings.

Though this is a standard practice in most studies, it is important to consider that future studies

that include an age-matched control group may benefit from obtaining measures for sensory

systems through additional assessments in both the control and clinical group. High variability

in balance measures in the control group may have been attributed to age-related changes

(Abrahamova & Hlavacka, 2008). Additionally, high variability in individuals with TBI may be

have been attributed to the plantar cutaneous sensation deficits post injury, as balance

impairments post-stroke have been associated with plantar cutaneous sensation deficits

(Parsons, Mansfield, Inness, & Patterson, 2016). Thus, the inclusion of other sensory

assessments such as monofilament assessments of the feet (Collins, Visscher, De Vet,

155

Zuurmond, & Perez, 2010) and joint proprioception (Han, Waddington, Adams, Anson, & Liu,

2016) will further inform differences in balance measures found between a control group and

individuals with TBI and within individuals with TBI.

In revisiting the conceptual model of this dissertation (Figure 2-3), the present work

addressed how sensorimotor impairments affect balance control and in turn inter-limb

coordination. Specifically, this dissertation identified how asymmetry affects balance control

and inter-limb coordination in balance control, and how inter-limb coordination with respect to

balance control mechanism was altered after TBI. Though the current study focused primarily

on the balance after TBI, there are many sequelae of TBI that need to be taken into

consideration in future studies. The interplay of cognitive and sensorimotor factors that may be

competing for resources may be further compromised in environments that increase anxiety or

in individuals that are predisposed to psychiatric conditions, such as depression.

6.5 Implications for Rehabilitation

From one study of this dissertation there is considerable evidence demonstrating that

individuals with moderate-to-severe TBI experience balance asymmetry at 2 months post-injury

and that this does not improve significantly across time. These findings support qualitative

observations of balance asymmetry of previous work (Newton, 1995). Given the motor

impairments found in balance control after TBI, it is important to note that rehabilitation

interventions that address balance asymmetry between the 2-month and 5-month window post

injury, when the majority of recovery is visible across time, may benefit individuals with

moderate-to-severe TBI. Unilateral lower-extremity strength training on both limbs (Vinstrup et

156

al., 2017) and symmetrical body-weight distribution training (Cheng, Wu, Liaw, Wong, & Tang,

2001) have led to balance asymmetry improvements in individuals with stroke, who typically

experience greater balance asymmetry (Cheng et al., 2001). Other types of rehabilitation that

have been proposed to improve balance in individuals with TBI include ballistic balance training

(Williams et al., 2016) or dynamic balance training that intend to improve specific white matter

regions (Drijkoningen, Caeyenberghs, Leunissen, et al., 2015). The combination of balance

training with a focus on asymmetry presents an attractive option for possible future

interventions.

The current dissertation also identified a high degree of coupling when preparing for

predictable and low magnitude perturbations, while unpredictable perturbations with variable

magnitudes produced lower levels of coupling. In addition, balance impairments corresponding

to both anticipatory and reactive balance control mechanisms demonstrated greater postural

sway amplitudes, which deviated from healthy controls. Identifying dynamic balance control

components that may be impaired, such as stepping reaction time, anticipatory postural

adjustments, muscle onset and magnitude is critical for further understanding balance

impairments after TBI. Furthermore, given the cognitive sequelae post-TBI (Azouvi, Vallat-

Azouvi, & Belmont, 2009; Dockree & Robertson, 2011; Rabinowitz & Levin, 2014), rehabilitation

interventions that increase salience of balance perturbations by pairing them with other

sensory stimuli or cues may benefit the balance performance of individuals with TBI. For

example, bouts of postural perturbations paired with auditory cues reduced reaction times

when postural perturbations were no longer present (Lakhani, Miyasike-Dasilva, Vette, &

McIlroy, 2013). In addition to an increased arousal, the study demonstrated that the CNS can

157

adapt to become more prepared with this type of exposure (Lakhani et al., 2013). Additionally,

increased somatosensory stimuli through haptic light touch contact reduced COP variability just

prior to predictive voluntary perturbations and also after predictive voluntary and external

reactive postural perturbations (Johannsen, Wing, & Hatzitaki, 2007). In consideration of the

findings of Study 2 that, on an individual level, improvements in balance measures over time

were at times dissociated from improvements in cognitive measures, the potential utility of

combining cuing with perturbations may only be appropriate for some individuals. Combining

additional sensory stimuli or cues with perturbations may only benefit individuals with TBI who

demonstrate improvements in cognitive measures, as the ability to process additional cognitive

cues or sensory information may positively affect postural timing and magnitude components

of dynamic balance control.

6.6 Summary and Conclusions

The overall purpose of this dissertation was to further understand the role of inter-limb

coordination in postural control and determine the extent to which inter-limb coordination is

altered in individuals following moderate-to-severe TBI. Outcomes from the studies presented

in this dissertation demonstrated that inter-limb synchronization decreased as the magnitude

of balance perturbations became more unpredictable. These results indicated that the CNS

scales the physiological readiness to the magnitude of the largest perturbation magnitude by

decoupling the inter-limb coordination in order to optimize balance responses. Furthermore,

balance control and inter-limb coordination was investigated in individuals with moderate-to-

severe TBI. The outcomes of the latter studies demonstrated balance asymmetry and poor

158

balance control in individuals with TBI. There were few measure of balance control that

demonstrated improvements across recovery including, net COP ML postural sway in both RMS

and power spectral density, duration in unipedal stance, and COP inter-limb coherence. The

majority of improvements in these balance measures were observed between 2-months and 5-

months post-injury, while COP inter-limb coherence demonstrated improvements by 12-

months post-injury. AP postural sway and balance asymmetry did not improve after TBI and the

majority of balance measures were outside the 95% confidence interval of healthy control

values, indicating that balance impairments persist a year after injury. Spectral analyses of the

COP balance data after TBI demonstrated that both anticipatory and reactive balance control

mechanisms, visible in low and high frequencies in the COP spectra, were also impaired after

injury. Two main conclusions can be drawn from these experiments: inter-limb coordination is

important to the preparation of postural instability; and, asymmetry in balance control may be

a driver to poor balance control and inter-limb coordination after TBI that affect both

anticipatory and reactive balance control mechanisms. Future studies examining balance

control after TBI need to incorporate dynamic balance tests, including anticipatory and reactive

balance responses, to further understand the risk of falling post-TBI and how to better provide

interventions for this population.

159

6.7 References

Abrahamova, D., & Hlavacka, F. (2008). Age-related changes of human balance during quiet

stance. Physiological Research, 57(6), 957-964.

Adkin, A. L., Campbell, A. D., Chua, R., & Carpenter, M. G. (2008). The influence of postural

threat on the cortical response to unpredictable and predictable postural perturbations.

Neuroscience Letters, 435(2), 120-125. doi:10.1016/j.neulet.2008.02.018

Adkin, A. L., Frank, J. S., Carpenter, M. G., & Peysar, G. W. (2000). Postural control is scaled to

level of postural threat. Gait and Posture, 12(2), 87-93.



doi:10.1016/j.humov.2003.12.002

Azouvi, P., Vallat-Azouvi, C., & Belmont, A. (2009). Cognitive deficits after traumatic coma.

Progress in Brain Research, 177, 89-110.

Bakker, M., Allum, J. H., Visser, J. E., Gruneberg, C., van de Warrenburg, B. P., Kremer, B. H., &

Bloem, B. R. (2006). Postural responses to multidirectional stance perturbations in

cerebellar ataxia. Experimental Neurology, 202(1), 21-35.

doi:10.1016/j.expneurol.2006.05.008

Baldissera, F. G., & Tesio, L. (2017). APAs Constraints to Voluntary Movements: The Case for

Limb Movements Coupling. Frontiers in Human Neuroscience, 11, 152.

doi:10.3389/fnhum.2017.00152

Bastian, A. J. (1997). Mechanisms of ataxia. Physical Therapy, 77(6), 672-673.

Campbell, M., & Parry, A. (2005). Balance disorder and traumatic brain injury: preliminary

findings of a multi-factorial observational study. Brain Injury, 19(13), 1095-1104.

doi:10.1080/02699050500188898

Carpenter, M. G., Frank, J. S., & Silcher, C. P. (1999). Surface height effects on postural control:

A hypothesis for a stiffness strategy for stance. Journal of Vestibular Research, 9(4), 277-

286.

Carpenter, M. G., Frank, J. S., Silcher, C. P., & Peysar, G. W. (2001). The influence of postural

threat on the control of upright stance. Experimental Brain Research, 138(2), 210-218.

doi:10.1007/s002210100681

Carr, L. J., Harrison, L. M., & Stephens, J. A. (1994). Evidence for bilateral innervation of certain

homologous motoneurone pools in man. Journal of Physiology, 475(2), 217-227.

160

Cheng, P. T., Wu, S. H., Liaw, M. Y., Wong, A. M., & Tang, F. T. (2001). Symmetrical body-weight

distribution training in stroke patients and its effect on fall prevention. Archives of

Physical Medicine and Rehabilitation, 82(12), 1650-1654. doi:10.1053/apmr.2001.26256




Christensen, B. K., Colella, B., Inness, E., Hebert, D., Monette, G., Bayley, M., & Green, R. E.

(2008). Recovery of cognitive function after traumatic brain injury: a multilevel modeling

analysis of Canadian outcomes. Archives of Physical Medicine and Rehabilitation, 89(12

Suppl), S3-15.

Collins, S., Visscher, P., De Vet, H. C., Zuurmond, W. W., & Perez, R. S. (2010). Reliability of the

Semmes Weinstein Monofilaments to measure coetaneous sensibility in the feet of

healthy subjects. Disability and Rehabilitation, 32(24), 2019-2027.

doi:10.3109/09638281003797406

Datta, A. K., Farmer, S. F., & Stephens, J. A. (1991). Central nervous pathways underlying

synchronization of human motor unit firing studied during voluntary contractions.

Journal of Physiology, 432, 401-425.




Debaere, F., Swinnen, S. P., Beatse, E., Sunaert, S., Van Hecke, P., & Duysens, J. (2001). Brain

Areas Involved in Interlimb Coordination: A Distributed Network. Neuroimage, 14(5),

947-958. doi:10.1006/nimg.2001.0892

Dockree, P. M., & Robertson, I. H. (2011). Electrophysiological markers of cognitive deficits in

traumatic brain injury: a review. International Journal of Psychophysiology, 82(1), 53-60.

doi:10.1016/j.ijpsycho.2011.01.004

Drijkoningen, D., Caeyenberghs, K., Leunissen, I., Vander Linden, C., Leemans, A., Sunaert, S., . . .

Swinnen, S. P. (2015). Training-induced improvements in postural control are

accompanied by alterations in cerebellar white matter in brain injured patients.

Neuroimage Clinical, 7, 240-251. doi:10.1016/j.nicl.2014.12.006




1332. doi:10.1089/neu.2014.3787

161

Enoka, R. M., & Pearson, K. G. (2013). The Moto Unit and Muscle Action. In E. Kandel, J. H.

Schwartz, T. M. Jessell, S. A. Siegelbaum, & A. J. Hudspeth (Eds.), Principles of Neural

Science. Toronto: McGraw Hill.

Farmer, S. F. (1998). Rhythmicity, synchronization and binding in human and primate motor

systems. Journal of Physiology, 509(1), 3-14.

Farmer, S. F., Bremner, F. D., Halliday, D. M., Rosenberg, J. R., & Stephens, J. A. (1993). The

frequency content of common synaptic inputs to motoneurones studied during

voluntary isometric contraction in man. Journal of Physiology, 470, 127-155.

Gaetz, M. (2004). The neurophysiology of brain injury. Clinical Neurophysiology, 115(1), 4-18.

doi:10.1016/s1388-2457(03)00258-x

Genthon, N., Rougier, P., Gissot, A. S., Froger, J., Pelissier, J., & Perennou, D. (2008).

Contribution of each lower limb to upright standing in stroke patients. Stroke, 39(6),

1793-1799. doi:10.1161/STROKEAHA.107.497701

Geurts, A. C., de Haart, M., van Nes, I. J., & Duysens, J. (2005). A review of standing balance

recovery from stroke. Gait and Posture, 22(3), 267-281.

doi:10.1016/j.gaitpost.2004.10.002



doi:10.1177/1545968308315600

Gibo, T. L., Bastian, A. J., & Okamura, A. M. (2013). Cerebellar ataxia impairs modulation of arm

stiffness during postural maintenance. Journal of Neurophysiology, 110(7), 1611-1620.

doi:10.1152/jn.00294.2013

Green, R. E., Christensen, B., Melo, B., Monette, G., Bayley, M., Hebert, D., . . . McIlroy, W.

(2006). Is there a trade-off between cognitive and motor recovery after traumatic brain

injury due to competition for limited neural resources? Brain and Cognition, 60(2), 199-

201.

Gueorguiva, R., & Krystal, J. H. (2004). Move over ANOVA: Progress in analyzing repeated-

measures data and its reflection in papers published in the Archives of General

Psychiatry. Archives of General Psychiatry, 61(3), 310-317.

Habib Perez, O., Singer, J. C., & Mochizuki, G. (2016). The effects of predictability on inter-limb

postural synchronization prior to bouts of postural instability. Gait and Posture, 46, 167-

172.

Han, J., Waddington, G., Adams, R., Anson, J., & Liu, Y. (2016). Assessing proprioception: A

critical review of methods. Journal of Sport and Health Science, 5(1), 80-90.

doi:10.1016/j.jshs.2014.10.004

162

Hay, D. C., & Wachowiak, M. P. (2017). Analysis of free moment and center of pressure

frequency components during quiet standing using magnitude squared coherence.

Human Movement Science, 54, 101-109. doi:10.1016/j.humov.2017.04.002

Henry, S. M., Fung, J., & Horak, F. B. (1998). EMG Response to maintain stance during

multidirectional surface translations. Journal of Neurophysiology, 80(4), 1939-1950.

Hetherington, C. R., Stuss, D. T., & Finlayson, M. A. J. (1996). Reaction time and variability 5 and

10 years after traumatic brain injury. Brain Injury, 10(7), 473-486.

Horak, F., Henry, S. M., & Shumway-Cook, A. (1997). Postural perturbations: New insights for

treatment of balance disorders. Physical Therapy, 77(5), 517-533.



Horak, F. B., & Nashner, L. M. (1996). Central programming of postural movements: Adaptation

to altered support-surface configurations. Journal of Neurophysiology, 55(6), 1369-1381.

Hue, O., Simoneau, M., Marcotte, J., Berrigan, F., Dore, J., Marceau, P., . . . Teasdale, N. (2007).

Body weight is a strong predictor of postural stability. Gait and Posture, 26(1), 32-38.

doi:10.1016/j.gaitpost.2006.07.005

Hurt, C. P., Rice, R. R., McIntosh, G. C., & Thaut, M. H. (1998). Rhythmic Auditory Stimulation in

Gait Training for Patients with Traumatic Brain Injury. Journal of Music Therapy, 35(4),

228-241.

Hurvitz, E. A., Richardson, J. K., Werner, R. A., Ruhl, A. M., & Dixon, M. R. (2000). Unipedal

Stance Testing as an Indicator of Fall Risk Among Older Outpatients. Archives of Physical

Medicine and Rehabilitation, 81(5), 587-591. doi:10.1053/mr.2ooO.6293

Inness, E. L., Mansfield, A., Bayley, M., & McIlroy, W. E. (2016). Reactive Stepping After Stroke:

Determinants of Time to Foot Off in the Paretic and Nonparetic Limb. Journal of

Neurologic Physical Therapy, 40(3), 196-202. doi:10.1097/NPT.0000000000000132

Inness, E. L., Mansfield, A., Lakhani, B., & McIlroy, W. E. (2014). Impaired Reactive Stepping

Among Patients Ready for Discharge From Inpatient Stroke Rehabilitation. Physical

Therapy, 94(12), 1755-1764.



Johannsen, L., Wing, A. M., & Hatzitaki, V. (2007). Effects of maintaining touch contact on

predictive and reactive balance. Journal of Neurophysiology, 97(4), 2686-2695.

doi:10.1152/jn.00038.2007

163

Jonsson, E., Seiger, A., & Hirschfeld, H. (2004). One-leg stance in healthy young and elderly

adults: a measure of postural steadiness? Clinical Biomechanics, 19(7), 688-694.

doi:10.1016/j.clinbiomech.2004.04.002

Kelso, J. A. S., Buchanan, J. J., & Wallace, S. A. (1991). Order parameters for the neural

organization of single, multijoint limb movement patterns. Experimental Brain Research,

85(2). doi:10.1007/bf00229420

Khan, F., Amatya, B., Judson, R., Chung, P., Truesdale, M., Elmalik, A., & Galea, M. P. (2016).

Factors associated with long-term functional and psychological outcomes in persons

with moderate to severe traumatic brain injury. Journal of Rehabilitation Medicine,

48(5), 442-448. doi:10.2340/16501977-2084

Khan, F., Baguley, I. J., & Cameron, I. J. (2003). Rehabilitation after traumatic brain injury.

Medical Journal of Australia, 178(6), 290-295.



116. doi:10.1016/j.jelekin.2011.11.005

Krishnan, V., Latash, M. L., & Aruin, A. S. (2012). Early and late components of feed-forward

postural adjustments to predictable perturbations. Clinical Neurophysiology, 123(5),

1016-1026. doi:10.1016/j.clinph.2011.09.014

Lakhani, B., Mansfield, A., Inness, E. L., & McIlroy, W. E. (2011). Compensatory stepping

responses in individuals with stroke: a pilot study. Physiotherapy Theory & Practice,

27(4), 299-309. doi:http://dx.doi.org/10.3109/09593985.2010.501848

Lakhani, B., Miyasike-Dasilva, V., Vette, A. H., & McIlroy, W. E. (2013). 'Priming' the brain to

generate rapid upper-limb reactions. Experimental Brain Research, 230(3), 261-270.

doi:10.1007/s00221-013-3650-x

Laxe, S., Zasler, N., Robles, V., Lopez-Blazquez, R., Tormos, J. M., & Bernabeu, M. (2014). ICF

profiling of patients with traumatic brain injury: an international professional survey.

Disability and Rehabilitation, 36(1), 82-88. doi:10.3109/09638288.2013.780641




962.

Maki, B. E., Holliday, P. J., & Topper, A. K. (1994). A prospective study of postural balance and

risk of falling in an ambulatory and independent elderly population. Journal of

Gerontology, 49(2), M72-84.



164

Maki, B. E., & Whitelaw, R. S. (1993). Influence of expectation and arousal on centre-of-

pressure responses to transient postural perturbations. Journal of Vestibular Research,

3(1), 25-29.

Manor, B., Hu, K., Zhao, P., Selim, M., Alsop, D., Novak, P., . . . Novak, V. (2010). Altered control

of postural sway following cerebral infarction: a cross-sectional analysis. Neurology,

74(6), 458-464. doi:10.1212/WNL.0b013e3181cef647




Mansfield, A., Inness, E. L., Lakhani, B., & McIlroy, W. E. (2012). Determinants of limb

preference for initiating compensatory stepping poststroke. Archives of Physical

Medicine and Rehabilitation, 93(7), 1179-1184. doi:10.1016/j.apmr.2012.02.006




doi:10.1177/1545968311429688

Mansfield, A., Wong, J. S., McIlroy, W. E., Biasin, L., Brunton, K., Bayley, M., & Inness, E. L.

(2015). Do measures of reactive balance control predict falls in people with stroke

returning to the community? Physiotherapy, 101(4), 373-380.

doi:10.1016/j.physio.2015.01.009

McFadyen, B. J., Swaine, B., Dumas, D., & Durand, A. (2003). Residual effects of a traumatic

brain injury on locomotor capacity. A first study of spatiotemporal patterns during

unobstructed and obstructed walking. Journal of Head Trauma Rehabilitation, 18(6),

512-525.

McIlroy, W. E., & Maki, B. E. (1993). Changes in early ‘automatic’ postural responses associated

with the prior-planning and execution of a compensatory step. Brain Research, 631(2),

203-211.

McKechnie, D., Pryor, J., & Fisher, M. J. (2015). Falls and fallers in traumatic brain injury (TBI)

rehabilitation settings: an integrative review. Disability and Rehabilitation, 37(24), 2291-

2299. doi:10.3109/09638288.2014.1002578

McKechnie, D., Pryor, J., & Fisher, M. J. (2017). An examination of patient characteristics that

contribute to falls in the inpatient traumatic brain injury rehabilitation setting. Disability

and Rehabilitation, 39(18), 1864-1871. doi:10.1080/09638288.2016.1212112

Mima, T., & Hallett, M. (1999). Corticomuscular coherence: a review. Journal of Clinical


165

Mitchell, S. L., Collins, J. J., De Luca, C. J., Burrows, A., & Lipsitz, L. A. (1995). Open-loop and

closed-loop postural control mechanisms in Parkinson's disease: increased mediolateral

activity during quiet standing. Neuroscience Letters, 197(2), 133-136.






doi:10.1152/jn.01322.2004

Mochizuki, G., Semmler, J. G., Ivanova, T. D., & Garland, S. J. (2006). Low-frequency common

modulation of soleus motor unit discharge is enhanced during postural control in

humans. Experimental Brain Research, 175(4), 584-595. doi:10.1007/s00221-006-0575-7






Niechwiej-Szwedo, E., Inness, E. L., Howe, J. A., Jaglal, S., McIlroy, W. E., & Verrier, M. C. (2007).

Changes in gait variability during different challenges to mobility in patients with

traumatic brain injury. Gait and Posture, 25(1), 70-77.

doi:10.1016/j.gaitpost.2006.01.002

Noble, J. W., Eng, J. J., & Boyd, L. A. (2014). Bilateral motor tasks involve more brain regions and

higher neural activation than unilateral tasks: an fMRI study. Experimental Brain

Research, 232(9), 2785-2795. doi:10.1007/s00221-014-3963-4

Osborn, A. J., Mathias, J. L., Fairweather-Schmidt, A. K., & Anstey, K. J. (2016). Anxiety and

comorbid depression following traumatic brain injury in a community-based sample of

young, middle-aged and older adults. Journal of Affective

Disorders(ttp://dx.doi.org/10.1016/j.jad.2016.09.045).

Parsons, S. L., Mansfield, A., Inness, E. L., & Patterson, K. K. (2016). The relationship of plantar

cutaneous sensation and standing balance post-stroke. Topics in Stroke Rehabilitation,

23(5), 326-332. doi:10.1080/10749357.2016.1162396




Povlishock, J. T., & Katz, D. I. (2005). Update of Neuropathology and Neurological Recovery

After Traumatic Brain Injury. Journal of Head Trauma and Rehabilitation, 20(1), 76-94.

166



Rabinowitz, A. R., & Levin, H. S. (2014). Cognitive sequelae of traumatic brain injury. Psychiatric

Clinics of North America, 37(1), 1-11. doi:10.1016/j.psc.2013.11.004

Rosenberg, J. R., Amjad, A. M., Breeze, P., Brinllinger, D. R., & Halliday, D. M. (1989). The fourier

approach to the identification of functional coupling between neuronal spike trains.

Progress in Biophysics and Molecular Biology, 53(1), 1-31.

Shumway-Cook, A., & Woollacott, M. H. (2012). Motor Control: Translating Research into

Clinical Practice: Wolters Kluwer Health, Lippincott Williams & Wilkins.

Sibley, K. M., Mochizuki, G., Frank, J. S., & McIlroy, W. E. (2010). The relationship between

physiological arousal and cortical and autonomic responses to postural instability.


Sidaros, A., Engberg, A. W., Sidaros, K., Liptrot, M. G., Herning, M., Petersen, P., . . . Rostrup, E.

(2008). Diffusion tensor imaging during recovery from severe traumatic brain injury and

relation to clinical outcome: a longitudinal study. Brain, 131(2), 559-572.

doi:10.1093/brain/awm294








786-795.

Spiliopoulou, S. I., Amiridis, I. G., Tsigganos, G., & Hatzitaki, V. (2013). Side-alternating vibration

training for balance and ankle muscle strength in untrained women. Journal of Athletic

Training, 48(5), 590-600. doi:10.4085/1062-6050-48.4.03

Springer, B. A., Marin, R., Cyhan, T., Roberts, H., & Gill, N. W. (2007). Normative values for the

unipedal stance test with eyes open and closed. Journal of Geriatric Physical Therapy,

30(1), 8-15.

Sturnieks, D. L., Delbaere, K., Brodie, M. A., & Lord, S. R. (2016). The influence of age, anxiety

and concern about falling on postural sway when standing at an elevated level. Human

Movement Science, 49, 206-215. doi:10.1016/j.humov.2016.06.014

Stuss, D. T., Pogue, J., Buckle, L., & Bondar, J. (1994). Characterization of stability of

performance in patients with traumatic brain injury: Variability and consistency on

reaction time tests. Neuropsychology, 8(3), 316-324.

167

Tinetti, M. E., Speechley, M., & Ginter, S. F. (1998). Risk factors for falls among elderly persons

living in the community The New England Journal of Medicine, 319(26), 1701-1707.

Van Deun, S., Stappaerts, K., Levin, O., Janssens, L., & Staes, F. (2011). Stability of measurement

outcomes for voluntary task performance in participants with chronic ankle instability

and healthy participants. Journal of Athletic Training, 46(4), 366-375.

Vieira, T. M., Minetto, M. A., Hodson-Tole, E. F., & Botter, A. (2013). How much does the human

medial gastrocnemius muscle contribute to ankle torques outside the sagittal plane?

Human Movement Science, 32(4), 753-767. doi:10.1016/j.humov.2013.03.003

Vinstrup, J., Calatayud, J., Jakobsen, M. D., Sundstrup, E., Jay, K., Brandt, M., . . . Andersen, L. L.

(2017). Electromyographic comparison of conventional machine strength training versus

bodyweight exercises in patients with chronic stroke. Topics in Stroke Rehabilitation,

24(4), 242-249. doi:10.1080/10749357.2016.1274466






44(7), 975-982. doi:10.1682/jrrd.2006.12.0158

Wang, P., Forbes, P. A., Croft, E. A., Van der Loos, H. F. M., & Blouin, J. S. (2015, 25-29 Aug.

2015). Shifting the balance of human standing: Inter-limb coordination for the control of

a robotic balance simulation. Paper presented at the 2015 37th Annual International

Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

Wang, Z., & Newell, K. M. (2012). Asymmetry of foot position and weight distribution channels

the inter-leg coordination dynamics of standing. Experimental Brain Research, 222(4),

333-344. doi:10.1007/s00221-012-3212-7

Williams, G., Ada, L., Hassett, L., Morris, M. E., Clark, R., Bryant, A. L., & Olver, J. (2016). Ballistic

strength training compared with usual care for improving mobility following traumatic

brain injury: protocol for a randomised, controlled trial. Journal of Physiotherapy, 62(3),

164. doi:10.1016/j.jphys.2016.04.003

Williams, G., & Morris, M. E. (2011). Tests of static balance do not predict mobility performance

following traumatic brain injury. Physiotherapy Canada, 63(1), 58-64.

doi:10.3138/ptc.2009-53


Posture, 3(4), 193-214.

168






World Health Organization. (2001). International classification of functioning, disability and

health: ICF. Geneva: World Health Organization.

Wright, A., Hannon, J., Hegedus, E. J., & Kavchak, A. E. (2012). Clinimetrics corner: a closer look

at the minimal clinically important difference (MCID). Journal of Manual and

Manipulative Therapy, 20(3), 160-166. doi:10.1179/2042618612Y.0000000001

169

Appendices

Appendix A – Relationship between inter-limb synchrony and mediolateral postural sway

variability

Appendix A Figure 1 – Scatter plots and lines of best fit displaying the relationship between

inter-limb synchrony and net COP mediolateral (ML) postural sway variability (RMS) for TBI

participants. Relationships between anteroposterior (AP) inter-limb synchrony (Rxy(0)) and net

COP ML RSM at T1 (A), T2 (B) and T3 (C) are plotted. Relationships between ML inter-limb

synchrony and net COP ML RMS at T1 (D), T2 (E), and T3 (F) are also plotted. Statistically

significant correlations were found in the relationships presented in panels A-E.

Documents

Inter-limb Coordination in Balance Control: Implications ... · from each limb in upright stance illustrates the importance of the integrity of the central nervous system (CNS) and