University of Groningen
Neural control of balance in increasingly difficult standing tasks
Nandi, Tulika
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Nandi, T. (2019). Neural control of balance in increasingly difficult standing tasks. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Neural control of balance in
increasingly difficult standing tasks
Promovendus
Tulika Nandi
The studies in this thesis are the result of a collaboration between the Center for Human Movement Sciences, part of the University Medical Center Groningen, University of Groningen, the Netherlands and the Division of Biokinesiology and Physical Therapy, part of the Herman Ostrow School of Den-tistry, University of Southern California, USA. The studies described in chapters 3 and 5 of this thesis were conducted at the Center for Human Movement Sciences and the studies described in chapters 2 and 5 were conducted at the Division of Biokinesiology and Physical Therapy.
This thesis was financially supported by: • University Medical Center Groningen • Graduate School of Medical Sciences
• Division of Biokinesiology and Physical Therapy, University of Southern California
Cover design: Piyal Sen Gupta www.E-Dezine.com
Cover Photo: bigstock-zen-stones-balance-81353279-e1452860672419 Human Footprints original pic: Human footprints free icon-Freepik/Pack: Footprints/Category: Gestures
Layout by: XML 2 Publish Printing by: Gildeprint, Enschede ISBN Print: 9789463234436 ISBN Digital: 9789463234443 Paranymphs: Hans Rijnks and Leslie De Joya © Copyright 2018, T. Nandi
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage or retrieval system, without written permission from the author.
Neural control of balance in
increasingly difficult standing tasks
PhD thesis
to obtain the degree of PhD at the University of Groningen
on the authority of the Rector Magnificus prof. E. Sterken
and in accordance with the decision by the College of Deans. This thesis will be defended in public on Wednesday 23 January 2019 at 12.45 hours
by
Tulika Nandi
born on 8 January 1988Supervisors Prof. T. Hortobágyi Prof. B.E. Fisher Co-supervisors Dr. C.J.C. Lamoth Dr. G.J. Salem
Assessment Committee Prof. J. van Dieën
Prof. W. Taube
Prof. M.A.J. de Koning-Tijssen Prof. E. Otten
your magical book cabinet was the first step in the journey that led to this book.
Page number
Chapter 1 General Introduction 9
Chapter 2 Increasing mediolateral standing sway is associated with increasing corticospinal excitability, and decreasing M1 inhibition and facilitation
21
Chapter 3 In standing, corticospinal excitability is proportional to COP velocity whereas M1 excitability is participant-specific
35
Chapter 4 Standing task difficulty related increase in agonist-agonist and agonist-antagonist common inputs are driven by corticospinal and subcortical inputs respectively
53
Chapter 5 Assessing balance confidence in young adults 71
Chapter 6 General Discussion 91
Appendices Summary 103
Nederlandse samenvatting 111
Acknowledgements 113
About the author 116
Chapter 1
As early as the 1880s, it was known that standing balance is altered by mechanical or sen-sory manipulations [1,2], and by the 1920s, it was regarded as a window into neuromuscu-lar control [3]. Clinicians and researchers often increase task difficulty by manipulating the base of support or visual input in order to detect age or pathology related balance deterio-ration, which is not apparent in quiet standing [4–6]. Indeed, when standing task difficulty increases, greater neural input is required to tune lower extremity muscle activation [7–13]. However, much is yet to be learnt about the brain areas and neurophysiological processes which underlie this increase in neural input, and how the input to multiple muscles is coordinated. Additionally, the neural basis for individual differences in balance control, driven by attributes like confidence, has not been examined. Greater insight into the neural control of increasingly difficult standing tasks in healthy young adults will contribute to a better understanding of the strategies used for managing balance deterioration induced by aging and/or pathology.
1. NeurAl CoNTrol of sTANdiNg bAlANCe: CorTiCAl iNPuTs
To musCles
Several brain areas contribute to the neural control of standing balance. During imagined standing, neuroimaging revealed activation in the cerebellum, basal ganglia, prefrontal cortex, premotor cortex, brainstem and thalamus [14–16]. However, the lack of actual muscle contraction is a major drawback of examining imagined standing and may explain why little or no activation is observed in the motor cortical areas, which are the focus of this thesis. While early animal studies suggested that balance is maintained using subcorti-cal reflexes [17,18], now there is ample evidence that descending inputs from the cortex are also important for maintaining balance. For example, electroencephalogram recordings during unconstrained standing revealed cortical activation in the fronto-central motor areas [10,11,19].
When task difficulty increases, it is essential to have more versatile muscle activation patterns to meet both expected and unexpected biomechanical demands. Therefore, the reliance on cortical descending commands which support flexible movement patterns, compared to stereotypical subcortical reflexes [13,20], is expected to increase when task difficulty increases. Specifically, it is hypothesized that when manipulations like base of support modification increase task difficulty, greater cortical inputs will be required to modulate or supplement the subcortical inputs to muscles. This thesis focuses on neural inputs to lower extremity muscles.
12 Chapter 1
2. effeCT of TAsk diffiCulTy oN siNgle musCle NeurAl
exCiTAbiliTy
When standing task difficulty increases, corticospinal excitability increases [7–9,12,13] and it is likely that both cortical and subcortical excitability changes contribute to this increase. Within the motor cortex, inhibition and facilitation, which are mediated by GABAa/ GABAb and glutamate respectively [21,22], must be carefully balanced to ensure optimal task specific corticospinal excitability. Indeed, the ratio between inhibition and facilitation is highly dynamic over time and is essential for determining whether cortical pyramidal neurons achieve action potential [22]. Therefore, both processes must be examined to make comprehensive inferences about cortical descending commands during standing. In standing, an increase in corticospinal excitability is accompanied by a decrease in GABAa inhibition [7,8] but GABAb inhibition has not been examined. Since GABAb is modulated independent of GABAa inhibition during upper extremity tasks [23], it is likely that they play different roles during motor control. GABAb receptor activation leads to more prolonged inhibition compared to GABAa, and GABAb activity may also influence glutamatergic neurons [22]. Therefore, it is important to examine both with respect to standing balance control. Additionally, the direction of change in glutamatergic intracortical facilitation in response to an increase in balance task difficulty remains uncertain given the equivocal findings from previous studies [8,24,25].
Also, little is known about the biomechanical correlates, and consequently the functional significance of cortical inhibition and facilitation for standing balance control. Previous studies employing anteroposterior (AP) or direction unspecific difficulty manipulations have found no association between cortical excitability and sway amplitude or velocity in young adults. As noted in section 1.1, the cortical contribution to balance control is expected to increase when task difficulty increases. Due to anatomical and biomechanical constraints, mediolateral (ML) control is inherently more complex than AP control [26], and indeed greater cortical activation is observed during ML compared to AP sway [27]. However, it is not known whether specific cortical neurophysiological processes, like inhibition and facilitation, are more closely associated with ML compared to AP sway dynamics.
3. effeCT of TAsk diffiCulTy oN CommoN NeurAl iNPuT To
mulTiPle musCles
In standing, balance is maintained by simultaneously activating multiple muscles in task specific combinations described as functional synergies [28,29]. Therefore, in addition to neural inputs to single muscles, it is important to examine the neural commands that
co-ordinate multiple muscles comprising a synergy. Such synergies may comprise agonists and antagonists which are co-activated to stiffen the limb. Alternatively, synergies may support reciprocal control where several agonists are co-activated to increase torque production but are not co-activated with antagonists. Such co-activation, whether supporting stiffness or reciprocal control, can be driven by common neural inputs to groups of muscles [30–33]. Reciprocal control allows for greater flexibility of responses to perturbations [34,35], and is indeed favored in functional synergies observed during standing [28,29]. However, it is not known whether common neural inputs favor stiffness or reciprocal control. Also, emerging evidence suggests that motor cortical areas contribute to common inputs in relatively difficult tasks [36,37], but the effect of task difficulty on such cortical common inputs has not been systematically examined. Additionally, it is suspected that the specific muscles pairs or groups, which receive common inputs in each task are determined by the biomechanical demands and configuration of each task.
4. iNdividuAl differeNCes iN NeurAl CoNTrol of sTANdiNg
bAlANCe
Even within a group of healthy young adults, there is a large variation in balance ca-pabilities[38] and likely in the underlying neural control strategies. One factor that can contribute to such individual differences is cognitive attributes like confidence and indeed, the association between confidence and balance capability is well documented [38–41]. However, to the best of our knowledge, the neural processes mediating the effect of confidence on muscle activation and postural sway, have not been examined.
Additionally, neural excitability, measured using transcranial magnetic stimulation (TMS), differs widely between individuals [42]. The recently introduced idea of intrinsic neural excitability [43,44] suggests that such differences are driven by factors like neurotransmit-ter concentration, synaptic strength etc. It is possible that such physiological differences are driven by individual attributes like previous motor experiences, which can induce neu-roplasticity in a manner similar to balance training [45–48]. Behaviorally, the reaction to in-creasing task difficulty is influenced by the amount of previous experience an individual has with motor tasks that challenge balance. Therefore, we suspect that differences in intrinsic excitability can influence the neural response to increasing task difficulty in standing. In this thesis we use excitability measured in a control task as an index of intrinsic excitability and examine how this intrinsic excitability influences the neural response to increasing task difficulty. Our findings were strengthened by reliability analyses which demonstrated that within individuals, excitability in both the control and more difficult tasks remains stable
14 Chapter 1
5. ComPlemeNTAry APProAChes To sTudyiNg sTANdiNg
bAlANCe CoNTrol
In this thesis cortical inputs to individual muscles are examined using transcranial magnetic stimulation (TMS), while inputs that coordinate multiple muscles are deduced using EMG-EMG coherence. These approaches complement each other by providing different types of information about the firing of spinal alpha motor neurons which finally drive muscle activation. The motor evoked potential (MEP) measured using TMS and coherence in the 6-15 Hz frequency range cannot distinguish between cortical and subcortical inputs to alpha motor neurons. However, outcomes described in the next paragraph allow us to examine the cortical contribution to lower extremity muscle activation in standing. Descending inputs from the primary motor cortex (M1) to alpha motor neurons can be measured using TMS. Specifically, the excitability of different M1 interneurons which use the various neurotransmitters described in section 1.2 can be measured. Short interval intracortical inhibition and long interval intracortical inhibition are indices of GABAa and GABAb activity, respectively [49]. Intracortical facilitation measures glutamate activity but may also be affected by other neurotransmitters [49]. On the other hand, coherence analysis allows us to examine commonalities in the firing pattern of alpha motor neurons targeting different muscles. Specifically, cortical inputs to alpha motor neurons drive the coherence between muscles in the higher 16-40 Hz range while coherence in the 0-5 Hz is driven by subcortical circuits [50,51]. In chapter 6 (sections 6.4 and 6.5) it is further discussed how this combination of techniques allowed us to make comprehensive infer-ences which could not be drawn from either technique individually.
6. Thesis Aims ANd ouTliNe
The different aspects of standing balance control examined in this thesis are depicted in Figure 1. The main objective was to examine the effect of task difficulty on the neural control of lower extremity muscles, with a focus on cortical inputs to muscles. In chapters 2 and 3, we used transcranial magnetic stimulation (TMS) to examine the neural inputs to single muscles and in chapter 4 we used EMG-EMG coherence to examine coordination of neural inputs to groups of muscles. With regards, to single muscles, the aim was to examine how M1 inhibitory and facilitatory processes contribute to the overall increase in corticospinal excitability, and to determine whether M1 excitability contributes to ML sway control. With regards to multiple lower extremity muscles, the aim was to deter-mine whether cortical common inputs support functional synergies favoring reciprocal or stiffness control. Additionally, we discuss how common inputs are organized to meet the
demands imposed by the biomechanical configuration of each task. Finally, we probed the neural basis for individual differences in neural control of standing balance using two approaches. In chapter 5, we examined how self-reported confidence influences the neural response to increasing task difficulty. Additionally, supplementary analysis of the data in chapter 3 provides preliminary clues about how intrinsic neural excitability influences the response to increasing task difficulty.
16 Chapter 1
7. refereNCes
[1] S.W. Mitchell, J.L. Morris, The tendon-jerk and muscle-jerk in disease, and especially in poste-rior sclerosis, Am. J. Med. Sci. 184 (1886) 363–372.
[2] W.R. Miles, Static equilibrium as a useful test of motor control, J Indust Hyg. 3 (1922) 316–331. [3] F.S. Fearing, The Factors Influencing Static Equilibrium. An Experimental Study of the Influence
of Height, Weight, and Position of the Feet on Amount of Sway, together with an Analysis of the Variability in the Records of One Reagent Over a Long Period of Time., J. Comp. Psychol. 4 (1924) 91.
[4] L.D. Bogle Thorbahn, R.A. Newton, Use of the Berg Balance Test to predict falls in elderly persons, Phys. Ther. 76 (1996) 576–583.
[5] L. Blum, N. Korner-Bitensky, Usefulness of the Berg Balance Scale in stroke rehabilitation: a systematic review, Phys. Ther. 88 (2008) 559–566.
[6] F.S. Fearing, The experimental study of the Romberg sign, J. Nerv. Ment. Dis. 61 (1925) 449–465.
[7] S. Papegaaij, W. Taube, M. Hogenhout, S. Baudry, T. Hortobágyi, Age-related decrease in motor cortical inhibition during standing under different sensory conditions., Front. Aging Neurosci. 6 (2014) 126.
[8] S. Papegaaij, S. Baudry, J. Négyesi, W. Taube, T. Hortobágyi, Intracortical inhibition in the soleus muscle is reduced during the control of upright standing in both young and old adults, Eur. J. Appl. Physiol. 116 (2016) 959–967.
[9] C.D. Tokuno, W. Taube, A.G. Cresswell, An enhanced level of motor cortical excitability during the control of human standing, Acta Physiol. 195 (2009) 385–395.
[10] T. Hülsdünker, A. Mierau, C. Neeb, H. Kleinöder, H.K. Strüder, Cortical processes associated with continuous balance control as revealed by EEG spectral power, Neurosci. Lett. 592 (2015) 1–5.
[11] T. Hülsdünker, A. Mierau, H.K. Strüder, Higher Balance Task Demands are Associated with an Increase in Individual Alpha Peak Frequency , Front. Hum. Neurosci. 9 (2016) 695.
[12] S. Baudry, J. Duchateau, Age-related influence of vision and proprioception on Ia presynaptic inhibition in soleus muscle during upright stance, J. Physiol. 590 (2012) 5541–5554.
[13] S. Baudry, F. Penzer, J. Duchateau, Input-output characteristics of soleus homonymous Ia af-ferents and corticospinal pathways during upright standing differ between young and elderly adults., Acta Physiol. (Oxf). 210 (2014) 667–77.
[14] K. Jahn, A. Deutschländer, T. Stephan, M. Strupp, M. Wiesmann, T. Brandt, Brain activation patterns during imagined stance and locomotion in functional magnetic resonance imaging, Neuroimage. 22 (2004) 1722–1731.
[15] A. Zwergal, J. Linn, G. Xiong, T. Brandt, M. Strupp, K. Jahn, Aging of human supraspinal locomotor and postural control in fMRI, Neurobiol. Aging. 33 (2012) 1073–1084.
[16] F. Malouin, C.L. Richards, P.L. Jackson, F. Dumas, J. Doyon, Brain activations during motor imagery of locomotor-related tasks: A PET study, Hum. Brain Mapp. 19 (2003) 47–62. [17] C.S. Sherrington, Flexion-reflex of the limb, crossed extension-reflex, and reflex stepping and
standing, J. Physiol. 40 (1910) 28–121.
[18] R. Magnus, Some results of studies in the physiology of posture, Lancet. 211 (1926) 531–536. [19] S. Slobounov, M. Hallett, S. Stanhope, H. Shibasaki, Role of cerebral cortex in human postural
[20] M.H. Trimble, D.M. Koceja, Effect of a Reduced Base of Support in Standing and Balance Training on the Soleus H-reflex, Int. J. Neurosci. 106 (2001) 1–20.
[21] G. Buzsáki, K. Kaila, M. Raichle, Inhibition and brain work., Neuron. 56 (2007) 771–83. [22] J.S. Isaacson, M. Scanziani, How inhibition shapes cortical activity, Neuron. 72 (2011) 231–243. [23] G.M. Opie, M.C. Ridding, J.G. Semmler, Age-related differences in pre- and post-synaptic
mo-tor cortex inhibition are task dependent, Brain Stimul. 8 (2015) 926–936.
[24] O. Soto, J. Valls-Solé, P. Shanahan, J. Rothwell, Reduction of intracortical inhibition in soleus muscle during postural activity., J. Neurophysiol. 96 (2006) 1711–7.
[25] S. Papegaaij, W. Taube, S. Baudry, E. Otten, T. Hortobágyi, Aging causes a reorganization of cortical and spinal control of posture, Front. Aging Neurosci. 6 (2014) 1–15.
[26] D.A. Winter, F. Prince, J.S. Frank, C. Powell, K.F. Zabjek, Unified theory regarding A/P and M/L balance in quiet stance, J. Neurophysiol. 75 (1996) 2334–2343.
[27] S. Slobounov, M. Hallett, C. Cao, K. Newell, Modulation of cortical activity as a result of voluntary postural sway direction: an EEG study., Neurosci. Lett. 442 (2008) 309–13.
[28] V. Krishnamoorthy, M.L. Latash, J.P. Scholz, V.M. Zatsiorsky, Muscle modes during shifts of the center of pressure by standing persons: Effect of instability and additional support, Exp. Brain Res. 157 (2004) 18–31.
[29] V. Krishnamoorthy, M.L. Latash, J.P. Scholz, V.M. Zatsiorsky, S. Goodman, V.M. Zatsiorsky, M.L. Latash, Muscle synergies during shifts of the center of pressure by standing persons: identifica-tion of muscle modes, Biol. Cybern. 89 (2003) 152–161.
[30] A. Danna-Dos-Santos, A.M. Degani, T.W. Boonstra, L. Mochizuki, A.M. Harney, M.M. Sch-meckpeper, L.C. Tabor, C.T. Leonard, The influence of visual information on multi-muscle control during quiet stance: a spectral analysis approach, Exp. Brain Res. 233 (2015) 657–669. [31] H. Obata, M.O. Abe, K. Masani, K. Nakazawa, Modulation between bilateral legs and within
unilateral muscle synergists of postural muscle activity changes with development and aging, Exp. Brain Res. 232 (2014) 1–11.
[32] A.M. Degani, C.T. Leonard, A. Danna-dos-Santos, The use of intermuscular coherence analysis as a novel approach to detect age-related changes on postural muscle synergy, Neurosci. Lett. 656 (2017) 108–113.
[33] X. García-Massó, M. Pellicer-Chenoll, L.M. Gonzalez, J.L. Toca-Herrera, The difficulty of the postural control task affects multi-muscle control during quiet standing, Exp. Brain Res. 234 (2016) 1977–1986.
[34] C. Grüneberg, B.R. Bloem, F. Honegger, J.H.J. Allum, The influence of artificially increased hip and trunk stiffness on balance control in man, Exp. Brain Res. 157 (2004) 472–485.
[35] N.P. Reeves, K.S. Narendra, J. Cholewicki, Spine stability: the six blind men and the elephant, Clin. Biomech. 22 (2007) 266–274.
[36] T. Watanabe, K. Saito, K. Ishida, S. Tanabe, I. Nojima, Coordination of plantar flexor muscles during bipedal and unipedal stances in young and elderly adults, Exp. Brain Res. (2018) 1–11. [37] T. Watanabe, K. Saito, K. Ishida, S. Tanabe, I. Nojima, Age-Related Declines in the Ability to
Modulate Common Input to Bilateral and Unilateral Plantar Flexors During Forward Postural Lean, Front. Hum. Neurosci. 12 (2018) 254.
[38] H. Kiers, J. van Dieën, H. Dekkers, H. Wittink, L. Vanhees, A systematic review of the relation-ship between physical activities in sports or daily life and postural sway in upright stance, Sport. Med. 43 (2013) 1171–1189.
community-18 Chapter 1
[40] K.M.C. Talley, J.F. Wyman, C.R. Gross, Psychometric properties of the activities-specific balance confidence scale and the survey of activities and fear of falling in older women, J. Am. Geriatr. Soc. 56 (2008) 328–333.
[41] S. Schepens, A. Goldberg, M. Wallace, The short version of the Activities-specific Balance Confidence (ABC) scale: its validity, reliability, and relationship to balance impairment and falls in older adults, Arch. Gerontol. Geriatr. 51 (2010) 9–12.
[42] M. Orth, A.H. Snijders, J.C. Rothwell, The variability of intracortical inhibition and facilitation, Clin. Neurophysiol. 114 (2003) 2362–2369.
[43] I. Greenhouse, M. King, S. Noah, R.J. Maddock, R.B. Ivry, Individual Differences in Resting Corticospinal Excitability Are Correlated with Reaction Time and GABA Content in Motor Cortex, J. Neurosci. 37 (2017) 2686–2696.
[44] T. Fedele, E. Blagovechtchenski, M. Nazarova, Z. Iscan, V. Moiseeva, V. V Nikulin, Long-range temporal correlations in the amplitude of alpha oscillations predict and reflect strength of intracortical facilitation: combined TMS and EEG study, Neuroscience. 331 (2016) 109–119. [45] M. Schubert, S. Beck, W. Taube, Balance training and ballistic strength training are associated
with task-specific corticospinal adaptations, Eur. J. Neurosci. 27 (2008) 2007–2018.
[46] S. Beck, W. Taube, M. Gruber, F. Amtage, a Gollhofer, M. Schubert, Task-specific changes in motor evoked potentials of lower limb muscles after different training interventions., Brain Res. 1179 (2007) 51–60.
[47] W. Taube, M. Gruber, S. Beck, M. Faist, a. Gollhofer, M. Schubert, Cortical and spinal ad-aptations induced by balance training: correlation between stance stability and corticospinal activation, Acta Physiol. 189 (2007) 347–358.
[48] W. Taube, M. Gruber, A. Gollhofer, Spinal and supraspinal adaptations associated with balance training and their functional relevance, Acta Physiol. 193 (2008) 101–116.
[49] P.M. Rossini, D. Burke, R. Chen, L.G. Cohen, Z. Daskalakis, R. Di Iorio, V. Di Lazzaro, F. Ferreri, P.B. Fitzgerald, M.S. George, Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: basic principles and procedures for routine clinical and research application. An updated report from an IFCN Committee, Clin. Neurophysiol. 126 (2015) 1071–1107.
[50] P. Grosse, M.J. Cassidy, P. Brown, EEG–EMG, MEG–EMG and EMG–EMG frequency analysis: physiological principles and clinical applications, Clin. Neurophysiol. 113 (2002) 1523–1531. [51] J.A. Norton, M.A. Gorassini, Changes in cortically related intermuscular coherence
accompa-nying improvements in locomotor skills in incomplete spinal cord injury, J. Neurophysiol. 95 (2006) 2580–2589.
Chapter 2
Increasing mediolateral
standing sway is associated
with increasing corticospinal
excitability, and decreasing M1
inhibition and facilitation
Tulika Nandia,b, Beth E. Fishera, Tibor Hortobágyib, George J. Salema
a Division of Biokinesiology and Physical Therapy,
University of Southern California, Los Angeles, CA, United States
b Center for Human Movement Sciences,
University of Groningen, University Medical Center Groningen, Groningen, The Netherlands
Chapter 2
22
AbsTrACT
In standing, corticospinal excitability increases and primary motor cortex (M1) inhibition decreases in response to anterior posterior or direction unspecific manipulations that increase task difficulty. However, mediolateral (ML) sway control requires greater active neural involvement. Therefore, the primary purpose of this study was to determine the pattern of change in neural excitability when ML postural task difficulty is manipulated and to test whether the neural excitability is proportional to ML sway magnitude across conditions. Tibialis anterior corticospinal excitability was quantified using motor evoked potential (MEP) and postural sway was indexed using ML center of pressure (COP) veloc-ity. Additionally, we examined inhibition and facilitation processes in the primary motor cortex using the paired pulse short interval intracortical inhibition (SICI) and intracortical facilitation (ICF) techniques respectively. Measurements were repeated in four conditions with quiet stance as a control. Differences between conditions were tested using one-way repeated measures ANOVAs, on log transformed data. Associations were quantified using Spearman’s Rank Correlation Coefficient. There was a significant main effect of condition on all the neural excitability measures with MEP (p < 0.001) being highest in the most difficult condition, and SICI (p = 0.01), ICF (p < 0.001) being lowest in the most difficult condition. Increasing ML COP velocity was significantly associated with increasing MEP amplitude (rho = 0.68, p < 0.001), but decreasing SICI (rho = 0.24, p =0.03) and ICF (rho = −0.54, p < 0.001). Our results show that both corticospinal and M1 excitability in standing are scaled in proportion to ML task difficulty.
1. iNTroduCTioN
During unperturbed standing, the body’s center of mass (COM) sways spontaneously, but remains within the base of support (BOS) as active and passive mechanisms generate appropriate counteractive forces and regulate the center of pressure (COP) path. Within limits, COP movements during quiet standing and larger fluctuations induced by decreas-ing the BOS size are considered normal in healthy adults [1–3]. However, excessive COP movements, especially in the mediolateral (ML) direction, predict falls in old adults [4–6] and are characteristic of individuals with Parkinson’s disease [7] and cerebellar deficits [8]. Also, modulation of both voltage and time-frequency based EEG measures is more pronounced during ML compared to AP sway [9]. Therefore, it is clear that AP and ML sway components are controlled using independent strategies [10] and ML control likely requires greater neural resources [9]. However, studies examining neural excitability in standing have employed manipulations only in the AP direction or without directional specificity [11–15]. Therefore, in this study we examined neural excitability changes in response to manipulation of ML difficulty indexed using COP movements.
While segmental inputs to postural control are well established [14,15], there is now evidence that the primary motor cortex (M1) also contributes to the control of sway dur-ing unperturbed, quiet standdur-ing [11,13,16] and more complex postural tasks [17–19]. Transcranial magnetic stimulation (TMS) studies demonstrate that when experimental ma-nipulations increase COP movements in standing, corticospinal excitability (CSE) increases and M1 inhibition decreases [11–13]. The overall increase in excitability can drive muscle activation and thereby create greater forces to counteract COM movements. CSE is a net outcome of multiple inhibitory and facilitatory neurophysiological processes, at various anatomical locations and an increase could be mediated by alteration in spinal excitability. The decrease in M1 inhibition provides more convincing evidence for cortical involvement in postural control. Therefore, in this study we examined two cortex specific measures of excitability, along with CSE. Also, it is unclear whether neural excitability is scaled in proportion to the magnitude of sway. Moderate correlations between AP or resultant COP movements, and H-reflex [15] or M1 inhibition [11] have been reported in older adults, but not in young.
We measured neural excitability using TMS and manipulated ML sway by altering the BOS and/or foot support, in young adults. Preliminary work confirmed that the selected manipulations systematically increased ML COP velocity, indicating an increase in difficulty. We tested the tibialis anterior (TA) because it is a primary ankle invertor which is essential for maintenance of ML balance. In fact, its contribution to ML control increases as BOS
Chapter 2
24
ability in four standing conditions with increasing ML COP velocity, and to test whether excitability is correlated with COP velocity. We hypothesized that an increase in task dif-ficulty would lead to increase in CSE and M1 facilitation and decrease in M1 inhibition, in proportion to the increase in ML COP velocity.
2. exPerimeNTAl ProCedures
2.1. Participants
Healthy adults (Age: 25.7 ± 4.2 years; 11 females, 9 males) volunteered to participate in the study. Three individuals were unable to complete the entire data collection due to fatigue and/or discomfort from prolonged stimulation. The number of participants included in each analysis is indicated in the tables and figures. Participants were excluded if they re-ported ongoing symptoms due to lower extremity injury, a history of neurological disorders, seizures, head trauma or unexplained loss of consciousness; or if they were pregnant; had metal implants or pacemakers; used medication known to lower seizure threshold or had blood relatives with a history of seizures. Written informed consent was obtained and all procedures were conducted in accordance with the Declaration of Helsinki. The study was approved by the Institutional Review Board of the University of Southern California, Health Sciences Campus. Foot dominance was determined using a 3-question inventory [21].
2.2. data acquisition
Electromyographic (EMG) signals were recorded from the dominant-side TA using a bipolar surface electrode (radius 12 mm, inter electrode distance 17 mm, Motion Lab Systems, Ba-ton Rouge, LA) which was aligned parallel to the muscle fibers and placed over the bulk of the muscle belly, located by palpating during voluntary contraction. The ground electrode was placed on the anterior tibial surface. Data were acquired at 15 kHz in order to simul-taneously capture the TMS pulse, and stored using Signal software (Signal v6, Cambridge Electronic Design Ltd, Cambridge UK). COP data were sampled at 1.5 kHz using AMTI force platforms (Model #OR6-6-1, Watertown, MA) embedded into the laboratory floor. Data were acquired and stored using Qualisys software (Qualisys Inc., Gothenburg, Sweden). TMS pulses were delivered using a double cone coil (110 mm) connected to a BiStim module (The Magstim Co., Whitland, UK) and two single-pulse magnetic stimulators (Mag-stim Model 2002). A lycra cap marked with a 1 cm grid ensured consistent manual coil positioning, and the current was directed posterior-to-anterior. The hotspot was defined as the location where the largest and most consistent motor evoked potentials (MEP) were obtained, and was located over the midline or 1–2 cm lateral in all participants. Active motor threshold (MT) was determined in standing by systematically varying the stimulation
intensity to fi nd the lowest stimulator output at which 3 out of 5 MEPs had peak-to-peak amplitude of at least 100 µV [11]. The average threshold was 56 ± 13% of maximal stimulator output. For measuring short interval intracortical inhibition (SICI) and intracorti-cal facilitation (ICF), a subthreshold (80% MT) conditioning pulse was applied followed by a supra-threshold (120% MT) test pulse, separated by inter-stimulus intervals (ISI) of 3 and 13 ms respectively. Ten paired pulses each for SICI and ICF protocols, and 10 single pulses at 120% MT were applied in random order, with a minimum of 5 s between stimuli.
2.3. Procedures
All data were acquired during a 2.5 h long lab visit. In pilot testing we examined several methods for eliciting TA maximal voluntary contraction (MVC) – isometric dorsifl exion and inversion with manual resistance in standing and sitting; participants standing on one foot with heel resting on the support surface, hands rested on a stable surface for balance and instructed to “raise their toes as high as possible” for 3 s. Participants inverted and dorsifl exed the foot during the latter and MVC was consistently higher than that obtained using the former. The procedure was repeated 3 times.
TMS and COP measurements were acquired with participants standing on a force plate. Four postural conditions were examined (listed from least to greatest postural diffi culty): 1) Standing on 2 feet, feet shoulder width apart (i.e., wide base; 2WB); 2) Standing on 2 feet, feet as close together as possible (i.e., narrow base; 2NB); 3) Standing with dominant foot on the fl oor (stance limb) and other foot on a solid block,∼30 cm high (1Step), and 4) Standing with dominant foot on the fl oor (stance limb) and other foot on an unstable spring (stiffness – 49.04 N/cm), ∼30 cm high (1Spring) (Fig. 1). Post-hoc analysis confi rmed that in 1Step and 1Spring, majority of the body weight i.e. 90.7 ± 3.6 and 94.5 ± 2.2% respectively was supported on the stance limb. The order of conditions was randomized
D) 1SPRING C) 1STEP
A) 2WB
Quiet Stance Narrow BaseA) 2NB
figure 1 Four conditions were used to manipulate
mediolateral center of pressure (ML COP) velocity: A) wide base with feet shoulder width apart (quiet stance; 2WB); B) narrow base condition with feet as close to-gether as possible (2NB); C) one foot supported on a stable block with >80% body weight on lower, domi-nant foot (1Step); D) one foot supported on an unsta-ble spring with >80% body weight on lower, dominant foot (1Spring).
Chapter 2
26
across participants, with 3–5 min of rest between conditions. For 2WB, the foot position was marked and maintained throughout testing. Participants were verbally instructed to stand as still as possible.
2.4. data analysis
EMG and COP data were processed using custom Matlab (The Mathworks, Natick, MA) codes. EMG data were down-sampled to 3.0 kHz, bandpass filtered using a 4th order Butterworth filter with 10 Hz and 1 kHz low pass and high pass cut off respectively, and rectified. We estimated background EMG (bEMG) from the mean voltage measured in a 100 ms window before the TMS stimulation artifact. For MVC trials, moving average with a 100 ms window was used to smooth the data, the peak was measured and the highest of 3 trials was selected.
COP data were filtered using a 4th order low pass Butterworth filter with 10 Hz cut off. ML velocity was calculated in a 2 s window before each TMS pulse, and averaged across windows to obtain a single estimate for each condition. Velocity was selected because it is predictive of falls [5,22] and more reliable than amplitude or area based measures [23,24]. Post-hoc analysis determined that participants were swaying laterally in the 1 s preceding TMS pulse application in approximately half the trials – 55 ± 10, 55 ± 7, 57 ± 9 and 55 ± 7% in 2WB, 2NB, 1Step and 1Spring respectively. Finally, directional bias of the manipula-tions was confirmed by comparing AP and ML COP velocity. The percentage increase from 2WB to 1Spring was much higher for ML (293%) than AP (203%) velocity.
Initial processing of the TMS data was done using Signal software (Signal v6, Cambridge Electronic Design Ltd, Cambridge UK). In each trial, the MEP amplitude was defined as the peak to peak voltage within a 25–65 ms window after application of the magnetic pulse. Visual inspection was used to ensure that the MEP was within this window. Sub-sequent processing was performed using Matlab (The Mathworks, Natick, MA). Peak to peak amplitudes from 10 single or paired pulse trials were averaged to estimate the test and conditioned MEPs respectively. The test MEP amplitude was used as an index of CSE. The following formulae were used to quantify SICI and ICF, using MEPs averaged over 10 trials –
1. SICI= (ConditionedSICI/Test MEP *100); smaller values indicate greater inhibition. 2. ICF = (ConditionedICF/Test MEP *100); higher values indicate greater facilitation.
2.5. statistical analyses
SPSS (Version 22, IBM Corp., Armonk, NY) software was used. When the Shapiro-Wilk test revealed that variables were not normally distributed in one or more conditions, values were log transformed for the ANOVA. Five one-way univariate ANOVAs, with condition as
a fixed factor, were used to test for effect of condition on bEMG, COP velocity, MEP, SICI and ICF. When the F test was significant, pairwise differences were tested using Tukey’s post hoc comparisons. Linear regression analyses with condition and bEMG as indepen-dent variables was used to test whether bEMG predicts MEP, SICI and ICF, in addition to condition. Spearman’s correlation coefficient was used to test for the associations between COP velocity and TMS measures, with data pooled across the four conditions. Curve fitting was applied to determine whether the variance in the data was better explained by linear or quadratic association. The significance level was set at 0.05.
3. resulTs
3.1. ml CoP velocity
Main effect of condition on velocity (p < 0.001) was significant. Velocity in all conditions was significantly higher than 2WB (p < 0.05, Table 1), with highest velocity in 1Spring.
3.2. bemg and Tms measures
Main effect of condition on bEMG (p < 0.001), MEP (p < 0.001), SICI (p= 0.01) and ICF (p < 0.001) was significant. In the most difficult condition – 1Spring, bEMG and MEP amplitude were highest, SICI and ICF were lowest (Table 1). Regression analyses revealed that condition was a significant predictor of MEP (p < 0.001), SICI (p= 0.04) and ICF (p= 0.03), while bEMG was a significant predictor of ICF (p =0.02) but not MEP (p = 0.72) and SICI (p= 0.21). MVC values were 0.36 ± 0.13 mV.
Table 1 Mediolateral center of pressure (ML COP) velocity, background EMG (bEMG), motor evoked
potential (MEP), short interval intracortical inhibition (SICI) and intracortical facilitation (ICF) in each condition: mean ± standard deviation.
f-value 2Wb 2Nb 1step 1spring
bEMG 13.72 2.91 ± 1.19 3.34 ± 1.81% 4.27 ± 2.12* 6.47 ± 2.31* ML COP velocity (cm/s) 81.92 0.43 ± 0.12 (n=20) 0.63 ± 0.12* (n=20) 0.80 ± 0.37* (n=19) 1.69 ± 0.43* (n=19) MEP (mV) 19.12 0.39 ± 0.20 (n=20) 0.56 ± 0.29 (n=20) 0.83 ± 0.40* (n=19) 1.14 ± 0.46* (n=19) SICI (%) 4.63 58.42 ± 25.51 (n=20) 59.51 ± 15.69 (n=20) 66.40 ± 15.25 (n=19) 75.13 ± 17.79* (n=19) ICF (%) 6.73 196.66 ± 110.60 (n=19) 171.33 ± 69.46 (n=19) 127.86 ± 22.73* (n=18) 116.95 ± 17.32* (n=17) * indicates significantly different from 2WB using post-hoc Tukey’s test (p<0.05)
Chapter 2
28
ML COP velocity was significantly linearly correlated with MEP (p < 0.001, rho= 0.68), SICI (p= 0.03, rho= 0.24) and ICF (p < 0.001, rho= −0.54) (Fig. 2). The association between ML COP velocity and SICI was best explained by a linear fit, while the associations with MEP and ICF were better explained by a quadratic fit. For MEP, quadratic (R2 =0.39, p < 0.001) and linear (R2 =0.35, p < 0.001) fits explained a similar percentage of variance. However, for ICF the quadratic fit (R2= 0.25, p < 0.001) explained a higher percentage of variance than the linear (R2 =0.15, p = 0.001) fit.
0 0.5 1 1.5 2 2.5 3 ML COP velocity (cm/s) 20 40 60 80 100 120 140 SICI (%)
B) Short Interval Intracortical Inhibition
rho=0.27 n=78 0 0.5 1 1.5 2 2.5 3 0 100 200 300 400 500 600 ICF (%) C) Intracortical Facilitation rho=-0.54 n=73 ML COP velocity (cm/s) 0 0.5 1 1.5 2 2.5 3 ML COP velocity (cm/s) 0 0.5 1 1.5 2 2.5 MEP (mV)
A) Motor Evoked Potential
rho=0.68 n=78
figure 2 Correlations between neural excitability and mediolateral center of pressure (ML COP) velocity; A)
motor evoked potential - ML COP velocity (p<0.001); B) short interval intracortical inhibition - ML COP velocity (p=0.03); C) intracortical facilitation - ML COP velocity ( p<0.001)
4. disCussioN
In partial agreement with our hypothesis, TA CSE increased and M1 inhibition decreased when postural task difficulty was manipulated by altering the BOS and foot support, in standing. Contrary to our expectation, M1 facilitation also decreased with increase in task difficulty. All three measures of neural excitability were correlated with ML COP velocity – i.e. excitability was scaled in proportion to postural sway.
4.1. increase in Cse, decrease in m1 inhibition and facilitation as
standing task difficulty increases
Independent of the direction of manipulation, reports suggest a consistent pattern of increasing CSE and decreasing M1 inhibition as postural task difficulty increases [11–13]. However, there are conflicting reports regarding M1 facilitation – ICF is lower in standing compared to sitting [16], but similar across conditions when sensory feedback or anterior trunk support are manipulated in standing [11,12]. As difficulty increases, greater internal forces must be generated so that COP shifts are sufficient and appropriate to ensure that the COM remains within the BOS. Theoretically, this can be achieved by increasing neural excitability. Therefore, our finding of simultaneous decrease in M1 inhibition and facilitation presents a conundrum. Specifically, the decrease in facilitation appears to be counterpro-ductive to the overall goal of increasing CSE. Parallel changes in excitatory and inhibitory activity have previously been observed in the upper extremity during the preparatory phase of reaction time tasks [25,26]. It has been proposed that these competing processes ensure that neural excitability is suitable for executing appropriate motor patterns, while inap-propriate movements are withheld [27]. In other words, each neurophysiological process plays a distinct role which is essential for successfully achieving the task goal. We propose that a similar principle applies to postural control – i.e. in difficult conditions low inhibition can increase TA activation and ensure that the mechanical demands of the posture are satisfied. Additionally, it ensures that the TA is ready to be activated in case of a perturba-tion. On the other hand, low facilitation prevents unnecessary muscle activity that could interfere with ongoing maintenance of balance. This becomes more critical when difficulty increases and inadvertent muscle activity could create self-generated perturbations with the potential to compromise balance. However, as evidenced by the quadratic association, ICF plateaus when difficulty exceeds a certain threshold. It is possible that in the most difficult postures the need to maintain high CSE outweighs the need to be cautious and prevent movement. This may explain why some postural control studies found changes in SICI without concurrent changes in ICF, depending on the conditions employed in a specific study.
Chapter 2
30
4.2. Correlations between CoP velocity and neural excitability
In line with the previously reported weak or insignificant correlations between AP COP and H-relfex, MEP or SICI in young adults [11,15,28], we found statistically significant but low to moderate strength associations between neural excitability and ML COP velocity. Though bEMG differed between conditions, regression analysis revealed that bEMG was a significant predictor only for ICF. Our initial hypothesis was based on the assumption that neural excitability determines the level of muscle activation which in turn drives COP adjustments. However, given that condition is a significant predictor of MEP and SICI, but bEMG is not, we hypothesize that TA excitability changes serve other roles, in addition to increasing ongoing contraction and COP movements.
Neural excitability changes without concomitant changes in muscle contraction [29,30], can reflect subliminal or subthreshold increases in excitability that prepares muscles for context-specific activation, but does not influence ongoing contraction. We hypothesize that the changes in TA excitability ensure that it is appropriate for the new context where perturbations pose a higher risk, and not for the sole purpose of influencing COP move-ments. Though not intended as such, in our study, the TMS pulses constituted a mechanical perturbation which was expected, but had unpredictable temporal spacing. Since partici-pants were explicitly instructed to stay as still as possible, it is likely that the motor control system initiated measures to prepare for the perturbation in advance. Additionally, in the context of postural control, there is an inherent instinct to avoid a fall and potential injury. Previous studies compared conditions in which a TMS pulse [31,32] or mechanical change [33,34] induced a perturbation and participants were instructed to either assist/disregard or resist it. In agreement with our study, CSE increased and M1 inhibition decreased when participants were asked to resist the perturbation. The simultaneous decrease in SICI and ICF also reinforces the hypothesis that M1 excitability in more difficult conditions is aimed at achieving various motor goals, besides maintaining COP movements. Both increasing COP movements to minimize COM movements, and preparing to respond to external perturbations require an increase in neural excitability. On the other hand, excessive excitability poses a risk of creating self-initiated perturbations. Therefore, the different neurophysiological processes must be optimally co-varied for effective postural control.
4.3. limitations
It has been reported that TA MEPs are smaller during forward compared to backward sway [13] and it is possible that excitability differs between medial and lateral sway. In our study, the proportion of trials with medial and lateral sway were approximately equal in all conditions. Controlling for sway direction is likely to decrease inter-trial variability within each condition and may even amplify the differences between conditions.
A limitation of using TMS to study postural control is that the coil could have a stabilizing effect and consequently affect neural excitability. Indeed, in a subset of participants we found higher COP velocity without the coil and addressed this issue by using velocity measured with the coil for all analyses.
The parallel decrease in SICI and ICF, despite low bEMG suggests that TA neural excitability is modulated in preparation for perturbations or a potential loss of balance, rather than for ongoing sway control. Though the TMS pulse acted as a perturbation, we were unable to quantify and comment on the quality of the biomechanical response. Therefore, direct be-havioral consequences of the changes in excitability remain unclear. Further investigations are required to determine whether this pattern of covariation in SICI and ICF translates into better postural performance and response to perturbations.
5. CoNClusioNs
Our study demonstrates that TA CSE increases, and M1 inhibition and facilitation decrease in proportion to ML COP velocity when postural task difficulty increases in standing. It confirms that the general pattern of neural excitability modulation is consistent across AP and ML postural manipulations. Additionally, it adds to the limited evidence that in stand-ing a part of the variability in neural excitability is explained by postural sway magnitude.
Chapter 2
32
6. refereNCes
[1] I.G. Amiridis, V. Hatzitaki, F. Arabatzi, Age-induced modifications of static postural control in humans, Neurosci. Lett. 350 (2003) 137–140.
[2] N. Benjuya, I. Melzer, J. Kaplanski, Aging-induced shifts from a reliance on sensory input to muscle cocontraction during balanced standing, Journals Gerontol. A, Biol. Sci. Med. Sci. 59 (2004) 166–171.
[3] T. Lemos, L.A. Imbiriba, C.D. Vargas, T.M. Vieira, Modulation of tibialis anterior muscle activity changes with upright stance width, J. Electromyogr. Kinesiol. 25 (2015) 168–174.
[4] B.E. Maki, P.J. Holliday, A.K. Topper, A Prospective Study of Postural Balance and Risk of Falling in An Ambulatory and Independent Elderly Population, J. Gerontol. 49 (1994) M72–M84. [5] M. Piirtola, P. Era, Force platform measurements as predictors of falls among older people - A
review, Gerontology. 52 (2006) 1–16.
[6] M.J. Hilliard, K.M. Martinez, I. Janssen, B. Edwards, M.L. Mille, Y. Zhang, M.W. Rogers, Lateral Balance Factors Predict Future Falls in Community-Living Older Adults, Arch. Phys. Med. Reha-bil. 89 (2008) 1708–1713.
[7] S.L. Mitchell, J.J. Collin, C.J. De Luca, A. Burrows, L.A. Lipsitz, Open-loop and closed-loop postural control mechanisms in Parkinson’s disease: increased mediolateral activity during quiet standing, Neurosci. Lett. 197 (1995) 133–136.
[8] P. Gatev, S. Thomas, J. Lou, M. Lim, M. Hallett, Effects of diminished and conflicting sensory information on balance in patients with cerebellar deficits, Mov. Disord. 11 (1996) 654–664. [9] S. Slobounov, M. Hallett, C. Cao, K. Newell, Modulation of cortical activity as a result of
voluntary postural sway direction: an EEG study., Neurosci. Lett. 442 (2008) 309–13.
[10] D.A. Winter, F. Prince, J.S. Frank, C. Powell, K.F. Zabjek, Unified theory regarding A/P and M/L balance in quiet stance, J. Neurophysiol. 75 (1996) 2334–2343.
[11] S. Papegaaij, W. Taube, M. Hogenhout, S. Baudry, T. Hortobágyi, Age-related decrease in motor cortical inhibition during standing under different sensory conditions., Front. Aging Neurosci. 6 (2014) 126.
[12] S. Papegaaij, S. Baudry, J. Négyesi, W. Taube, T. Hortobágyi, Intracortical inhibition in the soleus muscle is reduced during the control of upright standing in both young and old adults, Eur. J. Appl. Physiol. 116 (2016) 959–967.
[13] C.D. Tokuno, W. Taube, A.G. Cresswell, An enhanced level of motor cortical excitability during the control of human standing, Acta Physiol. 195 (2009) 385–395.
[14] C.D. Tokuno, M.G. Carpenter, A. Thorstensson, S.J. Garland, a G. Cresswell, Control of the triceps surae during the postural sway of quiet standing., Acta Physiol. (Oxf). 191 (2007) 229–36.
[15] S. Baudry, J. Duchateau, Age-related influence of vision and proprioception on Ia presynaptic inhibition in soleus muscle during upright stance, J. Physiol. 590 (2012) 5541–5554.
[16] O. Soto, J. Valls-Solé, P. Shanahan, J. Rothwell, Reduction of intracortical inhibition in soleus muscle during postural activity., J. Neurophysiol. 96 (2006) 1711–7.
[17] W. Taube, M. Schubert, M. Gruber, S. Beck, M. Faist, A. Gollhofer, Direct corticospinal path-ways contribute to neuromuscular control of perturbed stance., J. Appl. Physiol. 101 (2006) 420–9.
[18] J. V. Jacobs, K. Fujiwara, H. Tomita, N. Furune, K. Kunita, F.B. Horak, Changes in the activity of the cerebral cortex relate to postural response modification when warned of a perturbation, Clin. Neurophysiol. 119 (2008) 1431–1442.
[19] T.H. Petersen, K. Rosenberg, N.C. Petersen, J.B. Nielsen, Cortical involvement in anticipatory postural reactions in man., Exp. Brain Res. 193 (2009) 161–71.
[20] S. Sozzi, J.L. Honeine, M.C. Do, M. Schieppati, Leg muscle activity during tandem stance and the control of body balance in the frontal plane, Clin. Neurophysiol. 124 (2013) 1175–1186. [21] G. V Hebbal, V.R. Mysorekar, Evaluation of some tasks used for specifying handedness and
footedness, Percept. Mot. Skills. 102 (2006) 163–164.
[22] I. Melzer, N. Benjuya, J. Kaplanski, Postural stability in the elderly: A comparison between fallers and non-fallers, Age Ageing. 33 (2004) 602–607.
[23] A. Ruhe, R. Fejer, B. Walker, The test-retest reliability of centre of pressure measures in bipedal static task conditions - A systematic review of the literature, Gait Posture. 32 (2010) 436–445. [24] D. Lin, H. Seol, M.A. Nussbaum, M.L. Madigan, Reliability of COP-based postural sway
mea-sures and age-related differences, Gait Posture. 28 (2008) 337–342.
[25] K. Davranche, C. Tandonnet, B. Burle, C. Meynier, F. Vidal, T. Hasbroucq, The dual nature of time preparation: Neural activation and suppression revealed by transcranial magnetic stimula-tion of the motor cortex, Eur. J. Neurosci. 25 (2007) 3766–3774.
[26] C. Sinclair, G.R. Hammond, Excitatory and inhibitory processes in primary motor cortex during the foreperiod of a warned reaction time task are unrelated to response expectancy, Exp. Brain Res. 194 (2009) 103–113.
[27] J. Duque, L. Labruna, C. Cazares, R.B. Ivry, Dissociating the influence of response selection and task anticipation on corticospinal suppression during response preparation, Neuropsychologia. 65 (2014) 287–296.
[28] S. Baudry, S. Collignon, J. Duchateau, Influence of age and posture on spinal and corticospinal excitability, Exp. Gerontol. 69 (2015) 62–69.
[29] S.S. Kantak, G.F. Wittenberg, W.-W. Liao, L.S. Magder, M.W. Rogers, S.M. Waller, Posture-related modulations in motor cortical excitability of the proximal and distal arm muscles., Neurosci. Lett. 533 (2013) 65–70.
[30] H. Obata, H. Sekiguchi, T. Ohtsuki, K. Nakazawa, Posture-related modulation of cortical excit-ability in the tibialis anterior muscle in humans, Brain Res. 1577 (2014) 29–35.
[31] M. Camus, J. Pailhous, M. Bonnard, On-line flexibility of the cognitive tuning of corticospinal excitability: A TMS study in human gait, Brain Res. 1076 (2006) 144–149.
[32] M. Bonnard, L. Spieser, H.B. Meziane, J.B. De Graaf, J. Pailhous, Prior intention can locally tune inhibitory processes in the primary motor cortex: Direct evidence from combined TMS-EEG, Eur. J. Neurosci. 30 (2009) 913–923.
[33] S.G. Sangani, H.A. Raptis, A.G. Feldman, Subthreshold corticospinal control of anticipatory actions in humans, Behav. Brain Res. 224 (2011) 145–154.
[34] H.B. Meziane, L. Spieser, J. Pailhous, M. Bonnard, Corticospinal control of wrist muscles during expectation of a motor perturbation: A transcranial magnetic stimulation study, Behav. Brain Res. 198 (2009) 459–465.
Chapter 3
In Standing, Corticospinal
Excitability Is Proportional
to COP Velocity Whereas
M1 Excitability Is
Participant-Specific
Tulika Nandia,b, Claudine J. C. Lamotha, Helco G. van Keekena, Lisanne B. M. Bakkera, Iris Koka, George J. Salemb, Beth E. Fisherb and Tibor Hortobágyia
a Center for Human Movement Sciences,
University of Groningen, University Medical Center Groningen, Groningen, The Netherlands
bDivision of Biokinesiology and Physical Therapy,
University of Southern California, Los Angeles, CA, United States
Chapter 3
36
AbsTrACT
Reductions in the base of support (BOS) make standing difficult and require adjustments in the neural control of sway. In healthy young adults, we determined the effects of reductions in mediolateral (ML) BOS on peroneus longus (PL) motor evoked potential (MEP), intracortical facilitation (ICF), short interval intracortical inhibition (SICI) and long interval intracortical inhibition (LICI) using transcranial magnetic stimulation (TMS). We also examined whether participant-specific neural excitability influences the responses to increasing standing difficulty. Repeated measures ANOVA revealed that with increasing standing difficulty MEP size increased, SICI decreased (both p < 0.05) and ICF trended to decrease (p = 0.07). LICI decreased only in a sub-set of participants, demonstrating atypical facilitation. Spearman’s Rank Correlation showed a relationship of rho = 0.50 (p = 0.001) between MEP size and ML center of pressure (COP) velocity. Measures of M1 excitability did not correlate with COP velocity. LICI and ICF measured in the control task correlated with changes in LICI and ICF, i.e., the magnitude of response to increasing standing dif-ficulty. Therefore, corticospinal excitability as measured by MEP size contributes to ML sway control while cortical facilitation and inhibition are likely involved in other aspects of sway control while standing. Additionally, neural excitability in standing is determined by an interaction between task difficulty and participant-specific neural excitability.
1. iNTroduCTioN
Mechanical challenges and sensory manipulations of standing balance increase the spon-taneous movements of the center of mass [1,2]. Reductions in the base of support (BOS) make it difficult to maintain balance and require adjustments in the neural control of center of pressure (COP). In response to manipulations that challenge standing balance, fronto-parietal alpha and theta EEG power increases, indicating an increase in cortical activity [3,4]. Also, corticospinal excitability and primary motor cortex (M1) inhibition measured by transcranial magnetic stimulation (TMS), increases and decreases respectively [5–9]. Presumably such neural adjustments tune muscle contractions, adjust COP dynamics and consequently center of mass sway, thereby ensuring that balance is maintained. In contrast to anteroposterior (AP) and direction non-specific manipulations, we recently demonstrated that mediolateral (ML) manipulations of BOS produce correlated changes in the neural excitability of the tibialis anterior (TA) and COP velocity in young adults [8]. These findings are in line with EEG observations indicating that active neural control is greater during ML compared to AP sway [10]. However, the correlations were weak, possibly because the TA is also a primary dorsiflexor, which is essential for AP control. The peroneus longus (PL) may be physiologically and anatomically a more accurate target than the TA to determine the effects of ML manipulations on neural control of standing sway. Both PL and TA activity increase with ML sway, but PL activity is necessary only for ML control since plantarflexor forces are generated primarily by soleus and gastrocnemius [11]. Additionally, impaired PL control has been implicated in postural deficits associated with ankle instability [12,13]. However, neural excitability of the PL has not been examined in standing [14]. Thus, we determined the effects of ML standing task difficulty manipulation on corticospinal and M1 excitability of the PL, in healthy young adults. We expected to find an increase in corticospinal excitability and decrease in M1 GABAa inhibition and M1 facilitation, correlated with the increase in ML COP velocity as task difficulty increases. This expectation would lend support to the idea that active neural control, particularly cortical involvement in ML sway control, increases with task difficulty. Also, we examined, for the first time, M1 GABAb inhibition, which shows distinct task-specific modulation compared to GABAa inhibition [15].
Numerous studies have reported large between-participant variation in neural excitability of hand and leg muscles using TMS [8,16–18]. Such variation is found despite high test-retest reliability [19] leading to the idea of participant-specific ‘‘intrinsic neural excitability’’ [20]. Therefore, we considered the so far overlooked possibility that the neural modulation in response to changing task difficulty is dependent on excitability measured in the control
Chapter 3
38
stance, predicts changes in excitability as standing difficulty increases. We manipulated task difficulty by decreasing the ML BOS (wide, narrow, tandem, one leg). To minimize variation due to unreliability and strengthen our inferences, we examined task-specific reliability of each outcome variable to guide the main analyses.
2. mATeriAls ANd meThods
2.1 Participants
Fourteen healthy young adults aged 22.3 ± 1.7 years (mean ± SD, 12F) volunteered for the main study and data were acquired during a single 1.5 h long lab visit. Reliability of TMS outcomes was examined in another group of 15 young adults (22.1 ± 2.0, 11F) who visited the lab twice ~7 days apart. This study was carried out in accordance with the recommendations of the Medical Ethical Committee of the University Medical Center Groningen. The protocol was approved by the Medical Ethical Committee. All subjects gave written informed consent in accordance with the Declaration of Helsinki [21]. A safety questionnaire [22] was used to exclude individuals with history of neurological or orthopedic disorders, seizures, head trauma; suspicion of pregnancy; metal implants or pacemakers; used medication known to lower seizure threshold or had blood relatives with a history of seizures. We also determined foot dominance [23]. Level of physical activity [24] and mobility [25] were measured to ensure that our study sample had relatively similar physical activity levels, which can affect balance, and consequently our outcome measures. No participants were excluded based on these data.
2.2 Procedures
Measurement of TMS and COP outcomes was conducted in four tasks: (1) wide stance (feet shoulder width apart); (2) narrow stance (feet together); (3) tandem stance (dominant foot posterior), and (4) one leg stance (dominant foot). Participants wore socks and stood with arms crossed across the chest. Task order was randomized across participants, with 2–3 min of rest between tasks. Maximal voluntary contraction (MVC) was used to normal-ize and compare background EMG (bEMG) across tasks and participants. Two methods were used—manual resistance against ankle plantarflexion and eversion in sitting or heel rise while standing on one leg. Each method was repeated three times and the highest EMG obtained from all six trials was used as an estimate of MVC.
2.3 data acquisition
Wireless sensors (dimensions—37*26*15 mm, electrode material—silver; TrignoTM Wire-less System, Delsys, Natick, MA, USA) were used to record EMG from the dominant side PL. The signal was amplified 1000 times and sampled at 5000 Hz using data acquisition
interface and software (Power 1401 and Signal v5.11, Cambridge Electronic Design Ltd, Cambridge UK).
Magnetic pulses were applied using two single-pulse magnetic stimulators (Magstim Model 2002, The Magstim Co., Whitland, UK), a Bistim module and a double cone coil (110 mm). Participants wore a cloth cap marked with a grid and the coil was moved in 1 cm incre-ments to determine the hot-spot which was defined as the location where the largest and most consistent motor evoked potentials (MEPs) were obtained. The hot-spot was marked on the cap to ensure consistent positioning of the coil, which was held by the researcher. In standing, the active motor threshold (MT) was determined by systematically varying the stimulation intensity to find the lowest level of stimulator output at which 3 out of 5 MEPs had a peak-to-peak amplitude of at least 50 mV. For eliciting short interval intracortical inhibition (SICI, GABAa mediated) and intracortical facilitation (ICF), the conditioning and test pulse were set at 70% and 110% MT, respectively. For long interval intracortical inhibi-tion (LICI, GABAb mediated), the condiinhibi-tioning and test pulse were set at 120% and 110% MT respectively. An inter-stimulus interval (ISI) of 3, 13 and 100 ms was used for SICI, ICF and LICI, respectively [26]. These parameters were chosen based on extensive pilot testing which examined SICI, LICI and ICF using different combinations of intensities and ISIs. Ten paired pulses each for the SICI, LICI and ICF protocols, and 10 single pulses at 110% MT were applied in random order. There was an 8–10 s interval between pulses (or pulse pairs). COP location was calculated using force and moment data obtained using two force plates (Bertec 4060-08, Columbus, OH, USA) embedded in the floor, sampled at 200 Hz and acquired using a custom LabVIEW script (v2015, National Instruments, Austin, TX, USA).
2.4 data analyses
Data were analyzed using Matlab (The Mathworks, Natick, MA, USA). EMG was bandpass filtered using a 4th order dual pass Butterworth filter with 10 Hz and 1000 Hz high and low pass cut-offs, respectively. For the MVC trials, data were smoothed with a moving average with 100 ms non-overlapping windows, the peak voltage was measured and the highest of six trials used as an estimate of peak muscle activation during MVC. bEMG was estimated as the mean rectified signal over a 100 ms window before the TMS pulse and expressed as %MVC. Additionally, bEMG area was calculated by integrating the rectified EMG in the same window. MEP peak-to-peak amplitude was estimated from unrectified EMG, in a 100 ms window after application of the TMS pulse. For determination of MEP area, the filtered EMG was rectified, MEP onset was detected, and the data were integrated over a 100 ms window starting at onset. Onset was automatically detected in a 100 ms window after the TMS pulse, if the signal exceeded a bEMG+2SD threshold for at least three data points
Chapter 3
40
detected onset was visually inspected and manually corrected when required. Normalized MEP area was calculated by subtracting the trial specific bEMG area from MEP area [27]. For each of the three measures 8–10 trials each were averaged to obtain estimates of test MEP and conditioned MEPs for SICI, LICI and ICF. A few trials were discarded due to improper coil placement or technical errors in syncing the TMS data with force plate data. SICI, LICI and ICF were quantified as: Conditioned MEP/Test MEP * 100. Higher values indicate lower inhibition (SICI and LICI) and greater facilitation (ICF).
The COP location time series was low pass filtered using a 4th order Butterworth filter with a 5 Hz cut-off. Pilot testing showed that placement of the coil on the head alters COP velocity, even in the absence of stimulation. Therefore, COP data was extracted from a 2 s window before application of each pulse, when the coil was already positioned on the head. The distance between each pair of consecutive data points was summed to obtain the total distance and divided by time to obtain ML COP velocity. Velocity was averaged across 40 trials to obtain a single estimate for each standing task.
2.5 statistical analyses
All statistical analyses were conducted using SPSS (Version 24, IBM Corp., Armonk, NY, USA). The Shapiro-Wilk test revealed that several variables were not normally distributed, and these were log transformed for further analyses. Reliability of the TMS outcomes was estimated using intra-class correlation coefficients (ICC). ICC(2,k) i.e., a two-way random effects model for averaged measures (averaged over 8–10 trials) was used [28]. Categories of ICCs were based on a recent multi-center TMS reliability study: ICC >0.8: high and 0.5–0.8: moderate [29] and negative values were set to 0 [30]. Since reliability did not differ much between the three methods of MEP estimation, corrected area was used for all further analyses to minimize the effects of bEMG. For each outcome, one-way repeated measures ANOVA was used to determine differences between tasks only for TMS outcomes with at least moderate reliability i.e., ICC >0.5. For the ANOVAs, Greenhouse-Geisser correction was applied when sphericity was violated and Bonferroni’s post hoc tests were used for pairwise comparisons. If excitability was not normally distributed in one or more tasks, log transformed values were used for the ANOVA. Spearman’s rank correlation coefficients were used to test for linear associations between neural excitability and ML COP velocity with data pooled across all reliable tasks. Additionally, correlation between neural excitability and bEMG was estimated to test whether any differences in excitability between tasks was confounded by bEMG changes. All descriptive data are presented as mean (±SD).
