Pediatric Brain Stimulation: Mapping and Modulating the Developing Brain

By Donald L. Gilbert

Pediatric Brain Stimulation: Mapping and Modulating the Developing Brain presents the latest on this rapidly expanding field that has seen an exponential growth in publications over the past 10 years. Non-invasive modalities like TMS can painlessly map and measure complex neurophysiology in real patients. Neuromodulatory applications like rTMS and tDCS carry increasingly proven therapeutic applications. Rapidly advancing technological methodologies are increasing opportunities and indications.

Despite all these benefits, applications in the more plastic developing brains of children are only just emerging. This book provides a comprehensive overview of brain stimulation in children. Chapters include Transcranial Magnetic Stimulation (TMS) fundamentals, brain stimulation in pediatric neurological conditions, and invasive brain stimulation.

The main audience for this research will be those interested in applying brain stimulation technologies to advance clinical research and patient care, although a wide variety of clinicians and scientist will find this to be a valuable reference on brain stimulation with specific chapters on a variety of conditions.

• Provides an overview of recent findings and knowledge of pediatric brain stimulation and the developing brain
• Edited by renowned leaders in the field of pediatric brain stimulation
• Presents a great resource for basic and clinical scientists and practitioners in neuroscience, neurology, neurosurgery, and psychiatry

• Language: English
• Publisher: Academic Press
• Release date: May 4, 2016
• ISBN: 9780128020388

Book preview

Pediatric Brain Stimulation

Mapping and Modulating the Developing Brain

Editors

Adam Kirton

University of Calgary, Calgary, AB, Canada

Donald L. Gilbert

Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States

Table of Contents

Cover image

Title page

Copyright

List of Contributors

Foreword

Preface

Section I. Fundamentals of Nibs in Children

Chapter 1. TMS Basics: Single and Paired Pulse Neurophysiology

Introduction

How Does TMS Work?

Neuronal Structures Activated by TMS

TMS Neurophysiology

TMS Reliability

TMS Safety

Chapter 2. Assessing Normal Developmental Neurobiology With Brain Stimulation

Introduction

Normal Developmental Neurobiology

TMS-Evoked Parameters That Reflect Neuromotor Maturation

Assessing Neurophysiologic Functioning With Brain Stimulation in Development and Developmental Disorders

Outlook

Chapter 3. Neuroplasticity Protocols: Inducing and Measuring Change

Background (In Vitro and Animal Work)

Overview of Repetitive Transcranial Magnetic Stimulation (rTMS)

Types of rTMS

Study Design Considerations

Conclusion

Chapter 4. Therapeutic rTMS in Children

Introduction: From Biomarker of Plasticity and Excitability to Inducer of Therapeutic Changes?

Methods: Types of Repetitive Transcranial Magnetic Stimulation (rTMS)

Most Common Repetitive TMS Methods

Brief Overview of Clinical Trials and Indications in Young Adults

Brief Overview of Registered Clinical Trials Involving Children

Brief Overview of Published Clinical Trials Involving Children

Where to Next With rTMS Treatment in Children?

Chapter 5. Transcranial Direct-Current Stimulation (tDCS): Principles and Emerging Applications in Children

Introduction

Principles and Cortical Effects of Transcranial Direct-Current Stimulation

tDCS Methodology: Electrodes, Current, and Protocols

tDCS Dosage: Computational Modeling Evidence

tDCS Dosage: Neurophysiology Evidence

tDCS in Healthy Children to Enhance Motor Learning

Safety, Adverse Events, and Tolerability

Clinical Application of tDCS

Summary

Chapter 6. Insights Into Pediatric Brain Stimulation Protocols From Preclinical Research

Stimulating the Immature Brain

The Immature Brain Is Hyperexcitable

Inhibitory Control Develops With Age

GABAergic Dysmaturity in Neurodevelopmental Disease

Synaptic Plasticity Considerations in the Developing Brain

Relevance to Translational Brain Stimulation

Conclusion

Chapter 7. Pediatric Issues in Neuromodulation: Safety, Tolerability and Ethical Considerations

Brain Stimulation Techniques Used in Children

Direct Effects of Brain Stimulation in Children

Developmental Neuroplasticity

Side Effects of Stimulation

Ethical Application of Neuromodulation in Children

Neuro-Enhancement in Healthy Individuals

Advisements for Safe Administration of NIBS in Children

Conclusions

Section II. Nibs in Pediatric Neurological Conditions

Chapter 8. TMS Applications in ADHD and Developmental Disorders

Introduction

Key Considerations for TMS Research in Children With Developmental Disorders

TMS in Developmental Disorders: Practical Considerations and Representative Results

Chapter 9. TMS Mapping of Motor Development After Perinatal Brain Injury

Introduction

Developmental Aspects

Functional Aspects

Functional Imaging

MR Tractography

Clinical Relevance I: Functional Therapy

Clinical Relevance II: Epilepsy Surgery

Perspective: Brain Stimulation

Chapter 10. The Right Stimulation of the Right Circuits: Merging Understanding of Brain Stimulation Mechanisms and Systems Neuroscience for Effective Neuromodulation in Children

Introduction

Focal Stimulation to Known Circuits: Lessons From Deep Brain Stimulation for Parkinson’s Disease

Brain Stimulation in Pediatric Hemiparesis: Which Circuits?

Targeting Motor Cortex: Which Stimulation Modality?

Focal Stimulation Versus General Plasticity and Task-Specific Training

Challenges and Opportunities

Chapter 11. Therapeutic Brain Stimulation Trials in Children With Cerebral Palsy

Introduction

Scientific Evidence in Therapeutic Brain Stimulation in Children With CP: Summary of Current Interventional Trials

Trial Design

Conclusions

Chapter 12. Brain Stimulation in Children Born Preterm—Promises and Pitfalls

Introduction

The Motor Threshold

Short-Interval Intracortical Inhibition and Facilitation

Long-Interval Intracortical Inhibition and the Cortical Silent Period

Transcallosal Inhibition

Neuroplasticity Induced With Non-Invasive Brain Stimulation

Other Considerations

Where to Now?

Chapter 13. Brain Stimulation to Understand and Modulate the Autism Spectrum

Introduction

TMS Measures of ASD Pathophysiology

TMS as a Therapeutic Intervention in ASD

Conclusion

Chapter 14. Non-Invasive Brain Stimulation in Pediatric Epilepsy: Diagnostic and Therapeutic Uses

Introduction

Diagnostic Uses of TMS

Therapeutic Uses of TMS

Therapeutic Uses of Transcranial Direct-Current Stimulation

Conclusion

Chapter 15. Brain Stimulation in Pediatric Depression: Biological Mechanisms

Introduction

Preclinical

Neurophysiology

Neuroimaging

Peripheral Biomarkers

Neurocognition

Conclusion

Chapter 16. Brain Stimulation in Childhood Mental Health: Therapeutic Applications

Introduction

Applications

Discussion

Where Do We Stand and Where Do We Need to Go?

Conclusion

Chapter 17. Transcranial Magnetic Stimulation Neurophysiology of Pediatric Traumatic Brain Injury

Epidemiology and Clinical Spectrum of Traumatic Brain Injury

Outcome

Pathophysiology of Traumatic Brain Injury

Insights About TBI Using TMS: A Literature Review

Mild Traumatic Brain Injury

Severe Traumatic Brain Injury

Treatment Studies Using rTMS

TMS in Pediatric Mild Traumatic Brain Injury

Safety and Tolerability

Results

Summary

Chapter 18. Brain Stimulation Applications in Pediatric Headache and Pain Disorders

History of Neurostimulation in Headache and Pain

Pediatric Headache

Safety and Tolerability of Neurostimulation in Children

Pathophysiology of Head Pain and Migraine

Pediatric Data

Therapeutics

Future Avenues in Pediatric Brain Stimulation for Pain

Section III. Invasive Brain Stimulation in Children

Chapter 19. Deep Brain Stimulation in Children: Clinical Considerations

Historical Background

Primary Dystonias and DBS

Physiology of DBS Effect in Primary Dystonias

Efficacy of DBS on DYT-1-Positive and DYT-1-Negative Dystonias

Segmental Versus Generalized Dystonia and DBS

Effects of Switching DBS Off

DBS for Secondary Dystonias

DBS for Neurodegeneration With Brain Iron Accumulation (NBIA)

Primary and Secondary Dystonia Responses to DBS

Secondary Dystonias and Functional Imaging Before and After DBS

Patient Selection for DBS

Pooled Cohorts of DBS Data

Children With Dystonia and DBS

Technical Issues Influencing DBS Efficacy

Chapter 20. Deep Brain Stimulation Children – Surgical Considerations

Background

Technical Aspects of DBS

Clinical Indications

Treatment Efficacy

Physiology of Neuromodulation, Neuroplasticity, and Pediatric Considerations

Future Developments and Innovation

Chapter 21. Invasive Neuromodulation in Pediatric Epilepsy: VNS and Emerging Technologies

Introduction

Epilepsy in Children

Surgical Treatments for Epilepsy

Vagus Nerve Stimulation (VNS)

Trigeminal Nerve Stimulation

Intracranial Neurostimulation for Epilepsy

Conclusion

Chapter 22. Emerging Applications and Future Directions in Pediatric Neurostimulation

Transcranial MRI-Guided Focused Ultrasound

Robotics

Brain–Computer Interfaces (BCI)

Future Directions for Single- and Paired-Pulse TMS Studies

Brain Stimulation Biomarkers

Functional TMS

TMS for Biofeedback

Non-Invasive Low-Current Electrical Stimulation

Shifting Membrane Potentials With Transcranial Direct-Current Stimulation

Entraining or Disrupting Intrinsic Brain Oscillations With Transcranial Alternating-Current Stimulation

Conclusion

Index

Copyright

Academic Press is an imprint of Elsevier

125 London Wall, London EC2Y 5AS, UK

525 B Street, Suite 1800, San Diego, CA 92101-4495, USA

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, USA

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

Copyright © 2016 Elsevier Inc. 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 and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 978-0-12-802001-2

For Information on all Academic Press publications visit our website at https://www.elsevier.com/

Typeset by TNQ Books and Journals

www.tnq.co.in

List of Contributors

S.H. Ameis,     Campbell Family Mental Health Research Institute, CAMH; University of Toronto, Toronto, ON, Canada

K.M. Barlow,     University of Calgary, Calgary, AB, Canada

R.D. Bhardwaj,     Sanford Children’s Hospital, Sioux Falls, SD, United States

J.B. Carmel

Burke-Cornell Medical Research Institute, White Plains, NY, United States

Weill Cornell Medical College, New York, NY, United States

P. Ciechanski,     University of Calgary, Calgary, AB, Canada

P. Croarkin,     Mayo Clinic College of Medicine, Rochester, MN, United States

K.M. Friel

Burke-Cornell Medical Research Institute, White Plains, NY, United States

Weill Cornell Medical College, New York, NY, United States

R. Gersner,     Harvard Medical School, Boston, MA, United States

D.L. Gilbert,     Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States

B.T. Gillick,     University of Minnesota, Minneapolis, MN, United States

A.M. Gordon

Teachers College of Columbia University, New York, NY, United States

Columbia University Medical Center, New York, NY, United States

W.J. Hader,     University of Calgary, Calgary, AB, Canada

M.Q. Hameed,     Harvard Medical School, Boston, MA, United States

N.H. Jung,     Technische Universität München, Kinderzentrum München gemeinnützige GmbH, Munich, Germany

S.K. Kessler,     Children’s Hospital of Philadelphia, Philadelphia, PA, United States

A. Kirton,     University of Calgary, Calgary, AB, Canada

K. Limburg,     Technische Universität München, Kinderzentrum München gemeinnützige GmbH, Munich, Germany

J.-P. Lin,     Guy’s and St. Thomas’ NHS Foundation Trust, London, United Kingdom

K. Ma,     University of Calgary, Calgary, AB, Canada

F.P. MacMaster,     University of Calgary, Calgary, AB, Canada

V. Mall,     Technische Universität München, Kinderzentrum München gemeinnützige GmbH, Munich, Germany

J. Menk,     University of Minnesota, Minneapolis, MN, United States

L.M. Oberman,     Brown University, Providence, RI, United States

A. Pascual-Leone,     Harvard Medical School, Boston, MA, United States

E.V. Pedapati,     Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States

J.B. Pitcher,     The University of Adelaide, Adelaide, SA, Australia

T. Rajapakse,     Alberta Children’s Hospital, Calgary, AB, Canada

M. Ranjan,     University of Calgary, Calgary, AB, Canada

A. Rotenberg,     Harvard Medical School, Boston, MA, United States

K. Rudser,     University of Minnesota, Minneapolis, MN, United States

M.J. Sanchez,     Harvard Medical School, Boston, MA, United States

T.A. Seeger,     University of Calgary, Calgary, AB, Canada

M. Sembo,     University of Calgary, Calgary, AB, Canada

M. Staudt

Schön Klinik, Vogtareuth, Germany

University Children’s Hospital, Tübingen, Germany

S.W. Wu,     Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States

E. Zewdie,     University of Calgary, Calgary, AB, Canada

Foreword

Brain stimulation in various forms has been practiced for centuries, and it has gone in and out of favor, but recently with advances in methodologies, it has achieved a resurgence of interest. Electroconvulsive therapy (ECT) has been used over many decades for treatment of depression, but its side effects are concerning. It is the efficacy of deep brain stimulation (DBS) for various indications such as Parkinson’s disease that has pointed the modern direction for therapeutic brain stimulation and increased excitement. This advance was recently recognized with the Lasker prize. In the past two decades a large variety of non-invasive brain stimulation methods have been developed. First was high-voltage pulsed electrical stimulation. The demonstration that such stimulation could activate the motor cortex was met with astonishment. This was quickly followed by transcranial magnetic stimulation (TMS) which was welcomed since the electrical stimulation was painful. TMS has proven to be a superb tool for studying brain physiology, has some clinical utility for diagnosis, and is being explored for therapeutic benefit. Already TMS has achieved US FDA approval for the treatment of depression. Continuous low-voltage electrical stimulation, either direct current or alternating current (tDCS and tACS), is now also being used. It is growing fast in popularity since is it easy to do, cheap to manufacture, and seems also to have clear efficacy in altering brain function.

It should be noted in relation to therapy that some types of brain stimulation, like DBS, can be used continuously and work by online modulation of brain networks. In general, non-invasive brain stimulation is used only intermittently and any prolonged therapeutic effect would depend on brain plasticity to alter brain function.

All of this has applicability in children but has been developing more slowly. There is, of course, always less research in children than in adults. Some of this has to do with safety concerns. Children might well be more vulnerable in many ways, including the possibility of negatively interfering with brain development. Deep brain stimulation for the youngest children would be confounded by growth of the head. With TMS, which creates a loud noise with each pulse, there has been particular concern about possible hearing damage, in part because of the different anatomy of the external auditory canal in children. Recent studies of this latter issue have, fortunately, shown no problem as long as the study is conducted properly.

Nevertheless, the literature on brain stimulation in children has accumulated and it is timely to summarize the field. Dr. Kirton and Dr. Gilbert have put together a fine group of experts to cover the field broadly, from basic mechanisms, to physiology to therapy, including all types of brain stimulation. An area that was explored early on with TMS was the development of the corticospinal system in children, and this has allowed physiological studies of cerebral palsy with brain injury at various times. More recently there has been exploration of other childhood disorders. It is clear that there will be many applications, and a book like this should help firm up the foundation of the field and propel it forward.

Mark Hallett, MD,     National Institute of Neurological Disorders and Stroke, Bethesda, MD, United States

Preface

Children are not little adults and disorders of the developing brain are different. Non-invasive brain stimulation technologies are revolutionizing our understanding of human neurophysiology in health and disease. That the vast majority of progress has occurred in the mature brain is entirely reasonable, as initial uses of experimental devices require substantial safety testing before use in children.

Not long after these early accomplishments were realized the exciting potential to explore the fascinating elements of brain development was harnessed. During the 1990s, several forward-thinking pioneers began applying TMS methodologies to children, including preschoolers and even newborns. The results were nothing short of astounding, providing novel in vivo evidence of fundamental elements of brain development in health and disease previously only suggested by animal models, limited imaging, or simple theory.

Despite this early progress, neurophysiological applications in children continue to lag behind those in adults. A search of transcranial magnetic stimulation (TMS) in PubMed in 2015 demonstrates continued exponential scientific growth, yet <3% of studies are dedicated to children. This trend is even more pronounced when searching therapeutic brain stimulation applications and clinical trials.

The need for novel treatment interventions in childhood neurological and neuropsychiatric disease is pressing. Disorders are increasingly understood but treatment options often lag behind due to complexities of the developing brain, failure of drug companies to develop pediatric-specific pharmaceuticals, and other factors. The appeal of non-pharmacological, individualized, and targeted interventions is particularly appealing to both young patients and their parents.

The time to harness the full potential of non-invasive brain stimulation applications in children is now.

To facilitate this goal, we have attempted to pull together the global leaders in brain stimulation focused on the developing brain. To be clear, the majority of pioneers in virtually all aspects of non-invasive brain stimulation methodology and applications have come from the adult world. Many of the topics reviewed here carry a wealth of experience and data thanks to these many visionary scientists. That none of them are primary contributors to chapters in this volume in no way reflects their many seminal contributions. However, it is also intentional.

Our aim instead was to garner the knowledge and experience of those whose research, clinical care, and experience is focused on children. Validating and supporting this philosophy was the remarkably successful recruitment of authors we encountered. Almost without exception, each contributor responded immediately with keen interest, enthusiasm, and willingness to participate. As editors who also work in the world of pediatric neuroscience, this was not entirely surprising. In our experience, pediatricians and others focused on advancing the wellbeing of children tend to be of a highly generous and collaborative nature. The response we received certainly reflected this. This strong endorsement validated our initiating the project while further inspiring us to generate the best possible product.

The result is a volume that begins by attempting to orient less experienced readers in the fundamentals of currently available non-invasive brain stimulation technologies. Pediatric aspects are emphasized while providing the essential knowledge required to interpret and perform basic brain stimulation research. This is followed by a categorical approach to brain stimulation applications across the main neurological, psychiatric, and developmental disorders most often affecting children. These are intended to inspire those with an interest in these patient populations to pose new questions and conceive potential approaches to advancing understandings and treatment possibilities.

As summarized in the final chapter, the future holds many additional promising directions. We hope our readers will share our enthusiasm and contribute to what will certainly be exciting times to come with real potential to improve the lives of affected children and their families.

Dr. Adam Kirton

Dr. Donald L. Gilbert

Section I

Fundamentals of Nibs in Children

Outline

Chapter 1. TMS Basics: Single and Paired Pulse Neurophysiology

Chapter 2. Assessing Normal Developmental Neurobiology With Brain Stimulation

Chapter 3. Neuroplasticity Protocols: Inducing and Measuring Change

Chapter 4. Therapeutic rTMS in Children

Chapter 5. Transcranial Direct-Current Stimulation (tDCS): Principles and Emerging Applications in Children

Chapter 6. Insights Into Pediatric Brain Stimulation Protocols From Preclinical Research

Chapter 7. Pediatric Issues in Neuromodulation: Safety, Tolerability and Ethical Considerations

Chapter 1

TMS Basics

Single and Paired Pulse Neurophysiology

E. Zewdie     University of Calgary, Calgary, AB, Canada

A. Kirton     University of Calgary, Calgary, AB, Canada

Abstract

In this chapter, we briefly summarize the physics and biophysics of transcranial magnetic stimulation (TMS) and also overview the neuronal structures activated by TMS. The neurophysiology underlying single pulse, paired pulse, and repetitive TMS are discussed, as well as an example from experiments on pediatric population being provided. The reliability and safety of the techniques described in this chapter are also reviewed.

Keywords

Intracortical facilitation; Motoneurons; Motor cortex; Neurophysiology; TES; TMS

Outline

Introduction 4

How Does TMS Work? 4

Principle of TMS 4

Types of TMS Coil 6

Neuronal Structures Activated by TMS 7

Stimulation of the Motor Cortex 7

Descending Volleys 8

Descending Pathways from the Cerebral Cortex 9

Motor-Evoked Potential 10

Motor Thresholds 12

TMS Neurophysiology 12

Single-Pulse TMS 12

Stimulus Response Curve 12

Silent Periods 13

Paired-pulse TMS 14

Short-Interval Intercortical Inhibition (SICI) 14

Intracortical Facilitation (ICF) 15

Short-Interval Intracortical Facilitation (SICF) 15

Long-Interval Intracortical Inhibition (LICI) 16

Interhemispheric Inhibition (IHI) 16

TMS Paired With Sensory Activation 17

Short-Latency Afferent Inhibition (SAI) 17

Paired Associative Stimulation (PAS) 18

Repetitive TMS 18

TMS Reliability 19

TMS Safety 20

References 20

Introduction

Transcranial magnetic stimulation (TMS) is a non-invasive technique that painlessly delivers magnetic fields across the scalp and the skull to induce regional activation of cortical neuronal populations in the brain. Since its introduction by Barker in the 1980s, TMS has greatly advanced our ability to explore and understand neural circuitry and physiology in vivo. Using multiple patterns of stimulations (single, paired, or repetitive), TMS is capable of characterizing pathways, excitability of inhibitory and excitatory circuits and other neurophysiological elements, as well as measuring and modulating plasticity. Due to its safe nature and favorable tolerability, TMS has become an invaluable tool across virtually all areas of clinical neuroscience, though its applications in the developing brains of children are only just being realized.

How Does TMS Work?

Principle of TMS

Merton and Morton¹ were the first to build a high-voltage transcranial electrical stimulator (TES) that could be used to activate the cerebrum via surface electrodes placed over the scalp by passing an electric current in a perpendicular direction to the stimulating electrodes. TES works by transferring a short-lasting electric current from its source to the excitable neurons of the brain. Using this technique a very high voltage is required in order to stimulate the neuronal tissue of the brain as the skull and the scalp have high resistance. Although electrical stimulation has greatly advanced since (see Chapter 5: Transcranial Direct-Current Stimulation (tDCS): Principles and Emerging Applications in Children), magnetic stimulation, a less painful alternative brain stimulation technique, has emerged as an appealing alternative.

TMS was developed in the 1980s as a method to non-invasively stimulate the brain. The main principle of TMS is the induction of a short-lasting electric current in the cerebral cortex. Unlike TES, which involves direct transfer of current through electrodes, TMS uses electromagnetic induction to deliver current to the brain. A very short-lasting (150–300  μs) electric current is applied to a stimulating coil to produce a rapidly changing magnetic field which, in turn, induces a flow of electric current in nearby conductors – including the human brain. The current that is applied to the brain from TMS is governed by the Faraday–Henry law, which states that if an electric conductor is linked by a time-varying magnetic flux, a current is observed in the circuit. The magnetic flux is created due to the current that is passing through the stimulating inductor coil, and its magnitude depends on the material property of the coil and the amount of current that is passing through the coil.

The magnetic stimulator typically consists of a high-current pulse generator (capacitor), which produces a current of 5000  A when charged up to 2.8  kV, and a stimulating coil (inductor coil) producing magnetic pulses with a field strength of 1  T or more and a pulse duration of ∼1 ms² (Fig. 1.1).

When the capacitor discharges, apart from the electrical energy that is lost in the wiring, about 500  J of energy is transferred to the coil and then returned to the instrument to help reduce coil heating in about 100  μs, which means that, during the discharge, energy initially stored in the capacitor in the form of electrostatic charge is converted into magnetic energy in the stimulating coil in this time frame.² This leads to a rapid rise in the magnetic field (around 30  kT/s) which in turn induces current in the brain in the order of 1–20  mA/cm².

Figure 1.1  The principle of the magnetic stimulator.

The current generated, when the charge stored in the capacitors is discharged through the inductor coils, creates a magnetic field that can penetrate through the skull and induce secondary current in the opposite direction.

The effectiveness of the stimulation in activating the cortical neurons is determined by the waveform of the pulses and the induced current direction. Two types of waveforms are currently used – monophasic and biphasic. Biphasic stimulators are more effective than monophasic stimulators because a biphasic waveform produces currents with longer durations and can therefore produce more current to activate neurons.³ At the same time, coil orientation is an important determinant of which cortical neuronal populations are affected during TMS. The current that is induced by TMS in the brain tissue flows in a direction that is parallel to the stimulating coil but perpendicular to the magnetic field created by the coil. If the induced current has the same direction as the nerve signal in the axons it results in the most effective stimulation. Therefore, axons that are orientated parallel (or horizontal) to the stimulating coil placed on the surface over the motor cortex are most effectively activated by TMS.

Types of TMS Coil

TMS coils are designed in different shapes (see Fig. 1.2) to focus current in different parts of the cortex by varying the spatial field and depth of the induced electric field. The simplest design is a circular coil which typically has a diameter of 8–15  cm (Fig. 1.2A), with a maximum magnetic field directly under the center of the circle and a maximum induced current at the outer edge of the circle. Although this type of coil creates good penetration of the cerebral cortex, its spatial resolution is relatively low. The circular coil activates the motor cortex asymmetrically, with greater activation on the side where the coil current flows from posterior to anterior across the central sulcus.⁴ Hence, with the coil placed centrally on the vertex, the induced current predominantly stimulates the left motor cortex as the current flow will be posterior to anterior. Neurons located at a depth of 10.5  mm from the surface of the skull will be stimulated by a circular coil.⁴

More focal stimulation can be achieved by putting two smaller circular coils side-by-side in a coil type called a figure-of-eight coil (Fig. 1.2B). The magnetic fields created by currents flowing in opposite directions summate at the junction of the two circular coils and result in a more focal stimulation directly under the junction. Due to the smaller diameter of the coils, the penetration level is more limited at the center of each coil. Neurons located at a depth of 11.5  mm from the surface of the skull will be stimulated by a figure-of-eight coil.⁵ Typically, side-by-side coils range from very small flat coils for brain mapping work, such as figure-of-eight coils (Fig. 1.2B), to large contoured versions, such as bat-wing coils (Fig. 1.2C) or double cone coils (Fig. 1.2D), which may be used to stimulate deeper neural structures in the brain. Additional coils, such as those with cooling air circulation (Fig. 1.2E) have been designed to further advance the depth or other TMS delivery parameters.

Figure 1.2  TMS coil types.

(A) Circular coil has only a single winding with superficial current penetration. (B) Figure-of-eight coil has a flat shape with two adjacent windings and is used to activate more superficial hand areas of the cerebral cortex. (C) Bat-wing coil has a flat center with bent wings and is used for activating leg areas of the cerebral cortex. (D) Double-cone coil has two large cup-shaped windings positioned side-by-side in an angle and can activate deeper areas of the cortex more powerfully than the bat-wing coil. (E) Double air film coil has cold air circulating in order to cool the inductor coil. It is mainly used to deliver repetitive TMS.

Neuronal Structures Activated by TMS

Stimulation of the Motor Cortex

The cell bodies and axons of neurons in the cerebral cortex that respond to TMS⁶ depend on their size, location, orientation, and function. TMS can be delivered in a lateral–medial or anterior–posterior configuration to stimulate specific elements in the cerebral cortex. Compared to TES, TMS produces the most localized electric field. However, a localized field from TMS also decreases the ability to stimulate neurons buried deep in the brain because a magnetically induced field drops off faster than an electrically produced one.⁵ Electrical fields must have a component parallel to the neuron in order to stimulate it.⁷ In the motor cortex, specifically in Area 4, there are neurons on the surface that are oriented more vertically and neurons deeper within the central sulcus that are oriented more horizontally (and all configurations in between).

Given that magnetic stimulation produces a current flow that is parallel (in a horizontal direction) to the surface of the cortex, the neurons that are going to be stimulated directly are those that are parallel to the surface of the cortex. These neurons with horizontally aligned axons are typically located in deeper layers of the cortex. Since the strength of an electric field created by magnetic stimulation falls off with depth, these parallel neurons may be less stimulated if they are located very deep. In contrast, the electric field produced during electric stimulation has components both parallel and perpendicular (vertical) to the surface of the head. Therefore, it activated the axons or axon hillocks of the corticospinal tract (CST) neurons.

Descending Volleys

Evidence that multiple descending volleys in the CST can be activated by electrical stimulation was first proposed in the mid-1950s by Patton and Amassian.⁸ When a single, square-wave electrical shock was applied to the motor cortex of anesthetized cats and monkeys, multiple CST neurons were activated. The first wave was produced by direct excitation of CST neurons, while later waves originated from indirect activation via cortical interneurons. Therefore, the terms D- and I-waves are used to describe the first (Direct) and the later (Indirect) responses, respectively. The latency of D-waves agreed with the conduction time of large corticospinal fibers. On the other hand, I-waves were found to be more susceptible to cortical injuries than D-waves. In addition, when microstimulating deep white matter, the size of D-waves increased and I-waves only appeared when the stimulating electrode was applied to the gray matter. This was an important finding since it provided initial evidence that multiple descending volleys from I-waves were generated trans-synaptically through excitatory corticocortical fibers when stimulating the surface of the motor cortex with a single pulse.

Similarly, descending corticospinal volleys in response to transcranial stimulation were studied by Day and co-workers.⁹,¹⁰ They found that the latency of the descending volleys evoked by TMS varied markedly according to the type of stimulation and the stimulation intensity. An increase in stimulation intensity results in an increase in the probability of eliciting shorter-latency motor responses. In particular, at high stimulation intensities, the amount of short-latency volleys was typically increased.

Descending volleys from TMS were typically slower and more complex than the volleys from TES. The fact that TES typically induced firing a few milliseconds before TMS suggests that descending volleys induced by both techniques are likely triggered using different mechanisms. As corticospinal volleys from TMS are typically delayed, it is believed that these volleys are likely I-waves and caused by trans-synaptic activation. In agreement with direct microstimulation of the motor cortex in monkeys, it is predicted that TES predominantly triggers axonal depolarization of pyramidal neurons and leads to a descending volley in the form of a D-wave. As the directions of current flow from TES and TMS are different, it is predicted that the orientation of the current in the cortex is likely responsible for eliciting these different responses. It is believed that the horizontal currents from TMS likely activate interneurons, whereas more vertically orientated currents from TES are more likely to activate the axons of pyramidal neurons directly. Repetitive trans-synaptic activation of pyramidal neurons is likely responsible for the delayed and more numerous volleys observed using TMS.

Descending Pathways from the Cerebral Cortex

The corticospinal tract (CST) is the main descending motor pathway from the cerebral cortex to the spinal cord that can be activated by TMS. The CST originates from large pyramidal cells predominantly in the fifth layer of the cerebral cortex. Over 60% of CST fibers originate from the primary motor cortex, supplementary motor area, and premotor cortex.¹¹ A substantial proportion of the fibers (about 40%) originate from the primary somatosensory area and the parietal cortex.¹² The corticospinal tract is the only descending motor tract that is known to make monosynaptic connections with spinal motoneurons in humans.¹³,¹⁴ Unlike in cats where there are no functional corticomotoneuronal (CM) connections,¹⁵ in humans the presence of CM connections on upper limb¹⁶ and lower limb¹⁷ motoneurons has been demonstrated. Functionally, muscles that require greater precision, such as the index finger and thumb, have greater CM connections.¹⁸ In human lower limbs, the monosynaptic projection to the tibialis anterior muscle is particularly strong, comparable in magnitude to the finger muscles.¹⁹ This may be related to the precision required to clear the toes above the ground during the swing phase of human walking.²⁰

In addition to activating the cerebral cortex, TMS over the motor cortex can activate projections to other cortical²¹ and subcortical regions.²² For instance, TMS over the primary motor cortex (M1) can trans-synaptically activate corticoreticular²³ and corticovestibular²⁴ pathways, which can produce or influence spinal motor output via reticulospinal and vestibulospinal pathways, respectively. Single unit recordings from pontomedullary reticular formation in the brainstem of anesthetized monkeys revealed multiple latency responses evoked by TMS performed over M1.²³ The multiple latency responses may indicate the existence of multiple pathways from M1 to such subcortical structures.

Motor-Evoked Potential

As described above, TMS can generate descending volleys, which in turn activate motoneurons at the spinal cord level. When the action potential of spinal motoneurons arrives at the muscle, it results in muscle depolarization and contraction. Using a pair of surface electrodes over the muscle, it is possible to non-invasively detect the potential difference as muscle action potentials are activated. An electromyography (EMG) can detect the sum of this activation. When the EMG is changed due to the TMS activation, the resulting deflection (increase) of EMG is called the motor-evoked potential (MEP). MEP latency and amplitude reflect the functional integrity of the upper motoneurons, the CST, spinal interneurons, and muscle. MEP amplitudes are measured by subtracting the largest negative peak from the largest positive peak of EMG within the MEP window (Fig. 1.3). On the other hand, the latency of the MEP is measured as the time from TMS trigger to the onset of the MEP. It is possible to calculate the central conduction time by combining TMS latency with peripheral motor latencies. Loss of large myelinated motor axons can cause prolongation of the central conduction time.

A simple mapping procedure is typically used to determine the optimal cortical location for TMS application to generate MEP in the muscle of interest. The center of the coil can be approximated to the area of the motor strip expected to represent the muscle of interest. For hand muscles, this is typically close to the central reference point in the 10–20 EEG system (eg, C3 for the muscles of the right hand) (Fig. 1.4). Single pulse applications at modest stimulator output (eg, 30%) can be applied and slowly escalated until MEPs are seen. Using an output slightly above threshold, the coil can then be systematically mapped across the region to determine the location that gives the largest, most consistent MEP. This region can then be marked on the scalp as the hotspot and referenced for all subsequent experiments. Modern image-guided neuronavigation systems have further enhanced this process, providing individualized, real-time co-registration of individual anatomy with coil positioning, facilitating very accurate placement within and across experimental sessions.

Figure 1.3  Stimulus response curves.

(A) Average of 10 first dorsal interosseous MEPs evoked from incrementing intensities. (B) Mean peak-to-peak amplitude of MEP from the data shown in (A) to produce a TMS response curve. Error bars represent ± standard error (SE).

Figure 1.4  Silent period.

(A) An example motor-evoked potential from the FDI muscle of a 15-year-old participant at maximum stimulation intensity (150% RMT) followed by cortical silent period (cSP). (B) An example of ipsilateral EMG suppression called the ipsilateral silent period (iSP). The vertical arrow indicates the onset and offset of the silent period. The horizontal arrow indicates the duration of the silent period. The average EMG before the TMS is shown by the red line.

Motor Thresholds

In their earliest contribution, Merton and Morton¹ noticed that the threshold for evoking a motor response is very low when the muscle under test is contracting (active) compared to when the same muscle is relaxed (rest). This finding was vital for experiments that collect EMG responses to control the contraction level. However, it was necessary to characterize the mechanism and level (spinal or cortical) where it occurs. Voluntary contraction may increase the excitability of cortical neurons so that a given stimulus recruits a larger descending corticospinal volley. On the other hand, voluntary activation may also increase the excitability of spinal motoneurons, making it easier to discharge them with a given descending volley. In their earlier study of motor cortex stimulation, Rothwell and co-workers²⁵ investigated changes in muscle response to anodal cortical stimulation when a subject exerted a background voluntary contraction. The latency has significantly decreased and the size of peak-to-peak EMG response increased when the muscle was contracting compare to when the muscle was relaxing. The same effect was also seen when using TMS. Mazzochio and co-workers²⁶ showed that tonic voluntary contraction increased cortical excitability to magnetic stimulation, so that a given stimulus evoked a larger descending corticospinal volley.

Most MEP measurements are normalized across individuals to the resting or active threshold values. Resting motor threshold (RMT) is defined as the minimum TMS intensity (expressed as percentage of maximum stimulator output or MSO) that elicits reproducible MEP responses of at least 50  μV in 50% of 5–10 consecutive trials.²⁷ Similarly, active motor threshold (AMT) is defined as the minimum TMS intensity that elicits reproducible MEP responses of at least 200  μV in 50% of 5–10 consecutive trials while the muscle is actively contracting from 10% to 20% of the maximum voluntary contraction (MVC).²⁸ Visual feedback of EMG activity is often used to obtain consistent levels of contraction. Motor thresholds vary widely across individuals, are higher in young children, and may be influenced by numerous additional factors including the muscle being tested, medications, and level of alertness. As shown below, most additional TMS neurophysiology experiments are referenced to these individualized thresholds.

TMS Neurophysiology

Single-Pulse TMS

Stimulus Response Curve

Single-pulse TMS can be used to estimate the strength and excitability of the motor cortex and its descending pathways. To produce a TMS response curve, TMS intensities can be incrementally increased either in terms of the maximum stimulator output (usually from 30% to 100% of MSO in steps of 10%) or in proportion to RMT/AMT (usually from 100% to 150% of RMT or AMT). Fig. 1.3A demonstrates an example from a healthy 14-year-old. Each trace is an average of 10 traces given at each of the indicated TMS intensities in random order. By measuring the peak-to-peak MEP amplitude, it is possible to then produce the stimulus response curve (SRC) shown in Fig. 1.3B. SRC from healthy participants typically generates a sigmoidal curve with MEP size plateauing at higher stimulation intensities. The parameters of the SRC may correlate with specific neurophysiological elements. For instance, the maximum MEP can be measured as the largest response on the recruitment curve, and may reflect connectivity or excitability of the CST or changes in the cortical motor map. In addition, SRCs that have shifted leftwards or rightwards without a vertical change could be indicative of alterations in recruitment thresholds or cortical excitability.²⁹

Silent Periods

During voluntary background contraction, a temporary suppression in EMG activity has been reported following an MEP induced by TMS to the contralateral motor cortex.³⁰ This is called the cortical silent period (cSP). This change in excitation is observed through the duration of the cSP and is believed to reflect a lack of cortical drive related to recruitment of intracortical inhibitory circuits. To determine the amount of inhibition induced at spinal and supraspinal levels, Fuhr and co-workers³⁰ performed H reflex testing at different periods in the silent period. As the H reflex was dramatically depressed at the start of the silent period and strongly recovered by the end, these findings implied that reductions in excitability due to the motoneurons were only involved in inhibition during the start of the silent period. For this reason, the lack of myoelectric activity at the end of the silent period was likely associated with a lack of cortical drive following magnetic stimulation.

Although the cortical silent period is easily demonstrated using TMS in able-bodied subjects, various neurological disorders have been reported to significantly alter and sometimes even abolish the silent period.³¹ Changes in cortical silent period may be related to cortical reorganization following neurological injury. For this reason, reorganization and disinhibition in the motor cortex appear to be involved in preserving and enhancing motor function after neurological impairment.

An ipsilateral silent period (iSP) can also be measured when TMS is delivered to the motor cortex on the same side as the contracting hand muscle. The experimental paradigm is essentially the same as that described above for the cortical silent period. The ipsilateral silent period is thought to reflect activation of transcallosal pathways affecting inhibitory interneurons in the contralateral, active motor cortex.³²

Paired-pulse TMS

All of the above TMS paradigms involve the administration of a single pulse. The addition of one or more earlier pulses of defined intensity and timing can alter these responses. Such paired-pulse TMS can provide additional insight toward neurophysiological mechanisms in health and alterations in disease. The baseline single pulse is referred to as the test stimulus (TS), while the additional modifying pulse is the conditioning stimulus (CS). Conditioning stimuli strength may vary from less than (subthreshold) to greater than (suprathreshold) the RMT. The interstimulus interval (ISI) is the time in ms between the CS and TS and is the key determinant of the response observed.

Short-Interval Intercortical Inhibition (SICI)

Principal types of local intracortical inhibition can be studied using paired-pulse TMS. Kujirai and co-workers³³ discovered that the response evoked by suprathreshold test stimulus (TS) given 1–6  ms after subthreshold conditioning stimulus (CS) is inhibited, compared to the response evoked by the test stimulus alone. This inhibitory phenomenon is called short-interval intracortical inhibition (SICI). As shown in the EMG data from a single 12-year-old participant (Fig. 1.5A), the test stimulus alone elicits an EMG response of about 1-mV peak-to-peak amplitude. The superimposed responses by TS alone (gray line) and paired pulses (solid line) are given at 2  ms after a conditioning stimulus. The paired-pulse response was significantly reduced. In addition, this inhibition is present for ISI of 1–6  ms. This range of ISIs, where the percentage control size (paired-pulse response divided by test alone response) is below 100%, describes the SICI phenomenon.

Figure 1.5  Short-interval intercortical inhibition and facilitation in the FDI muscle.

Raw traces showing test MEP (gray traces) and (A) SICI (ISI  =  2  ms) and (B) ICF (ISI  =  10  ms) shown by black traces. Each of the traces is an average of 10 EMG traces.

Intracortical Facilitation (ICF)

Using paired-pulse TMS, two categories of facilitations within M1 have been identified. Using subthreshold conditioning stimulus (CS) to condition suprathreshold test stimulus (TS) at interstimulus interval (ISI) of 6–25  ms reveals TS MEP facilitation, which is known as intracortical facilitation (ICF). In the same experiment, where Kujrai and co-workers³³ described SICI, ICF was also exhibited for ISIs above 6  ms. As shown in Fig. 1.5B, conditioning a test response by a prior conditioning stimulus at 10  ms ISI, can result in an increase in MEP amplitude. Since these ISIs are longer than where SICI happens, it is not called short intracortical facilitation. Subthreshold facilitation of spinal motoneurons was suggested to be the cause of the facilitation of test responses and hence the cause of ICF. Though the exact mechanism has not yet been clarified, the possible mechanism mediating ICF has been suggested by Herwig and co-workers³⁴ to be the induction of slow excitatory postsynaptic potentials and/or induced intracellular signaling cascades enhancing excitability of pyramidal tract neurons by trans-synaptic activation of metabotrope receptors. However, this suggestion was based only on physiological theory and was not supported by any data.

Short-Interval Intracortical Facilitation (SICF)

Over relatively short intervals, the second type of facilitatory interaction can be demonstrated within M1, called short-interval intracortical facilitation (SICF). This occurs when a suprathreshold stimulus is followed by a subthreshold stimulus.³⁵ This phenomenon can also be exhibited when two stimuli near motor threshold are given consecutively. In relaxed subjects, when the interval between the stimuli was around 1.0–1.5  ms, 2.5–3.0  ms, or 4.5  ms or later, the size of the response to the pair of stimuli was much greater than the algebraic sum of the response to each stimulus alone. The first, second, and third peaks of facilitations were observed when the second stimulus is fixed to 70% RMT and the first stimulus was 70%, 90%, and 100% RMT, respectively.³⁶ SICF is also called I-wave facilitation as the three peaks of facilitations observed may correspond to the generation of I-waves. Di Lazzaro and co-workers tested the cortical involvement in SICF and the relation with I-waves by recording descending motor volleys directly from the cervical epidural space of five conscious patients, who had a stimulator implanted in the cervical cord for the treatment of intractable pain.³⁷ Test stimulus was set at an intensity of 2% (of stimulator output) above active motor threshold (AMT) and condition stimulus was set at AMT. At ISI of 1, 1.2, and 4  ms the amplitude of the total volleys (the sum of individual I-waves minus the responses to test stimulus alone) and of the EMG response is larger after paired stimulation. More importantly, it is clearly shown that the I1-wave is virtually unaffected, while there is a significant increase in the I2- and I3-waves. The ISIs where SICF was observed using EMG recordings might be slightly longer (∼0.5  ms) than ISIs where SICF was observed using epidural recordings. This increase could be due to the distance difference of where the recordings are taken, since epidural recordings are taken closer to the stimulation than EMG recordings.

Long-Interval Intracortical Inhibition (LICI)

The second type of local intracortical inhibitory phenomenon that occurs at longer ISI is known as long-interval intracortical inhibition (LICI). LICI is shown by suppression of the MEP to a test stimulus when it is preceded 50–200  ms by a conditioning stimulus that is above motor threshold.³⁸ Wassermann and co-workers delivered paired stimuli to motor cortex with a circular coil at 1.1 AMT, with various ISIs ranging between 20–200  ms. The active muscle experiments were performed while holding 10% maximum voluntary contraction of wrist extensors of the right arm. The ratio conditioned MEP to the MEP evoked by test stimulus alone was below 100% at all ISIs between 20 and 200  ms, indicating the activation of LICI. The mechanism underlying LICI is similar to that of the cortical silent period (CSP), since the inhibition in both measurements is induced by a suprathreshold stimulus. CSP may represent the duration of inhibition induced by suprathreshold CS, while LICI may reflect its magnitude. Since the spinal inhibitory mechanisms are exerted mainly during the early part (up to 50  ms) of CSP, LICI at early ISIs is presumably mediated by the supraspinal mechanisms, but partial spinal influences cannot be entirely excluded.³⁸

Interhemispheric Inhibition (IHI)

In this stimulation paradigm, TMS over the primary motor cortex (M1) of one hemisphere affects the response of the opposite hemisphere to the TMS. Accordingly, this methodology is unique in that it requires the simultaneous placement of two separate TMS coils, one over each motor cortex. Ferbert and colleagues³⁹ stimulated the left hemisphere M1 (condition stimulus) followed by simulation of the right hemisphere M1 (test stimulus). The MEP in the distal hand muscles by test TMS was inhibited at all ISIs between 6 and 30  ms. They proposed this inhibition is produced at the cortical level via a transcallosal route. The claim was later supported by epidural recordings where I2- and later I-wave are suppressed by conditioning stimuli on the other side of hemisphere, but not D-wave or I1-wave.³⁷ More detailed studies of IHI have defined both short- (approximately 8–10  ms ISI) and long- (40–50  ms ISI) interval versions of IHI.⁴⁰ Mechanisms of IHI may be similar to those of the ipsilateral silent period described above, though differences have also been reported. Our studies of typically developing children suggest similar IHI mechanisms are present by school-age though relative symmetry between the dominant and non-dominant directions may only be established in adolescence (unpublished observation).

TMS Paired With Sensory Activation

Sensory inputs are integrated with motor control commands at spinal, subcortical, and cortical levels and as a result directly or indirectly influence changes in neuronal excitability and reorganization as assessed using TMS. For instance, single-pulse TMS conditioned by sensory afferent inputs allows measurements of corticospinal excitability changes related to specific sensory inputs that are activated artificially through electrical or mechanical stimulation, or naturally through various motor tasks.⁴¹

Short-Latency Afferent Inhibition (SAI)

SAI is a TMS paradigm that can be used to investigate sensory–motor integration. In this paradigm, a peripheral nerve that is stimulated prior to the activation of the motor cortex reduces the size of the motor-evoked potentials (MEPs) elicited by TMS.⁴² SAI requires a minimum ISI that is close to the latency of the N20 component of a somatosensory-evoked potential, and lasts for about 7–8  ms.⁴² The pathway mediating SAI is considered to be of cortical origin based on evidence of peripheral nerve electrical stimulation 19  ms prior to TMS of the motor cortex suppressed responses evoked by TMS but not by TES. Similarly, epidural recordings from the cervical epidural space of five patients during TMS over the motor cortex confirmed that the most prominent effect of SAI was on the I2- and I3-waves, whilst the D- and I1-waves were not affected by SAI.⁴²

Sensory input can also facilitate on-going EMG activity through transcortical reflex pathways.⁴³ Using TMS, it was possible to show that MEPs evoked in the flexor pollicis longus muscle were facilitated if they were evoked during the period of the long-latency stretch reflex and not if they were evoked during the short-latency stretch reflex, even when the sizes of the two reflex components were approximately equal.⁴⁴ Furthermore, TMS at an intensity that was below threshold for evoking MEPs during the short-latency reflex period could produce a response if given within the long-latency period. Similar effects were not present when low-intensity TES was