Neuropsychological, Neurophysiological and Laboratory Markers of ...
Inhibition of Gamma Oscillations as a Neurophysiological ...€¦ · Natasha Radhu and Mawahib...
Transcript of Inhibition of Gamma Oscillations as a Neurophysiological ...€¦ · Natasha Radhu and Mawahib...
Inhibition of Gamma Oscillations as a Neurophysiological
Endophenotype of Schizophrenia
by
Natasha Radhu
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy
Graduate Department of Institute of Medical Science
University of Toronto
© Copyright by Natasha Radhu 2015
ii
Inhibition of Gamma Oscillations as a
Neurophysiological Endophenotype of Schizophrenia
Natasha Radhu
Doctor of Philosophy
Graduate Department of Institute of Medical Science
University of Toronto
2015
Abstract
Background
The pathophysiology of schizophrenia (SCZ) has not been fully elucidated. Studies have
demonstrated that SCZ patients have impairments in the dorsolateral prefrontal cortex
(DLPFC). Two findings have been shown in the DLPFC: deficits in GABAergic inhibitiory
neurotransmission and abnormal inhibition of cortical oscillations. Thus, we aimed to assess
frontal inhibition as a potential endophenotype of SCZ.
Objectives
The first objective was to quantitatively assess transcranial magnetic stimulation (TMS)
motor cortex measures of inhibition and excitation in obsessive-compulsive disorder (OCD),
major depressive disorder (MDD) and SCZ, as a meta-analysis. The second objective was to
evaluate the inhibition of overall and gamma oscillations in the DLPFC and motor cortex
iii
using TMS and EEG in SCZ and OCD. The final objective was to evaluate the inhibition of
overall and gamma oscillations in SCZ patients, OCD patients, and their unaffected first-
degree relatives with TMS and EEG.
Hypotheses
First, we hypothesized that motor cortex inhibitory deficits would be a ubiquitous finding
across OCD, MDD and SCZ patients. Second, we hypothesized that patients with SCZ would
show deficits in overall and gamma inhibition in the DLPFC compared to healthy subjects
and patients with OCD. Lastly, it was hypothesized that frontal inhibition in first-degree
relatives of SCZ would be intermediate of healthy subjects and their related probands.
Results
The first study showed that motor inhibitory deficits were a ubiquitous finding across OCD,
MDD, and SCZ. The second paper found that SCZ patients demonstrated inhibitory deficits
in the DLPFC (overall and gamma inhibition), not observed in OCD patients. This was found
to be independent of illness severity and medication. The final study demonstrated deficits in
frontal inhibition in SCZ patients, which were significantly less than their unaffected first-
degree relatives. No differences were found between first-degree relatives of SCZ and
healthy controls. First-degree relatives were intermediate of their related probands and
healthy controls. We did not show inhibition deficits in OCD patients and their first-degree
relatives.
iv
Conclusions
Frontal inhibition (measured via TMS and EEG) may be an essential neurophysiological
process that is impaired in SCZ. Multi-site trials are needed to investigate inhibition as a
potential endophenotype for SCZ.
v
Acknowledgements
I am extremely thankful to everyone who has been for me through every step of the way
during my PhD and would like to express my acknowledgements to each of you.
First and foremost, I would like to thank my PhD supervisor, Dr. Zafiris Jeffrey Daskalakis. I
am so fortunate to have Dr. Daskalakis as my mentor as he is a very well-rounded and
brilliant scientist, physician, and teacher. He has been a strong influence in my life not only
academically, while also demonstrating strong commendable family values. I am very
grateful for his guidance throughout my graduate education and his continuous admiration
for my abilities in pursuing a career in neuroscience. I look forward to future collaborations
with Dr. Daskalakis.
I would like to thank two esteemed members of my program advisory committee
Dr. Robert Chen and Dr. James Kennedy who have always provided their extensive feedback
and have continually set aside time to provide their expertise regarding my PhD project.
Also, I am very thankful to my examiners, Dr. Richard Staines, Dr. Sean Kidd, and
Dr. Daniel Mueller for their valued suggestions and comments to help strengthen my PhD
dissertation. Thank you to our collaborators on this work: Dr. Daniel Blumberger,
Dr. Faranak Farzan, Dr. Danilo De Jesus, Dr. Margaret Richter, Dr. Lakshmi Ravindran, Dr.
Paul Fitzgerald and Dr. Tiffany Greenwood. Also, thank you to Anosha Zanjani for applying
her creative artistic talent to this work. All of the figures in this dissertation are beautiful due
to Anosha’s dedicated efforts and endless perfection.
vi
To the entire team at the Temerty Centre for Therapeutic Intervention, thank you for your
hard work and assistance during my doctoral work. The lab environment has always been
both productive and fun. In particular, many thanks to Dr. Luis Garcia Dominguez for his
superb Matlab abilities and always thinking critically about the data. I am very appreciative
to Mawahib Semeralul for all of her prompt help and for making the heritability project a true
victory. I am very enthusiastic to be a co-CEO with Mawahib one day.
My family’s support has made this work exceptionally possible and would be impossible
without them. I would like to thank my mother Usha Radhu and my father Prem Radhu for
instilling in me the value of education, for teaching me their morals and always allowing for
me to put my education before everything else. My parents have given me unconditional love
and extreme generosity throughout my graduate career; words can’t express how thankful I
am to both my mother and father for giving me the opportunity to pursue my dreams. I would
also like to thank Anita Idrees (my sister), Adnan Idrees (my brother-in-law), Jasmine Idrees
(my niece), Isha Idrees (my niece), Reena Radhu (my sister), Mark Snyder (my brother-in-
law), and Anya Snyder (my niece) as their infinite support and love has been so important to
me throughout my PhD journey, I am forever appreciative.
Last but definitely not the least; I would like to thank my husband, Manoj Gandhi. His
passion for mathematics, his drive for education, his zest for life and his endless love to me
have all been significant motivators to my success. I am very excited to begin a new chapter
in my life with Manoj Gandhi right by my side.
vii
Contributions
Literature Review Chapters
Chapter 1, Inhibition of the Cortex Using Transcranial Magnetic Stimulation in Psychiatric
Populations: Current and Future Directions
Authors: Radhu N, Ravindran LN, Levinson AJ, Daskalakis ZJ.
All authors of this paper published in the Journal of Psychiatry and Neuroscience (Natasha
Radhu, Lakshmi Ravindran, Andrea Levinson and Zafiris Daskalakis) reviewed the article
critically, approved the final version for publication and assisted with writing the paper.
Natasha Radhu and Lakshmi Ravindran conducted the systematic literature search. Natasha
Radhu, Lakshmi Ravindran and Zafiris Daskalakis designed the study, analyzed the papers
and organized the structure of the paper. Anosha Zanjani created and designed the figures for
this paper.
Chapter 2, Neurophysiological Measurements Associated with Transcranial Magnetic
Stimulation
Authors: Radhu N, Blumberger DM, Zanjani A, Daskalakis ZJ.
All authors of this book chapter published by Oxford University Press (Natasha Radhu,
Daniel Blumberger, Anosha Zanjani and Zafiris Daskalakis) reviewed the chapter critically,
approved the final version for publication and assisted with writing the chapter. Natasha
Radhu conducted the systematic literature review search. Natasha Radhu and Zafiris
Daskalakis designed the study, analyzed the papers and organized the structure of the
chapter. Anosha Zanjani created and designed the figures for this book chapter.
Chapter 3, Schizophrenia and Their First-Degree Relatives
Author: Radhu N
This chapter was written solely by Natasha Radhu to serve as part of the literature review for
the dissertation.
viii
Chapter 4, Research Aims and Hypotheses of the Experiments
Author: Radhu N
This chapter was written solely by Natasha Radhu to state the objectives and hypotheses for
the three original research articles.
Original Research Articles
Chapter 5, A Meta-Analysis of Cortical Inhibition and Excitability Using Transcranial
Magnetic Stimulation in Psychiatric Disorders
Authors: Radhu N, de Jesus DR, Ravindran LN, Zanjani A, Fitzgerald PB, Daskalakis ZJ.
All authors of the meta-analysis published in the Journal of Clinical Neurophysiology
(Natasha Radhu, Danilo de Jesus, Lakshmi Ravindran, Anosha Zanjani, Paul Fitzgerald,
Zafiris Daskalakis) reviewed the paper critically, approved the final version for publication
and contributed with writing the paper. Natasha Radhu and Zafiris Daskalakis designed the
study. Natasha Radhu acquired the data from published studies, analyzed the data, interpreted
the results and created the forest plots for illustrating the effect size data. Anosha Zanjani
provided expertise with the effect size (Hedge’s G) analysis and assisted with all figures for
this publication.
Chapter 6, Evidence for Inhibitory Deficits in the Prefrontal Cortex in Schizophrenia
Authors: Radhu N, Garcia Dominguez L, Farzan F, Richter MA, Semeralul MO, Chen R,
Fitzgerald PB, Daskalakis ZJ.
All authors of this paper published in Brain (Natasha Radhu, Luis Garcia Dominguez,
Faranak Farzan, Margaret Richter, Mawahib Semeralul, Robert Chen, Paul Fitzgerald, Zafiris
Daskalakis) reviewed the paper critically, approved the final version for publication and
contributed with writing the paper. Natasha Radhu and Zafiris Daskalakis designed the study
and interpreted the findings. Natasha Radhu and Mawahib Semeralul completed the
recruitment. Mawahib Semeralul screened all research participants using a structured clinical
interview and TMS safety screener questionnaire. Natasha Radhu acquired and pre-processed
the data. Natasha Radhu and Luis Garcia Dominguez completed the final analyses of the
ix
data. Luis Garcia Dominguez created the figures for this publication. Faranak Farzan
provided consultation and expertise with the EEG analysis.
Chapter 7, Investigating the Heritability of Cortical Inhibition in First-Degree Relatives and
Probands in Schizophrenia
Authors: Radhu N, Garcia Dominguez L, Greenwood TA, Farzan F, Semeralul MO, Richter
MA, Kennedy JL, Blumberger DM, Chen R, Fitzgerald PB, Daskalakis ZJ.
Manuscript Submitted
All authors of the manuscript (Natasha Radhu, Luis Garcia Dominguez, Tiffany Greenwood,
Faranak Farzan, Mawahib Semeralul, Margaret Richter, James Kennedy, Daniel Blumberger,
Robert Chen, Paul Fitzgerald, and Zafiris Daskalakis) reviewed the paper critically, approved
the final version for publication and contributed with writing the paper. Natasha Radhu and
Zafiris Daskalakis designed the study as well as completed the interpretation of the results.
Natasha Radhu and Mawahib Semeralul completed the recruitment for the study. Mawahib
Semeralul screened all research participants using a structured clinical interview and TMS
safety screener questionnaire. Natasha Radhu acquired and pre-processed the data. Natasha
Radhu and Luis Garcia Dominguez designed and completed the final analyses of the data.
Tiffany Greenwood completed the heritability analyses and provided genetic expertise.
James Kennedy provided genetic expertise with the data. Luis Garcia Dominguez created the
figures for this publication. Faranak Farzan provided consultation and expertise with the
EEG analysis.
Chapter 8, General Discussion, Future Directions and Conclusions
Author: Radhu N
This chapter was written solely by Natasha Radhu to serve as the discussion for the
dissertation.
x
Table of Contents
Abstract…………………………………………………………………………………….ii-iv
Acknowledgments………………………………………………………………………….v-vi
Contributions……………………………………………………...………………………vii-ix
List of Abbreviations………………………………………………………………..…xviii-xix
List of Tables……………………………………………………………………...…………xx
List of Figures…………………………………………………………………………xxi-xxiv
Chapter 1 Inhibition of the Cortex using Transcranial Magnetic Stimulation in
Psychiatric Populations: Current and Future Directions………………...…………..1
1.1 Abstract……………………………………………………………………………………2
1.2 Introduction………………………………………………………………………………..3
1.3 Evaluating Cortical Inhibition with Transcranial Magnetic Stimulation………………….4
1.3.1 Cortical Silent Period……………………………………………………………………6
1.3.2 Short Interval Cortical Inhibition……………………………………………………..…6
1.3.3 Long Interval Cortical Inhibition………………………………………………………..7
1.3.4 Interhemispheric Inhibition……………………………………………………………...7
1.4 Transcranial Magnetic Stimulation as a Method to Measure Excitability………………...8
1.4.1 Resting Motor Threshold………………………………………………………………..8
1.4.2 Active Motor Threshold…………………………………………………………………9
1.4.3 Intracortical Facilitation…………………………………………………………………9
1.5 Motor Cortex Inhibition in Psychiatric Disorders……………………………………….11
1.5.1 Schizophrenia…………………………………………………………………………..11
1.5.2 Bipolar Disorder………………………………………………………………………..13
xi
1.5.3 Major Depressive Disorder………….....………………………………………………14
1.5.4 Obsessive-Compulsive Disorder……………………………….………………………15
1.5.5 What Are the Implications of These Findings?…..……………………………………16
1.5.6 Applications beyond the Motor Cortex…………………………...……………………17
1.6 TMS-EEG Studies of Inhibition………………………………………………………....18
1.6.1 Assessing Connectivity with TMS and EEG…………………………………………..18
1.7 Functional Consequences of Disordered Inhibition……………………………….……..19
1.8 Limitations……………………………………………………………………………….21
1.9 Summary of Findings…………………………………………………………………….22
Chapter 2 Neurophysiological Measurements Associated with Transcranial
Magnetic Stimulation …………………………………………………………..…..........24
2.1 Abstract…………………………………………………………………………………..25
2.2 Introduction………………………………………………………………………………26
2.3 Overview of TMS Technology…………………………………………………………..26
2.3.1 Motor Cortex TMS Studies in Psychiatric Illnesses and Clinical Utility……………...26
2.4 Applications of TMS and EEG……………………………………………………..……27
2.4.1 Overview of EEG………………………………………………………………………27
2.4.2 Overview of TMS and EEG…………………………………………………………....28
2.4.3 Advantages of TMS and EEG………………………………………………………….30
2.4.4 Rhythms of the Brain as Measured by EEG…………………………………………...30
2.4.5 Single-Pulse TMS Combined With EEG …………………………………………...…31
2.4.6 Paired-Pulse TMS Combined With EEG………………………………………………32
2.4.7 Application of TMS-EEG in Sleep Studies……………………………………………33
2.4.8 Application of TMS-EEG in Loss of Consciousness Studies……………………….…34
2.5 How Can We Apply Combined TMS and EEG in Psychiatric Disorders…………….....34
2.6 Limitations...............................................................................................................…......37
Chapter 3 Schizophrenia and Their First-Degree Relatives…………………….....38
3.1 Abstract…………………………………………………………………………………..39
xii
3.2 Introduction………………………………………………………………………………40
3.3 The Relationship between GABA-Mediated Inhibition and Gamma Oscillations……....40
3.3.1 Gamma Oscillations, GABAergic Inhibition and Working Memory………………….42
3.4 Inhibitory Deficits in Schizophrenia……………………………………………………..43
3.4.1 NMDA-receptor Hypofunction Hypothesis……………………………………………43
3.4.2 Neuroanatomic Findings ………………………………………………………………44
3.4.3 Magnetic Resonance Spectroscopy Studies…………………………………………....45
3.4.4 The N100 Response Evoked by TMS……………..…………………………………...45
3.4.5 Main Findings…………………………………...……………………………………..46
3.5 Endophenotypes in Schizophrenia……………………………………………………….46
3.5.1 Heritability of Schizophrenia…………………………………………………………..47
3.5.2 Applications of Endophenotypes………………………………………………………47
3.5.3 P50 Suppression……………………………………………………………………......48
3.5.4 Anti Saccade Paradigm………………………………………………………………...49
3.5.5 Prepulse Inhibition……………………………………………………………………..49
3.5.6 Auditory Event-Related Potentials…………………………………………………….50
3.5.7 Auditory-Related N100 Findings…………………………………………..........……..51
3.5.8 Mismatch Negativity…………………………………………………………………...51
3.6 The Bipolar-Schizophrenia Network on Intermediate Phenotypes Research……………52
3.6.1 Neuroimaging and EEG B-SNIP Studies……………………………………………...53
3.7 Transcranial Magnetic Stimulation Studies with First-Degree Relatives of Schizophrenia
Patients ………………………………………………………………………………………54
3.8 Outline of the Dissertation…………………………………………………………..…...55
Chapter 4 Research Aims and Hypotheses of the Experiments………………......56
4.1 Introduction………………………………………………………………………………57
4.2 A Meta-Analysis of Cortical Inhibition and Excitability Using Transcranial Magnetic
Stimulation in Psychiatric Disorders (Chapter 5) Objectives and Hypotheses………………58
4.2.1 Objective 1, Inhibitory Deficits Present in Severe Psychiatric Disorders……………..58
4.2.2 Hypothesis 1, Inhibitory Deficits Present in Severe Psychiatric Disorders……………58
xiii
4.3 Evidence for Inhibitory Deficits in the Prefrontal Cortex in Schizophrenia (Chapter 6)
Objectives and Hypotheses………………………………………………………………..…58
4.3.1 Objective 1, Replication of Frontal Inhibitory Deficits………………………………..59
4.3.2 Hypothesis 1, Replication of Frontal Inhibitory Deficits………………………………59
4.3.3 Objective 2, Diagnostic Specificity of Frontal Inhibitory Deficits…………………….59
4.3.4 Hypothesis 2, Diagnostic Specificity of Frontal Inhibitory Deficits…………………..59
4.3.5 Objective 3, Large Effect Size Differences via Cohen’s D……………………………60
4.3.6 Hypothesis 3, Large Effect Size Differences via Cohen’s D…………………………..60
4.3.7 Objective 4, Trait Stability of Frontal Inhibition………………………………………60
4.3.8 Hypothesis 4, Trait Stability of Frontal Inhibition……………………………………..60
4.4 Investigating the Heritability of Cortical Inhibition in First-Degree Relatives and
Probands in Schizophrenia (Chapter 7) Objectives and Hypotheses………….…………..…60
4.4.1 Objective 1, Assessing Inhibition in Unaffected First-Degree Relatives…………...…60
4.4.2 Hypothesis 1, Frontal Inhibition in Unaffected First-Degree Relatives……………….61
Chapter 5 A Meta-Analysis of Cortical Inhibition and Excitability Using
Transcranial Magnetic Stimulation in Psychiatric Disorders…………………...62
5.1 Abstract…………………………………………………………………………………..63
5.2 Introduction………………………………………………………………………………64
5.3 Inhibitory TMS Paradigms…………………………………………...………………….64
5.3.1 Excitatory TMS Paradigms………………………………………………………….…65
5.4 Applications within Psychiatric Disorders…………………………………………….…65
5.5 Methods……………………………………………………………………………….….66
5.5.1 Data Sources…………………………………………………………………………...66
5.5.2 Study Selection……………………………………………………………………...…66
5.5.3 Data Extraction…………………………………………………………………….…..67
5.5.4 Hedge's g Calculation for the Meta-Analysis………………………………………….67
5.5.5 Test of Heterogeneity………………………………………………………………..…67
5.5.6 N Fail-Safe …………………………………………………………………………….68
5.6 Results……………………………………………………………………………………68
xiv
5.6.1 Patients with OCD……………………………………………………………………..69
5.6.2 OCD - Resting Motor Threshold………………………………………………………69
5.6.3 OCD – SICI………………….………………………………………………………...69
5.6.4 OCD - Intracortical Facilitation…………………………………………………..……70
5.6.5 OCD - CSP …………………………………………………………………………….70
5.7 Patients with MDD………………………………………………………………………72
5.7.1 MDD - Resting Motor Threshold……………………………………………………...72
5.7.2 MDD – SICI……………………………………………………………………………72
5.7.3 MDD - Intracortical Facilitation……………………………………………………….73
5.7.4 MDD - CSP………………………………………………………………………….…73
5.7.5 MDD - Motor Evoked Potential Amplitude………………………………...…………73
5.8 Patients with SCZ……………………………………………………………………..…75
5.8.1 SCZ - Resting Motor Threshold…………………………………………………….....75
5.8.2 SCZ – SICI……………………………………………………………………………..76
5.8.3 SCZ - Intracortical Facilitation……………………………………………......……….78
5.8.4 SCZ – CSP……………………………………………………………………………..79
5.8.5 SCZ - Motor Evoked Potential Amplitude………………………………………….…80
5.9 Discussion……………………………………………………………………………..…80
5.9.1 Clinical Implications…………………………………………………………………...82
5.9.2 Limitations……………………………………………………………………………..83
5.9.3 Main Findings Summarized….………………………………………………….……..84
Chapter 6 Evidence for Inhibitory Deficits in the Prefrontal Cortex in
Schizophrenia.......................................................................................................................85
6.1 Abstract…………………………………………………………………………………..86
6.2 Introduction ……………………………………………………………………………...87
6.3 Materials and Methods….………………………………………………………………..90
6.3.1 Clinical Severity………………………………………………………………………..96
6.3.2 Transcranial Magnetic Stimulation Data Recording…………………………………...96
6.3.3 Localization of the Motor Cortex……………………………………………………...96
xv
6.3.4 Localization of the DLPFC………………………………………………………..…...96
6.3.5 Electromyography Recording……………………………………………………….…97
6.3.6 EEG Recording and Pre-Processing………………………………………………...…97
6.4 Results………………………………………………………………………………..…103
6.4.1 Comparing Single and Paired-Pulse Conditions (Within-Group Analysis)……...…..103
6.4.2 Between-Group Results for DLPFC Stimulation…………………………………….104
6.4.3 Local Grid of Electrodes Analysis for DLPFC Stimulation………………………….104
6.4.4 Effect Size for DLPFC Stimulation…………………………………………………..105
6.4.5 Motor Cortex Stimulation………………………………………………………….....105
6.4.6 Local Grid of Electrodes Analysis for Motor Cortex Stimulation…………………....105
6.4.7 Sum of T-Score Topology………...……………………………………………….…105
6.4.8 Effect of Medication Treatments…………………………………………………..…106
6.4.9 Clinical Severity Correlation Analysis..................................................................…...106
6.5 Discussion ………………………………………………………………………..…….113
6.6 Frontal LICI Deficits in SCZ…………………………………………………………...113
6.6.1 Frontal Gamma LICI Deficits in SCZ……………………………………………..…114
6.6.2 Neurophysiology of OCD………………………………………………………….....115
6.7 Advancements in Analyses ………………………………………………………….....116
6.8 Limitations …………………………………………………………….……………….117
6.9 Clinical Implications…………………………………………………….…………...…117
Chapter 7 Investigating the Heritability of Cortical Inhibition in First-Degree
Relatives and Probands in Schizophrenia……………………………..…………….118
7.1 Abstract………………………………………………………...……………………….119
7.2 Introduction……………………………………………………………………………..120
7.3 Materials and Methods……………………………………………………………….…123
7.3.1 Clinical Severity………………………………………………………………………127
7.3.2 Data Recording……………………………………………………………………….128
7.3.3 Transcranial Magnetic Stimulation…………………………………………………...128
7.3.4 Localization of the Motor Cortex…………………………………………………….128
xvi
7.3.5 Localization of the DLPFC…………………………………………………..……….128
7.3.6 Electromyography Recording………………………………………………..……….129
7.3.7 EEG Recording and Pre-Processing……………………………………………….…129
7.4 Post-Processing Analyses …………………………………………………...…………130
7.4.1 Calculating Inhibition by Subject…………………..…………………...……………130
7.4.2 Between-Group Analyses………………………………….....………………………131
7.4.3 Assessing Clinical Severity and Effects of Medication Analyses in Schizophrenia
Patients............................................................................................................................…...132
7.4.4 Evaluating Clinical Severity in First-Degree Relatives of Schizophrenia…………....132
7.4.5 Stratification of Age in First-Degree Relatives of Schizophrenia Patients………...…132
7.4.6 The Heritability of Inhibition in Schizophrenia………………………...…………….133
7.5 Results………………………………………………………………..…………………133
7.5.1 Frontal Overall (2-50 Hz) Inhibition………………………………………………….133
7.5.2 Frontal Gamma (30-50 Hz) Inhibition………………………………………………..134
7.5.3 Motor Cortex Overall (2-50 Hz) Inhibition ……………………………………..…...136
7.5.4 Motor Cortex Gamma (30-50 Hz) Inhibition………………………… ……………...137
7.5.5 Stratification of Age in First-Degree Relatives of Schizophrenia Patients…………...137
7.5.6 Clinical Severity Analysis in First-Degree Relatives of Schizophrenia Patients….....137
7.5.7 Effect of Antipsychotic Medications and Anti-Deprssant Medications……….......…137
7.5.8 Clinical Severity in Schizophrenia Patients……………………………………..……138
7.6 Discussion…………………………………………………………….......…………….141
7.6.1 Frontal Inhibition in First-Degree Relatives of Schizophrenia Probands…..……..….141
7.6.2 Endophenotypes (Intermediate Phenotypes) in Schizophrenia……………………….142
7.6.3 Frontal Inhibitory Deficits in Schizophrenia……………………………………...….143
7.7 Limitations…………………………………………………………………………..….144
7.8 Summarizing the Main Findings …….……………………..……………………..……145
Chapter 8 General Discussion, Future Directions and Conclusions…..………..146
8.1 Summary of the Literature Review……………………………………………………..147
8.1.1 Summary of the Original Research…………………………………………….……..147
8.2 How Can Neuroscience Revolutionize Psychiatry?........................................................148
xvii
8.2.1 Combined TMS and EEG ……………………………………………………………149
8.2.2 Why Implement TMS-EEG?........................................................................................149
8.3 Advancements in Analyses ……………………………………………….……………149
8.3.1 Detecting and Removing Artifacts in EEG ………………………………….……….150
8.3.2 Cluster-Based Analyses………………………………………………………..……..150
8.3.3 LICI Analyses…………………………………………………………….…………..151
8.4 Pathophysiology of Schizophrenia………………………..……………………………152
8.4.1 GABAergic Deficits in Schizophrenia………………...…...…………………………152
8.5 Pathophysiology of OCD…………………………………………………………….…154
8.6 Neurophysiological Biomarkers in Clinical Practice…………………….……………..156
8.6.1 Assessing Biological First-Degree Relatives of Patients………………….………….157
8.7 Limitations………………………………………………………………………...……158
8.8 Future Directions…………………………………………………………...…………..160
8.9 Conclusions……………………………………………………………………………..161
References or Bibliography……………………………………………………………..….162
xviii
List of Abbreviations
APB abductor pollicis brevis
BD bipolar disorder
CI cortical inhibition
CS conditioning stimulus
CSP cortical silent period
DF degrees of freedom
DLPFC dorsolateral prefrontal cortex
DSM diagnostic and statistical manual of mental disorders
ECT electroconvulsive therapy
EEG electroencephalography
EMG electromyography
EPSP excitatory post synaptic potential
fMRI functional magnetic resonance imaging
GABA gamma-aminobutyric acid
GAD glutamic acid decarboxylase
ICA independent component analysis
ICF intracortical facilitation
IHI interhemispheric inhibition
IPSP inhibitory post synaptic potential
ISI interstimulus interval
ISP interhemispheric signal propagation
LICI long interval cortical inhibition
LTP long-term potentiation
MDD major depressive disorder
MEP motor evoked potential
MRI magnetic resonance image
MRS magnetic resonance spectroscopy
mV millivolt
xix
NMDA N-methyl D-aspartate
OCD obsessive-compulsive disorder
PAS paired associative stimulation
PPI prepulse inhibition
RMT resting motor threshold
S1 stimulus one
S2 stimulus two
SCZ schizophrenia
SICF short interval cortical facilitation
SICI short interval cortical inhibition
SSRI Selective serotonin reuptake inhibitor
tDCs transcranial direct current stimulation
TS test stimulus
TRD treatment-resistant depression
TMS transcranial magnetic stimulation
YBOCS Yale-Brown obsessive compulsive scale
WM working memory
5-HT serotonin
xx
List of Tables
Table 1. Summary of TMS Paradigms……………………………...………………………...4
Table 2. Summary of Significant TMS Findings in Psychiatric Populations………………..23
Table 3. Number of Included and Excluded Studies………………………………………...68
Table 4. Summary of Significant Hedge's g Results in Psychiatric Populations….…………84
Table 5. Description of the Psychotropic Medications Displayed as Number of
Subjects/Dose(s)………………………………………………………………………...…...91
5A. Patients with Schizophrenia Medication Details………………………………………..91
5B. Patients with Obsessive-Compulsive Disorder Medication Details. ……………………93
Table 6. Diagnostic Information for Schizophrenia and Obsessive-Compulsive Disorder
patients……………………………………………………………………………………….95
Table 7. All p-values of the between-group comparisons by site of stimulation, frequency
band, and electrode grids. The primary analysis is the first line; the second line displays the
pooled variance analysis…………………………………………..…………………..……101
Table 8. Description of the Psychotropic Medications Displayed as Number of
Subjects/Dose(s)…………………………………………………………………………....125
Table 8A. Patients with Schizophrenia Medication Details…………………………......…125
Table 8B. Patients with Obsessive-Compulsive Disorder Medication Details …………….126
Table 9. Diagnostic Information for Schizophrenia and Obsessive-Compulsive Disorder
patients………………………………………………………………………………….......127
xxi
List of Figures Figure 1. Surface electromyography recordings from a right hand muscle…………….........10
(1A) A single test stimulus (TS) applied to the left motor cortex producing a motor evoked
potential (MEP).
(1B) The cortical silent period (CSP) is induced following a 40% suprathreshold TS applied
to the left motor cortex while the right hand muscle is tonically activated. The CSP starts at
the onset of the MEP and ends with the return of motor activity.
(1C) Short-interval cortical inhibition (SICI) occurs when a conditioning stimulus (CS)
precedes the TS by 2 ms to and inhibits the MEP produced by the TS.
(1D) Long-interval cortical inhibition (LICI) occurs when the CS precedes the TS by 100 ms
and inhibits the MEP produced by the TS.
(1E) Intracortical facilitation (ICF) occurs when the CS precedes the TS by 20 ms,
facilitating the MEP produced by the TS.
Figure 2. A single TMS pulse is applied to the motor cortex activating cortical tissues
associated with the abductor pollicis brevis muscle, eliciting a motor evoked potential at the
periphery captured through electromyography……………………………………………....22
Figure 3. This illustration demonstrates that patients with SCZ have selective deficits in the
inhibition of gamma (30 - 50 Hz) oscillations in the dorsolateral prefrontal cortex compared
to healthy controls using interleaved TMS and EEG……………………………………...…35
Figure 4. Forest Plot of the Hedge's g Analysis For All Studies That Included Patients With
Obsessive-Compulsive Disorder Compared to Healthy Controls…………………….……..71
(A) Resting Motor Threshold
(B) Short-interval Cortical Inhibition
(C) Intracortical Facilitation
(D) Cortical Silent Period
Figure 5. Forest Plot of the Hedge's g Analysis For All studies That Included Patients With
Major Depressive Disorder Compared to Healthy Controls. ………………………………..74
(A) Resting Motor Threshold
(B) Short-interval Cortical Inhibition
(C) Intracortical Facilitation
xxii
(D) Cortical Silent Period
Figure 6. Forest Plot of Resting Motor Threshold Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls……………………...76
Figure 7. Forest Plot of Short-Interval Cortical Inhibition Hedge's g Analysis For All Studies
That Included Patients With Schizophrenia Compared to Healthy Controls. ………………77
Figure 8. Forest Plot of Intracortical Facilitation Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls……………………...78
Figure 9. Forest Plot of Cortical Silent Period Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls……………………...79
Figure 10. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in healthy subjects when
stimulating the dorsolateral prefrontal cortex.……………………………….......…………107
Figure 11. Statistical Significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in schizophrenia patients when
stimulating the dorsolateral prefrontal cortex......…………………………………………..107
Figure 12. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in obsessive-compulsive disorder
patients when stimulating the dorsolateral prefrontal cortex……………………………….108
Figure 13. Strength of inhibition by electrode. Each value consists of the sum of all t-scores
of inhibition in the major cluster of inhibition in the time-frequency maps for each electrode.
These plots show the three groups by frequency bands in the dorsolateral prefrontal cortex.
Values have been normalized within each frequency band. The color bar is omitted since
only the pattern matters, as the actual sum is dependent on the resolution of the time-
frequency-spatial domain………………………………………………..………………….108
Figure 14. Strength of inhibition by electrode. Each value consists of the sum of all t-scores
of inhibition in the major cluster of inhibition in the time-frequency maps for each electrode.
These plots show the three groups by frequency bands in the motor cortex. Values have been
normalized within each frequency band. The color bar is omitted since only the pattern
matters, as the actual sum is dependent on the resolution of the time-frequency-spatial
domain.……………………………………………………………………………………...109
xxiii
Figure 15. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in healthy subjects when
stimulating the motor cortex. ……………………………………………………………...110
Figure 16. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in schizophrenia patients when
stimulating the motor cortex………………………………………………...…...…………110
Figure 17. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in obsessive-compulsive disorder
patients when stimulating the motor cortex.………………………………………………..111
Figure 18. Effect size (Cohen’s d) of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode when comparing schizophrenia
and healthy subjects in the dorsolateral prefrontal cortex. Red corresponds to when healthy
subjects show greater inhibition, blue corresponds to when SCZ show greater inhibition...112
Figure 19. The index of frontal inhibition from a cluster analysis at different thresholds of p-
values in healthy controls, first-degree relatives of schizophrenia patients and their related
probands. A cluster analysis was performed for each group by sampling a subset 19 of
subjects with replacement. The procedure was repeated 2000 times for each threshold. Error
bars indicate one standard deviation………………………………….……………..…..….135
Figure 20.The index of frontal inhibition from a cluster analysis at different thresholds of p-
values in healthy controls, first-degree relatives of obsessive-compulsive disorder patients
and their related probands. A cluster analysis was performed for each group by sampling a
subset 13 of subjects with replacement. The procedure was repeated 2000 times for each
threshold. Error bars indicate one standard deviation………………………………………136
Figure 21. The frequency of significant values for each group, summarized from subject data,
on each voxel for all the nine central electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2).
The threshold for significance was chosen to be p < 0.01. Each graph corresponds to healthy
subjects, first-degree relatives of SCZ patients, and their related probands. The stimulation
area was the dorsolateral prefrontal cortex. Values are masked over the left bottom area (dark
navy blue) indicating that those windows of the wavelet analysis, which contains points from
the pre-stimulus interval……………………………...…………………………………….139
xxiv
Figure 22. The frequency of significant values for each group, summarized from subject data,
on each voxel for all the nine central electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2).
The threshold for significance was chosen to be p < 0.01. Each graph corresponds to healthy
subjects, first-degree relatives of obsessive-compulsive disorder patients, and their related
probands. The stimulation area was the dorsolateral prefrontal cortex. Values are masked
over the left bottom area (dark navy blue) indicating that those windows of the wavelet
analysis, which contains points from the pre-stimulus interval.............................................140
1
Chapter 1
Inhibition of the Cortex using Transcranial Magnetic Stimulation in
Psychiatric Populations: Current and Future Directions
Contents of this chapter have been reprinted by permission from the Journal of Psychiatry
and Neuroscience
“Inhibition of the Cortex using Transcranial Magnetic Stimulation in Psychiatric Populations:
Current and Future Directions” Reprinted from the Journal of Psychiatry and Neuroscience
November 2012; (37) (6), Pages 369-378 by permission of the publisher. © 2014 Canadian
Medical Association.
Radhu N, Ravindran LN, Levinson AJ, Daskalakis ZJ. (2012). Inhibition of the cortex using
transcranial magnetic stimulation in psychiatric populations: current and future directions.
Journal of Psychiatry & Neuroscience. 37(6): 369-378.
A link to the published paper can be found at:
http://www.jpn.ca/vol37-issue6/37-6-369/
2
1.1 Abstract Several lines of evidence suggest that deficits in gamma-aminobutyric acid (GABA)
inhibitory neurotransmission are implicated in the pathophysiology of schizophrenia (SCZ),
bipolar disorder (BD), major depressive disorder (MDD), and obsessive-compulsive disorder
(OCD). Cortical inhibition (CI) refers to a neurophysiologic process, whereby GABA
inhibitory interneurons selectively attenuate pyramidal neurons. Transcranial magnetic
stimulation (TMS) represents a non-invasive technique to measure CI, excitability, and
plasticity in the cortex. These measures were traditionally limited to the motor cortex which
is a significant limitation when non-motor neurophysiological processes are of primary
interest. Recently, TMS has been combined with electroencephalography (EEG) to derive
such measurements directly from the cortex. This review will focus on neurophysiological
studies related to inhibitory and excitatory paradigms linking dysfunctional GABAergic
neurotransmission to disease states. We review evidence that suggests CI deficits among
psychiatric populations and conclude by discussing the future directions of TMS,
demonstrating the potential to identify biological markers of neuropsychiatric disorders.
3
1.2 Introduction TMS is an important neurophysiological tool that allows researchers to non-invasively study
the cortex of healthy individuals and in patients with neuropsychiatric disorders [Barker et
al., 1985]. TMS is used to understand the neurobiology of cognitive function, behaviour and
emotional processing [McClintock et al., 2011] by assessing neurophysiological markers of
inhibition, excitation and plasticity [Classen et al., 1998; Kujirai et al., 1993]. In 1985,
Barker et al. introduced TMS as a tool for investigating the functional state of the motor
pathways in patients with neurological disorders and healthy participants [Barker et al.,
1985]. It involves the generation of a magnetic field through the use of an electromagnetic
coil connected to a TMS device which induces an electrical current in the brain [Wagner et
al., 2007]. They demonstrated that a single TMS pulse applied to the motor cortex could
activate cortical tissues associated with the hand or leg muscles and this activation could
elicit motor evoked potentials (MEP) at the periphery captured through electromyography
(EMG) [Barker et al., 1985] (figure 1A). Recently, TMS has been combined with EEG to
evaluate the effects of electromagnetic induction on cortical oscillations (figure 3) [Fuggetta
et al., 2005; Paus et al., 2001; Rosanova et al., 2009]. This review will emphasize the
neurophysiological evidence underlying psychiatric disorders through the application of
TMS and demonstrate the functional consequences of disordered inhibition. A literature
search was performed using PubMed from 1990 through December 2011, Ovid Medline
from 1990 through December 2011, Embase Psychiatry from 1990 through December 2011,
and PsycINFO from 1990 through December 2011. The following search terms were
used: transcranial magnetic stimulation, TMS, TMS-EEG, psychiatry, psychiatric
disorder, neuropsychiatric disorder, schizophrenia, bipolar
disorder, mania, depression, major depressive disorder, obsessive-compulsive disorder,
cortical inhibition, cortical silent period, short interval cortical inhibition, long interval
cortical inhibition, interhemispheric inhibition, cortical excitability, resting motor threshold,
active motor threshold, intracortical facilitation, motor evoked potential amplitude,
interhemispheric signal propagation, plasticity, paired-associative stimulation, and use-
dependent plasticity.
4
1.3 Evaluating Cortical Inhibition with Transcranial Magnetic
Stimulation CI refers to a neurophysiological process, whereby gamma-aminobutyric acid (GABA)
inhibitory interneurons selectively attenuate the activity of other neurons (e.g., pyramidal
neurons) in the cortex [Daskalakis et al., 2007]. Pyramidal cell activity is coordinated
through a balance of inhibitory postsynaptic potentials (IPSPs) and excitatory postsynaptic
potentials (EPSPs) [Krnjevic, 1997]. IPSPs are generated by GABAergic interneurons
terminating on the pyramidal cell [Krnjevic, 1997]. GABA is the main inhibitory
neurotransmitter in the brain regulating the modulation of cortical excitability and neural
plasticity [DeFelipe et al., 1986; Schieber and Hibbard, 1993]. The following TMS
paradigms will be described implicating GABAergic inhibitory neurotransmission: cortical
silent period (CSP), short interval cortical inhibition (SICI), long interval cortical inhibition
(LICI), and interhemispheric inhibition (IHI). Table 1 provides a summary of these TMS
paradigms.
Paradigm Definition Neurotransmitter
System Involved
Cortical
Silent Period
(CSP)
CSP is measured by stimulating the
contralateral motor cortex with a single
suprathreshold pulse in a moderately tonically
active muscle (i.e., 20% of maximum
contraction), resulting in the interruption of
voluntary muscle contraction [Cantello et al.,
1992; Kujirai et al., 1993].
GABAB
Short-Interval
Cortical
Inhibition
(SICI)
SICI is a paired pulse paradigm, whereby, a
subthreshold conditioning stimulus (CS) is
applied to the motor cortex before a
suprathreshold test stimulus (TS) at
interstimulus intervals (ISIs) between 1
millisecond (ms) to 4 ms. The subthreshold
CS suppresses the motor-evoked potential
GABAA
5
(MEP) produced by the TS [Cantello et al.,
1992; Kujirai et al., 1993].
Long-Interval
Cortical
Inhibition
(LICI)
LICI refers to the pairing of a suprathreshold
CS followed by a suprathreshold TS at long
ISIs (e.g. 50 - 100 ms), resulting in inhibition
of the MEP produced by the TS [Claus et al.,
1992; Valls-Sole et al., 1992].
GABAB
Interhemisph
eric Inhibtion
(IHI)
IHI is measured using two magnetic
stimulating coils, whereby, a suprathreshold
TMS pulse delivered to one hemisphere can
inhibit the MEP response to a suprathreshold
TMS pulse delivered within 6 to 50 ms to the
opposite hemisphere [Ferbert et al., 1992;
Gerloff et al., 1998].
GABAB
Resting
Motor
Threshold
(RMT)
RMT is defined as the minimal intensity
(single pulse) that produces a MEP > 50 μV in
5 of 10 trials in a relaxed muscle [Kujirai et
al., 1993].
Glutamate
Active Motor
Threshold
(AMT)
AMT is defined as the first intensity (single
pulse) that produces an MEP of > 100 μV in 5
of 10 trials in an isometrically moderately
active muscle [Chen et al., 1998].
Glutamate
Intracortical
Facilitation
(ICF)
ICF is a paired pulse paradigm that can be
used to index excitability of the excitatory
circuits in the motor cortex, whereby,
conditioning stimuli are applied to the motor
cortex before the TS usually at ISIs between
7ms – 20ms [Nakamura et al., 1997].
Glutamatergic
NMDA receptors
Table 1. Summary of TMS Paradigms.
6
1.3.1 Cortical Silent Period CSP is measured by stimulating the contralateral motor cortex in a moderately tonically
active muscle (i.e., 20% of maximum contraction) with TMS intensities of 110% - 160% of
the resting motor threshold (RMT), resulting in the interruption of voluntary muscle
contraction [Cantello et al., 1992; Kujirai et al., 1993] (figure 1B). The duration of CSP is
measured from MEP onset to the return of any voluntary EMG activity, ending with a
deflection in the EMG waveform [Tergau et al., 1999]. Studies have demonstrated that CSP
is related to GABAB receptor-mediated inhibitory neurotransmission as it displays a similar
time course to the GABAB receptor-induced IPSP, approximately 150 to 200 ms post
stimulus [McCormick, 1989; Roick et al., 1993; Siebner et al., 1998; Werhahn et al., 1999b].
For instance, administration of tiagabine, a GABA reuptake inhibitor, leads to an increased
concentration of GABA in the synaptic cleft and predominately activates GABAB receptors
[Thompson and Gahwiler, 1992], which has resulted in a dose-dependent prolongation of the
CSP [Werhahn et al., 1999b]. Furthermore, baclofen, a potentiator of GABAB receptor-
mediated inhibitory neurotransmission, was also found to lengthen the CSP [Siebner et al.,
1998].
1.3.2 Short Interval Cortical Inhibition The SICI paradigm was first reported by Kujirai et al. [Kujirai et al., 1993], involves a
subthreshold conditioning stimulus (CS) set at 80% of RMT that precedes a suprathreshold
test stimulus (TS), adjusted to produce an average MEP of 0.5–1.5 millivolt (mV) peak-to-
peak amplitude in the contralateral muscle (figure 1C) [Cantello et al., 1992; Kujirai et al.,
1993]. To measure SICI conditioning stimuli are applied to the motor cortex before the TS at
interstimulus intervals (ISIs) between 1 millisecond (ms) to 4 ms as the subthreshold CS
suppresses the MEP produced by the TS (figure 1C). Research suggests that SICI is related to
GABAA receptor-mediated inhibitory neurotransmission [Ziemann et al., 1996b] as it has
been demonstrated that SICI is increased by medications that facilitate GABAA
neurotransmission such as lorazepam [Ziemann et al., 1996a]. Baclofen (GABAB agonist) has
also been shown to decrease SICI [McDonnell et al., 2006], possibly related to presynaptic
GABAB autoreceptors [Daskalakis et al., 2002b]. Moreover, SICI is related to GABAA
7
receptor-mediated inhibitory neurotransmission as it displays a similar time course to the
GABAA receptor-induced IPSP. For example, Wang and Buzsaki [Wang and Buzsaki, 1996]
showed through computer simulations that the synaptic time constant for GABAA receptors
approximately ranges from 10 to 25 ms, confirming that SICI is associated with activity of
GABAA receptor-mediated inhibitory neurotransmission.
1.3.3 Long Interval Cortical Inhibition LICI refers to the pairing of a suprathreshold CS followed by a suprathreshold TS at long
ISIs (e.g. 50 - 100 ms), resulting in inhibition of the MEP produced by the TS (figure 1D)
[Claus et al., 1992; Valls-Sole et al., 1992]. Studies strongly suggest that LICI is mediated by
slow IPSPs via activation of GABAB receptors [McDonnell et al., 2006; Sanger et al., 2001;
Werhahn et al., 1999a]. For example, 50 mg of baclofen orally administered to 9 healthy
participants, resulted in enhanced LICI implying that the increase in LICI is likely a result of
increased GABAB receptor- mediated IPSPs [McDonnell et al., 2006]. Also, LICI is optimal
when the CS precedes the TS by 100 to 150 ms [Sanger et al., 2001] comparable to the time
course of the GABAB receptor activation, shown to typically peak around 150 to 200 ms post
stimulus [McCormick, 1989]. More recently, a significant positive relationship has been
shown between the suppression of MEP amplitudes in LICI (with an ISI of 100 ms), and the
duration of the silent period in the CSP paradigm in healthy individuals [Farzan et al.,
2010b], providing evidence for the mediation of the GABA B receptor in both LICI and CSP.
1.3.4 Interhemispheric Inhibition IHI is measured using two magnetic stimulating coils, whereby a suprathreshold TMS pulse
delivered to one hemisphere can inhibit the MEP response to a suprathreshold TMS pulse
delivered within 6 to 50 ms to the opposite hemisphere [Ferbert et al., 1992; Gerloff et al.,
1998]. Inhibitory GABAergic neurons mainly serve local circuits [Somogyi et al., 1998]; IHI
may be mediated through excitatory axons that cross the corpus callosum to act on local
inhibitory neurons in the contralateral motor cortex [Berlucci, 1990]. Daskalakis et al.
[Daskalakis et al., 2002b] demonstrated that SICI is reduced in the presence of IHI.
Furthermore, IHI is reduced in the presence of LICI when matched for test MEP amplitude
8
but no significant change is seen when matched for TS intensity. These results demonstrate
that IHI may be related to GABAB activity. This is consistent with Ziemann et al. who
showed that lorazepam increased SICI but did not change IHI, suggesting that IHI is not
related to GABAA activity [Ziemann et al., 1996a].
1.4 Transcranial Magnetic Stimulation as a Method to
Measure Excitability Glutamate and aspartate are the main excitatory neurotransmitters within the central nervous
system [Monaghan et al., 1989]. EPSPs in neurons of rat sensorimotor cortex are mediated
by α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), N-methyl-D-aspartate
(NMDA) and kainate receptors [Hwa and Avoli, 1992]. Voltage-gated sodium channels are
vital in regulating axon excitability [Hodgkin and Huxley, 1952], while non-ionotropic non-
NMDA glutamate receptors are responsible for fast excitatory synaptic neurotransmission
within the neocortex [Douglas and Martin, 1998]. The following excitatory paradigms will be
discussed which include: RMT, active motor threshold (AMT), and intracortical facilitation
(ICF).
1.4.1 Resting Motor Threshold TMS permits assessment of the RMT, defined as the minimal intensity that produces a MEP
> 50 μV in 5 of 10 trials in a relaxed muscle [Kujirai et al., 1993]. The RMT is a global
measure of corticospinal excitability and depends on the excitability of axons activated by
the TMS pulse, as well as the excitability of synaptic connections at both the cortical and
spinal level [Paulus et al., 2008a]. The RMT depends on glutamatergic synaptic excitability
[Paulus et al., 2008a]. It has been shown that drugs which block voltage-gated sodium
channels, in particular anticonvulsants such as carbamazepine, lamotrigine and losigamone,
increase RMT [Ziemann et al., 1996b]. By contrast, NMDA antagonists such as ketamine
reduce RMT [Di Lazzaro et al., 2003]. Lastly, drugs with GABAergic properties such as
vigabatrin (GABA analogue), baclofen (GABAB receptor agonist) and gabapentin (GABA
analogue) do not affect motor threshold [Ziemann et al., 1996b].
9
1.4.2 Active Motor Threshold The AMT is defined as the first intensity that produces an MEP of > 100 μV in 5 of 10 trials
in an isometrically moderately active muscle [Chen et al., 1998]. The AMT is measured
during muscle contraction, where corticospinal neurons and spinal motor-neurons are very
close to firing threshold [Paulus et al., 2008a].
1.4.3 Intracortical Facilitation ICF is a paired pulse paradigm that can be used to index excitability of the excitatory circuits
in the motor cortex, whereby, conditioning stimuli are applied to the motor cortex before the
TS at ISIs usually between 7ms – 20ms (figure 1E). It has been shown that ICF originates
from EPSPs transmitted by glutamatergic NMDA receptors [Nakamura et al., 1997]. In fact,
the latency of onset of the EPSP mediated by the NMDA receptor is approximately 10 ms,
which is consistent with the time course of ICF [Kujirai et al., 1993; Ziemann et al., 1996c].
This is supported by the majority of pharmacological studies which demonstrated that
NMDA receptor antagonists such as dextromethorphan and memantine decrease ICF
[Schwenkreis et al., 1999; Ziemann et al., 1998]. Benzodiazepines such as lorazepam
(GABAA agonist) decreases ICF [Ziemann et al., 1996a] and baclofen (GABAB agonist)
increases ICF [Ziemann et al., 1996b]. Lastly, it has also been suggested that ICF is not
exclusively mediated by excitatory interneurons, but rather by a net balance between
inhibition and excitability [Daskalakis et al., 2004].
10
Figure 1. Surface electromyography recordings from a right hand muscle.
Figure 1A. A single test stimulus (TS) applied to the left motor cortex producing a motor
evoked potential (MEP).
Figure 1B. The cortical silent period (CSP) is induced following a 40% suprathreshold TS
applied to the left motor cortex while the right hand muscle is tonically activated. The CSP
starts at the onset of the MEP and ends with the return of motor activity.
Figure 1C. Short-interval cortical inhibition (SICI) occurs when a conditioning stimulus (CS)
precedes the TS by 2 ms to and inhibits the MEP produced by the TS.
Figure 1D. Long-interval cortical inhibition (LICI) occurs when the CS precedes the TS by
100 ms and inhibits the MEP produced by the TS.
11
Figure 1E. Intracortical facilitation (ICF) occurs when the CS precedes the TS by 20 ms,
facilitating the MEP produced by the TS.
1.5 Motor Cortex Inhibition in Psychiatric Disorders Several lines of evidence suggest that deficits in GABA functioning are implicated in the
pathophysiology of SCZ, BD, MDD, and OCD. The integration of TMS with EMG has
offered a valuable tool for the assessment of the pathological processes associated with
psychiatric disorders.
1.5.1 Schizophrenia SCZ is a severe psychiatric illness characterized by delusions, hallucinations, disorganized
thinking and often life-long disability [van Os and Kapur, 2009]. SCZ is a debilitating
disorder that exacts enormous personal, social and economic costs [van Os and Kapur, 2009].
CI may represent an important mechanism responsible for the symptoms observed in patients
with SCZ. Several lines of evidence suggest that abnormalities in CI are an important
neurophysiological mechanism in SCZ and these impairments have been shown to be related
to GABAergic deficits. Benes et al. [Benes et al., 1991] first reported that patients with SCZ
have morphologic changes in cortical GABA interneurons by demonstrating a decreased
density of non-pyramidal cells (i.e., interneurons) in anterior cingulate layers II-VI and in
prefrontal cortex layer II.
Research has shown CI abnormalities in SCZ using TMS. For example, Daskalakis et al.
[Daskalakis et al., 2002a] measured MT, SICI, ICF, CSP and IHI in 15 unmedicated patients
with SCZ (14 medication-naïve and 1 medication-free for over 1 year), 15 medicated SCZ
patients and 15 healthy controls. They found that unmedicated SCZ patients had significantly
lower CI compared with healthy controls in measures of SICI, CSP and IHI providing TMS
evidence for deficient GABAergic neurotransmission in SCZ. Similarly, Fitzgerald et al.
[Fitzgerald et al., 2002b] found comparable results in 22 medicated patients with SCZ
compared with 21 healthy controls. They demonstrated significantly lower SICI and CSP
within SCZ group compared with healthy controls. Fitzgerald et al. [Fitzgerald et al., 2002a]
12
also evaluated IHI in 25 patients with SCZ and 20 healthy controls. They similarly
demonstrated a significant decrease in IHI in patients independent of medication dose. More
recent studies have also demonstrated deficits in CI using TMS in patients with SCZ. For
example, Daskalakis et al. [Daskalakis et al., 2008b] reported that 10 clozapine-treated
patients with SCZ had significantly longer CSPs compared with 10 healthy participants and 6
unmedicated SCZ patients. A subsequent study by Liu et al. [Liu et al., 2009] with a large
sample of 78 SCZ patients and 38 healthy controls confirmed that clozapine-treated SCZ
patients demonstrated a longer CSP and reduced SICI compared with healthy control
participants. However, patients treated with other antipsychotics and unmedicated patients
demonstrated a significantly shorter CSP duration. These findings suggest that CI is involved
in the pathophysiology of SCZ and that clozapine may potentiate GABAB receptor-mediated
inhibitory neurotransmission. Additionally, across all SCZ patients in this study, CSP was
inversely related to negative symptoms, while SICI was inversely associated with positive
symptoms, highlighting the role of both GABAB and GABAA receptor-mediated inhibitory
neurotransmission in SCZ. This finding is consistent with recent neurochemical evidence
demonstrating that there is a direct link between clozapine and the GABAB receptor [Wu et
al., 2011]. Wu et al. showed through autoradiography, synaptic membrane binding, and
HEK293 expression of the GABAB receptor, they showed that clozapine facilitates the
binding of CGP54626A (a specific high-affinity antagonist) to the GABAB receptor.
Furthermore, Wobrock et al. [Wobrock et al., 2010] examined 12 first-episode SCZ patients
with a history of comorbid cannabis use and 17 without. They found that patients with a
history of comorbid cannabis use had lower SICI and increased ICF but no significant
differences were found in RMT and CSP. Comorbid cannabis abuse was suggested to
potentiate the reduced SICI and enhanced ICF observed in first episode SCZ patients. This
finding is consistent with a previous study, in which Fitzgerald et al. [Fitzgerald et al.,
2009b] found that heavy and light users of cannabis demonstrated significantly decreased
SICI compared to healthy controls. Taken together, these studies provide evidence to confirm
that SCZ is associated with CI deficits in the motor cortex. Future studies are necessary to
advance current knowledge by identifying biological markers of both illness and treatment
response to developing a deeper understanding of the neurophysiological mechanisms
underlying SCZ.
13
1.5.2 Bipolar Disorder
BD is a serious psychiatric illness with prevalence estimates of 2.4% worldwide [Merikangas
et al., 2011]. It is characterized by periods of mania or hypomania alternating with phases of
depression [Benazzi, 2007] and is associated with an early age at onset, usually between 16
to 26 years [Javaid et al., 2011; Manchia et al., 2008]. Suicide and suicide attempts are
significant contributors to premature mortality and disability [Goodwin et al., 2003].
Relatively little work has been done to understand the neurophysiological underpinnings of
this disease. Limited neuroanatomical evidence demonstrates that BD patients have impaired
cortical inhibitory neurotransmission [Benes et al., 1998]. Benes and Berretta found that the
density of cortical GABA interneurons, which mediate CI, is reduced in the anterior
cingulate cortex among patients with BD [Benes and Berretta, 2001] and also found a 30%
decrease in cortical inhibitory GABAergic interneurons in BD, compared with a 16%
decrease in patients with SCZ [Benes and Berretta, 2001]. The data suggests a loss of
GABAergic interneurons in both BD and SCZ. However, there is little in vivo
neurophysiological evidence supporting such impairments in BD. Levinson et al. [Levinson
et al., 2007] used TMS to evaluate SICI, CSP and IHI in 15 BD patients (13 medicated with a
single mood stabilizer and two unmedicated) compared to 15 healthy control participants.
They found that BD patients demonstrated significant deficits in SICI, CSP and IHI
compared with healthy individuals. The authors concluded that GABAergic inhibitory
neurotransmission is deficient in the motor cortex of patients with BD. Furthermore, the
majority of patients were medicated and the evidence suggested that these inhibitory deficits
were attenuated with treatment. Nevertheless, additional studies are needed with large
unmedicated samples, and more severely ill patient populations. It would be hypothesized
that any inhibitory deficits would be magnified under these conditions.
14
1.5.3 Major Depressive Disorder MDD is one of the most prevalent psychiatric disorders, and is estimated to affect 16.6% of
individuals in their lifetime [Kessler et al., 2005]. It not only affects physical and cognitive
functions but also has a profound impact on psychosocial well-being [Kessler et al., 2005].
Preclinical work has demonstrated that chronic stress induces compensatory changes in the
GABAergic system in animal models [Acosta and Rubio, 1994]. Evidence suggests that
MDD may be associated with abnormalities in cortical excitability, and more specifically
deficits in CI. For example, Fitzgerald et al. [Fitzgerald et al., 2004a] assessed cortical
excitability prior to a trial of repetitive TMS (rTMS) treatment in MDD patients. This study
included 60 patients with treatment-resistant depression (TRD), of which, 46 were medicated
during the trial (antidepressants, mood stabilizers, and antipsychotics). The authors found a
decreased SICI of the right motor cortex (1 ms ISI) and reported that an increased CSP in the
left motor cortex predicted a poorer response to rTMS treatment. Bajbouj et al. [Bajbouj et
al., 2006b] assessed 20 patients with MDD who had been washed off of medication for at
least 4 weeks compared with 20 healthy participants. They found reduced SICI and CSP in
patients with MDD, consistent with the hypothesis of deficient GABAergic tone in
depression. Similarly, Lefacheur [Lefaucheur et al., 2008] demonstrated that MDD patients
showed a reduced excitability of both excitatory (RMT, ICF) and inhibitory (CSP, SICI)
processes in the left hemisphere when compared to healthy controls. More recently, Levinson
et al. [Levinson et al., 2010] examined CI in 25 medicated individuals with TRD, 19
medicated euthymic partcipants, 16 unmedicated depressed patients and 25 healthy controls
and found that all patients with MDD, regardless of symptom or medication state,
demonstrated significant CSP deficits compared with healthy participants. Patients with TRD
also demonstrated significant deficits in SICI compared with healthy participants. The
findings above all held true after controlling for benzodiazepine use which has been shown to
affect TMS parameters [Levinson et al., 2010]. Since all MDD patients showed CSP
abnormalities but only TRD subjects additionally demonstrated SICI reductions, the authors
concluded that the depressed state may be overall associated with GABAB deficits, but severe
symptomatology, as seen in TRD, may be associated with greater deficits in both GABAA
and GABAB neurotransmission. Taken together, the above findings suggest that MDD is
15
associated with deficits in GABAergic inhibitory neurotransmission and abnormalities in
inhibitory functions of the motor cortex. Future studies are needed to explore these findings
further in cortical regions that are more closely associated with the pathophysiology of this
disorder (e.g., the dorsolateral prefrontal cortex (DLPFC)).
1.5.4 Obsessive-Compulsive Disorder OCD is a serious psychiatric illness characterized by the presence of recurrent, intrusive and
thoughts, impulses or images (obsessions) that are often also accompanied by repetitive
rituals or behaviours (compulsions) designed to counteract the associated anxiety. As
obsessive thoughts and/or rituals may cause great distress and take up significant time during
the day, OCD often leads to pronounced psychosocial impairment [Eisen et al., 2006]. It is
estimated to affect up to 2.5% of the world’s population [Karno et al., 1988; Kessler et al.,
2005]. Although, its pathophysiology remains to be fully elucidated, research suggests that
OCD may involve inhibitory deficits in orbitofrontal striatal circuits [Menzies et al., 2008].
One preliminary study found decreased SICI in 12 patients with OCD without a history of
comorbid tics or Tourette’s syndrome compared to 12 healthy control participants
[Greenberg et al., 1998], implicating a role for GABAA inhibitory neurotransmission in
OCD. These results were expanded with 9 medicated and 7 unmedicated OCD patients
compared to 11 healthy control participants [Greenberg et al., 2000]. In this case, both RMT
and AMT were found to be significantly lower in OCD patients compared to controls.
Similarly, SICI was significantly lower in OCD patients relative to healthy individuals and
this difference remained significant even when the same comparison was made with only
unmedicated OCD patients; no differences were found for SICI between unmedicated and
medicated OCD patients. By contrast, there were no differences in ICF or CSP detected
between patient and control groups. Furthermore, OCD patients with tics had significantly
less SICI than those without tics. The authors concluded that OCD, in the presence or
absence of comorbid tics, was characterized by deficient SICI.
16
More recently, Richter et al. [Richter et al., 2012] assessed a larger sample of OCD patients,
as they compared 34 OCD patients (23 medicated and 11 unmedicated) to 34 healthy
individuals. In contrast to the previous study, no overall difference was found in RMT
between OCD and healthy groups; however, RMT was significantly lower in the medicated
compared to the unmedicated OCD population. Additionally, CSP duration was also found to
be significantly shorter in OCD but no further differences were detected between the OCD
subgroups. Finally, although this study failed to detect differences in SICI between OCD and
healthy individuals, OCD subjects were found to have a significantly greater ICF, regardless
of medication status. No correlations were found between illness severity and TMS
parameters in either medicated or unmedicated OCD populations. In this case, the results
suggest that OCD is associated with deficient CSP and excessive ICF, regardless of
medication state, reflecting abnormalities in GABAB and NMDA-mediated
neurotransmission, consistent with several genetic studies that have been reported in this
disorder [Arnold et al., 2006; Dickel et al., 2006; Samuels et al., 2011; Stewart et al., 2007;
Voyiaziakis et al., 2011; Zai et al., 2005]. The authors suggested that differences between
their results and those previously published could be due to the greater number of
unmedicated OCD patients and elevated symptom severity included in their data or to the
different stimulation intensities used to elicit measures. The discrepant findings in the limited
number of studies highlight the need for further research to better characterize the potential
abnormalities seen in OCD.
1.5.5 What Are the Implications of These Findings? These findings provide compelling evidence to suggest that GABAergic inhibitory deficits
are closely involved in the pathophysiology of SCZ, MDD, OCD and BD. Taken together,
research has suggested that patients with SCZ have demonstrated impairments in SICI, CSP
and IHI. Moreover, two studies have showed that clozapine-treated SCZ patients
demonstrated significantly longer CSP durations, implicating the role of the GABAB receptor
in clozapine and demonstrating a specific response profile of treatment in this disorder
[Daskalakis et al., 2008b; Liu et al., 2009]. Similarly, previous research has demonstrated
that selective serotonin reuptake antidepressants normalize GABAergic deficits in depression
17
as assessed by the TMS paradigms through enhanced SICI and decreased ICF [Manganotti et
al., 2001; Minelli et al., 2010]. Along the same lines, OCD patients have demonstrated
decreased SICI, CSP and enhanced ICF, independent of medication status [Greenberg et al.,
2000; Greenberg et al., 1998; Richter et al., 2012]. However, BD patients have showed
impairments in SICI, CSP and IHI, not ameliorated with treatment [Levinson et al., 2007].
These studies suggest an overall lack of GABAergic inhibitory neurotransmission in these
psychiatric disorders, however, each may have a distinct profile of response to treatment.
Research in this direction is needed for more objectively based diagnostic methods and novel
treatment.
1.5.6 Applications beyond the Motor Cortex The neurophysiological studies mentioned above demonstrate the conventional approaches to
measuring CI and excitability of the motor cortex. Such approaches have been used to
demonstrate important neurophysiological findings in both healthy and diseased states.
However, the restriction of such recordings over the motor cortex is of limited interest since
the pathophysiology of many psychiatric disorders is associated with non-motor brain
regions. As a result, it is important to evaluate the neurophysiology of brain regions that are
more proximal to the underlying phenotype (e.g., the DLPFC). Recently, TMS has been
combined with EEG to derive inhibitory measurements directly from the DLPFC and the
motor cortex in healthy subjects [Daskalakis et al., 2008c; Farzan et al., 2010b; Fitzgerald et
al., 2008]. LICI can be measured using a combination of paired-pulse TMS and EEG to study
how GABAB receptors modulate oscillations in the brain in both the motor cortex and
DLPFC with high test-retest reliability [Daskalakis et al., 2008c; Farzan et al., 2010b;
Fitzgerald et al., 2008]. LICI using TMS-EEG is defined using the area under rectified
unconditioned and conditioned waveforms for averaged EEG recordings between 50-150 ms
post-TS. This interval was chosen as it represents the earliest artifact free data (i.e., 50 ms
post TS) and reflects the duration of GABAB receptor-mediated IPSPs (i.e., 250 ms post CS)
[Deisz, 1999b]. Gamma oscillations (30 to 50 Hz) in the cortex are generated as a result of
rapid firing of output pyramidal neurons. Inhibitory interneurons exert fine control over the
firing of pyramidal neuronal networks, which translates into high frequency gamma
18
oscillatory activity on EEG [Sohal et al., 2009]. Several reports also suggest that different
GABA receptor subtypes are active during different phases of gamma oscillations. It has
been shown that GABAA IPSPs contribute to generation of gamma oscillations and GABAB
IPSPs contribute to the modulation of gamma oscillations [Bartos et al., 2007; Whittington et
al., 1995].
1.6 TMS-EEG Studies of Inhibition Two studies have used combined TMS and EEG to examine the pathophysiology of SCZ.
Farzan et al. [Farzan et al., 2010a] demonstrated that overall LICI using TMS-EEG in SCZ
patients did not differ significantly in any region compared with BD patients and healthy
controls. However, when the evoked EEG response was filtered into different frequency
bands, they found a significant deficit in the inhibition of gamma oscillations in the DLPFC
of SCZ patients relative to BD patients and healthy controls, but no inhibitory deficit was
found within the motor cortex. The authors concluded that this selective deficit in the
inhibition of gamma oscillations demonstrates that the DLPFC is a region in the brain closely
related to the pathophysiology of SCZ. Deficits in frontal gamma inhibition of the DLPFC
are consistent with neurophysiological evidence of frontal impairments implicated in SCZ as
deficits in cognitive functions, such as working memory (WM), a major feature of SCZ
[Weinberger et al., 1986]. An earlier study provided additional support for this finding as
Ferrarelli et al. [Ferrarelli et al., 2008] also demonstrated a decrease in EEG-evoked
responses in the gamma band when TMS was applied directly to the frontal cortex,
suggesting frontal gamma deficits in SCZ patients. Taken together, these studies point to
important new directions in which TMS-EEG can provide new insights into the
neurophysiological underpinnings of SCZ.
1.6.1 Assessing Connectivity with TMS and EEG Current pathophysiological theories of SCZ emphasize the role of altered brain connectivity
[Friston, 1998; Stephan et al., 2006]. This disconnectivity may manifest anatomically,
through structural changes of association fibers at the cellular level, or functionally through
aberrant control of synaptic plasticity[Stephan et al., 2006]. TMS combined with EEG can be
19
used to evaluate the connectivity between and within hemispheres [Voineskos et al., 2010],
providing potential to ascertain functional connectivity between cortical regions
[Bloom and Hynd, 2005; Gazzaniga, 2000]. Voineskos et al. [Voineskos et al., 2010]
examined the relationship between microstructural integrity of subdivisions of the corpus
callosum with TMS-induced interhemispheric signal propagation (ISP) using a single-pulse
paradigm. They found a significant inverse relationship between microstructural integrity of
genu fibers of the corpus callosum and TMS-induced ISP from left to right DLPFC. Further,
they found a significant inverse relationship between microstructural integrity of callosal
motor fibers with TMS-induced ISP from left to right motor cortex. The authors concluded
that the examination of corpus callosum microstructure in relation to TMS-induced ISP may
provide novel insight into the neurobiological mechanisms of severe psychiatric disorders,
such as SCZ. Research has shown that during early cortical development reelin plays an
important role in lamination of the cortex. Reelin is a protein that regulates cortical
pyramidal neurons, interneurons, and Purkinje cell positioning [Curran and D'Arcangelo,
1998; Rice and Curran, 2001]. In SCZ, reelin was found to be decreased in layers I and II of
the prefrontal cortex [Impagnatiello et al., 1998]. Furthermore, Costa et al.[Costa et al., 2001]
found in patients with SCZ a downregulation of reelin expression and attenuated dendritic
spine expression that in turn reduce cortico-cortical connectivity and glutamic acid
decarboxylase 67 expression. These findings may explain the deficits in GABAergic
inhibitory neurotransmission and the subtle disruptions in connectivity found in SCZ. Future
research may consider evaluating the relationship between LICI and ISP, hypothesizing a
strong relationship between deficits frontal gamma inhibition and a lack of TMS-induced ISP
in patients with SCZ.
1.7 Functional Consequences of Disordered Inhibition Plasticity in the human cortex involves the functional reorganization of synaptic connections
in an effort to change or to adapt throughout life is characterized by processes involved in
learning, memory and neural repair [Hallett, 2000]. Evidence suggests that neural plasticity
may also be a corollary of CI as mechanisms mediating plasticity include unmasking existing
cortico-cortical connections [Schieber and Hibbard, 1993] by removing cortical inhibitory
20
neurotransmission [Jacobs and Donoghue, 1991]. For example, in humans administration of a
GABAergic agonist disrupts plasticity [Butefisch et al., 2000]. Abnormalities in brain
plasticity, possibly related to abnormal CI, have been proposed to underlie the
pathophysiology of SCZ [Fitzgerald et al., 2004b; Oxley et al., 2004]. We will discuss paired
associative stimulation (PAS) and use-dependent plasticity as a way of measuring plasticity
in the cortex. The evidence suggests that decreased neural plasticity is even more pronounced
in SCZ patients with impaired CI.
PAS represents a neurophysiologic paradigm that involves peripheral nerve stimulation of
the median nerve, followed by TMS of the contralateral motor cortex. PAS has been shown
to result in long-term potentiation-like activity (PAS-LTP) if peripheral nerve stimulation
precedes TMS by 25 ms (PAS-25) [Stefan et al., 2000]. Rajji et al. [Rajji et al., 2011]
demonstrated MEP potentiation after PAS-25 which was associated with enhanced motor
learning at 1-week post-PAS in healthy participants. Moreover, Frantseva et al. [Frantseva et
al., 2008] demonstrated that SCZ patients showed deficits in MEP facilitation indicating
disrupted LTP-like plasticity associated with impaired motor skill learning compared to
healthy participants. This study highlighted the role of PAS-TMS in the motor regions to
assess synaptic plasticity in SCZ patients. The authors concluded that SCZ patients
demonstrated impaired LTP-like plasticity which may be associated with deficits in learning
and memory.
Use-dependent plasticity involves the use of a TMS paradigm which can measure neural
plasticity in the cortex [Classen et al., 1998]. The spontaneous direction of TMS-induced
thumb movements is measured in two axes (x and y). As a result, use-dependent plasticity is
assessed using a task in which individuals are trained to perform a simple motor task
opposite to the direction of TMS-induced thumb movement. TMS is then reapplied to the
cortex while evaluating the direction of the induced thumb movement over time. Classen et
al. [Classen et al., 1998] found that immediately after training, the direction of TMS-induced
movements followed the direction of training. Both GABA and NMDA receptor-mediated
neurotransmission play an important role in use-dependent plasticity [Butefisch et al., 2000].
Daskalakis et al. [Daskalakis et al., 2008a] evaluated use-dependent plasticity in 14
21
medicated and six unmedicated patients with SCZ compared with 12 healthy participants. A
significant reduction of use-dependent plasticity was demonstrated in SCZ compared with
healthy participants. That is, SCZ patients demonstrated significantly small angular
deviations in the 5-10 minutes of post-training period versus pre-training compared with
controls. The authors concluded that such abnormalities may be related to dysfunctional
neurophysiological brain processes, including LTP, that exist as a result of disturbances of
GABA, NMDA, and dopamine neurotransmission. These findings potentially account for the
aberrant motor performance demonstrated in SCZ. Taken together, these studies provide
preliminary evidence for a diminution of the neurophysiological process that mediate neural
plasticity in SCZ.
1.8 Limitations The aforementioned studies relating to deficits in CI in SCZ, BD, MDD and OCD are limited
in several ways. Numerous studies are limited to measuring the motor cortex as the exact
mechanism underlying the generation and modulation of the TMS-evoked MEPs remains
unclear. Additional limitations include small samples, differences in the TMS methodologies
between research groups, heterogeneous populations studied and an overall lack of diagnostic
specificity. Furthermore, it has been shown that medication may affect outcomes of TMS
measures. As such, the inclusion of medicated individuals on various classes of psychotropic
agents in these studies may be a significant confounder of results. Addressing these issues
systematically in future research by assessing a large sample of unmedicated psychiatric
populations will allow for a greater confidence in results and provide a more stable evidence
base for elucidating biological markers involved in psychiatric illnesses.
There are several methodological limitations of using combined TMS and EEG paradigms to
measure the cortex in human participants. TMS-evoked EEG responses may be contaminated
by muscular activity, indirect cranial reflexes, and somatosensory evoked potentials [Pope et
al., 2009; Whitham et al., 2008; Whitham et al., 2007] producing artifact in the recordings.
22
1.9 Summary of Findings TMS has provided us with the ability to evaluate cortical processes such as inhibition,
excitation and plasticity in healthy participants, which has led to invaluable evidence in
elucidating the pathophysiology of neuropsychiatric disorders (figure 2).
Figure 2. A single TMS pulse is applied to the motor cortex activating cortical tissues
associated with the abductor pollicis brevis muscle, eliciting a motor evoked potential at the
periphery captured through electromyography.
Taken together, the literature has demonstrated that disorders such as SCZ, BD, MDD and
OCD are characterized by abnormalities in CI, highlighting the lack of GABAergic
inhibitory neurotransmission (summary of findings in table 2). It is important to assess the
neurophysiology of brain regions that are more proximal to the underlying phenotype (e.g.,
the DLPFC). Additionally, the ability to evaluate the response profiles of different oscillatory
frequency bands via EEG in response to TMS may ultimately serve as a key method to
identify endophenotypes of psychiatric illness. Endophenotypes are valuable as they are
presumably upstream in the pathophysiology of the illness and closer to the genetic variation
underlying complex psychiatric disorders [Gottesman and Gould, 2003]. Compared to
23
current subjective clinical diagnoses, endophenotypes are objective, quantifiable, and
heritable. They also allow for measurement of aberrant neural circuitry [Braff et al., 2008;
Gottesman and Gould, 2003]. In conclusion, there is a great need to better understand the
neurobiological underpinnings of psychiatric disorders for more objective diagnosis and for
the potential of treatment discovery.
Disorder Main Findings
SCZ Impairments in SICI, CSP, IHI [Daskalakis et al., 2002a; Fitzgerald et
al., 2002a; Fitzgerald et al., 2002b].
Two studies have shown that clozapine-treated SCZ patients
demonstrated potentiated CSP durations [Daskalakis et al., 2008b; Liu
et al., 2009].
SCZ patients with history of comorbid cannabis show decreased SICI
and increased ICF [Wobrock et al., 2010].
Deficits in gamma inhibition of the DLPFC using LICI with combined
TMS and EEG [Farzan et al., 2010a].
BD Impairments in SICI, CSP, IHI [Levinson et al., 2007].
MDD Deficits in SICI and CSP [Bajbouj et al., 2006b; Fitzgerald et al., 2004a;
Lefaucheur et al., 2008; Levinson et al., 2010].
Impairments in ICF and RMT [Lefaucheur et al., 2008].
OCD Decreased SICI, CSP and enhanced ICF [Greenberg et al., 2000;
Greenberg et al., 1998; Richter et al.].
Table 2. Summary of Significant TMS Findings in Psychiatric Populations.
24
Chapter 2
Neurophysiological Measurements Associated with Transcranial
Magnetic Stimulation.
Contents of this chapter have been reprinted by permission from Oxford University Press
Radhu N, Blumberger DM, Zanjani A, Daskalakis ZJ. (2014). Neurophysiological
Measurements Associated with Transcranial Magnetic Stimulation. Clinical Guide to the
Administration of Transcranial Magnetic Stimulation for Neuropsychiatric Disorders. (1):
98-116.
A link to the published chapter can be found at:
http://oxfordmedicine.com/view/10.1093/med/9780199926480.001.0001/med-
9780199926480-chapter-8
25
2.1 Abstract Previous studies have used TMS as a way to evaluate severe psychiatric disorders. However,
these measures were conventionally limited to the motor cortex. This past decade has seen
significant improvements in the concurrent use of TMS and EEG to assess cortical network
properties such as CI, cortical excitability, plasticity and connectivity in non-motor regions.
New hardware solutions and advanced data processing techniques have allowed substantial
reduction of the TMS-induced artifact on EEG recordings. In this chapter, the past, present
and future status of TMS-EEG research is discussed. First, a description of the working
principle of TMS is provided. Second, commonly used TMS paradigms are defined while
providing evidence for the cortical mechanisms underlying each method. Finally, this chapter
highlights the application of these novel cutting-edge techniques vis à vis healthy and disease
states to provide a clear platform from which diagnostic procedures can be developed.
26
2.2 Introduction TMS is a cutting-edge non-invasive neurophysiological tool used to investigate the cortex in
healthy and disease states [Barker et al., 1985]. TMS is a useful method to further understand
the neurobiology of cognitive function, behavior and emotional processing [McClintock et
al., 2011]. It involves the generation of a magnetic field through the use of an
electromagnetic coil connected to a TMS device which induces an electrical current in the
brain [Wagner et al., 2007]. TMS is used as an investigational tool as it assesses a variety of
cortical phenomena including CI, excitation and plasticity [Classen et al., 1998; Kujirai et al.,
1993]. Assessing the cortical phenomena using TMS provides valuable insights into the
neurophysiological substrates underlying psychiatric and neurological disorders. However,
the restriction of such recordings to the motor cortex is of limited interest as the
pathophysiology of many neuropsychiatric disorders is associated with the frontal cortex.
Thus, evaluating the neurophysiology of brain regions that are more proximal to the
underlying phenotype is essential.
2.3 Overview of TMS Technology TMS capitalizes on the ability of time-varying magnetic fields to induce eddy currents in
biological tissue via Faraday’s principle of electromagnetic induction. TMS fields pass
through the scalp unimpeded and non-invasively stimulates brain areas compared to more
invasive transcranial electrical stimulation [Hallett, 2000]. Conventional approaches to
measure cortical neurophysiology involve stimulation of the motor cortex while using MEPs
as the primary dependant variable of interest, which is measured in the periphery through
EMG. Such approaches have been used to demonstrate important neurophysiological
findings in both healthy and disease states, which will be discussed in this chapter.
2.3.1 Motor Cortex TMS Studies in Psychiatric Illnesses and Clinical
Utility A series of studies have reported that TMS paradigms that generate a functional index of
GABA inhibitory neurotransmission from the cortex of healthy human subjects have
27
demonstrated a distinct and consistent pattern of deficiency in severe psychiatric disorders.
These paradigms show high test-retest reliability and large effect size differences between
healthy subjects and patient populations. These tests are relatively easy to perform,
inexpensive, and easy to interpret. Several lines of evidence suggest that CI is impaired in
these disorders. For example, previous TMS studies have demonstrated deficits in CI
assessed from the motor cortex in patients with OCD [Greenberg et al., 2000; Greenberg et
al., 1998; Richter et al., 2012], MDD [Bajbouj et al., 2006b; Fitzgerald et al., 2004a;
Lefaucheur et al., 2008; Levinson et al., 2010], SCZ [Daskalakis et al., 2002a; Daskalakis et
al., 2008b; Fitzgerald et al., 2002a; Fitzgerald et al., 2002b; Fitzgerald et al., 2003; Liu et al.,
2009; Wobrock et al., 2010; Wobrock et al., 2009; Wobrock et al., 2008] and BD [Levinson
et al., 2007]. Collectively, these studies provide evidence to suggest that impairments in
GABA inhibitory neurotransmission are a ubiquitous finding in severe psychiatric illnesses.
TMS paradigms hold potential as biomarkers of psychiatric disorders and treatment response.
Biomarker development will lead to strategies that prevent manifestation of the illness and
increase our understanding of the underlying neurobiological mechanisms. However, further
replication of findings is required. The use of TMS to establish molecular engagement of
novel psychopharmacological and somatic treatments (i.e., electroconvulsive therapy (ECT),
repetitive TMS, magnetic seizure therapy, transcranial direct current stimulation or cognitive
behaviour therapy), particularly within the GABA and glutamate circuits, are other potential
biomarker roles for these tests. Conceivably TMS measures of GABAergic and glutamatergic
functioning could be used as biological markers of novel treatments that are aimed at
enhancing inhibition or decreasing facilitation in the cortex.
2.4 Applications of TMS and EEG 2.4.1 Overview of EEG In the 1920s, the psychiatrist Hans Berger recorded brain waves from the surface of the
human scalp and coined the technique as EEG [Buzsaki, 2006; Swartz and Goldensohn,
1998]. Electrical activity of the cortex is measured by placing multiple electrodes along the
scalp, these electrodes record electrical signals that are primarily generated by coordinated
28
output of neurons from the scalp’s surface [Nunez and Srinivasan, 2006]. Cortical potentials
recorded through EEG represent the oscillatory activity of underlying neuronal activity
[Nunez and Srinivasan, 2006]. Such recordings at rest can be used clinically to diagnose
tumours, seizures, encephalopathies, and brain death and can be potentially used as
biological markers of neuropsychiatric illnesses [Babiloni et al., 2011; Sponheim et al., 2000;
Tot et al., 2002; Venables et al., 2009]. By contrast, when sensory stimuli are presented to
patients, evoked activity that is of greater electrical power is produced and recorded at the
scalp surface when compared to resting EEG recordings. Such activity can be used to
evaluate the neurophysiological mechanisms involved in the processing of emotional or
cognitive stimuli.
2.4.2 Overview of TMS and EEG The past decade has seen significant developments in the concurrent use of TMS and EEG to
directly assess cortical network properties such as CI, excitability and connectivity.
Simultaneous EEG recording during TMS stimulation was previously unattainable because
of the technological shortcomings of EEG amplifiers that would saturate for a long duration
due to the large artifact produced by the magnetic stimulation. For example, application of a
single TMS pulse would result in artifact lasting for several seconds after the pulse. Such
long lasting artifact blocks the window of time during which neurophysiological processes
such as CI occur. Through advances in EEG amplifier technology, researchers have
conducted a series of studies to examine TMS paradigms in the motor cortex through
simultaneous EEG and EMG as well as in frontal brain regions through EEG recordings.
A significant electromagnetic artifact field is generated by the TMS (at the site of
stimulation) and is several-fold larger than that produced by sensory evoked potentials on
EEG recordings [Ilmoniemi et al., 1997]. Several developments in EEG amplifier technology
have led to a reduction of this artifact. First, Ilmoniemi and colleagues reported that
decoupling of the electrode from the amplifier at or immediately before TMS can markedly
reduce the impact of the TMS stimulus artifact on EEG recordings [Ilmoniemi et al., 1997].
This was achieved through a sample-and-hold circuit that maintains amplifier output at a
29
constant level during stimulus delivery [Ilmoniemi et al., 1997]. They showed that this
modification permitted amplifier recovery within 100 μsec after the TMS [Ilmoniemi et al.,
1997]. Second, in a traditional alternating-current (AC) coupled EEG amplifier, the typical
500 mV and 50 μsec TMS pulse prevent the signal from returning to zero immediately after
the pulse. Rather, the signal that is recorded is followed by a negative deflection that can take
seconds to return to its initial state. With the introduction of direct-current coupled EEG
amplifiers, this prolonged negative swing is eliminated and immediately returns to its linear
range after the stimulus stops. Direct-current coupling has become available only in recent
years with the introduction of fast 24-bit analogue digital converter resolution (i.e., 24nV/bit)
that is superior to the older 16-bit analogue digital converter resolution that was limited to 6.1
μV/bit, a resolution that fails to limit the TMS stimulus artifact. The third modification is to
record EEG at very high sampling rates (e.g., 20 kHz) to permit full characterization of the
TMS pulse and limit the stimulus artifact that is produced on the recordings. By using any of
these strategies, EEG recording can become TMS compatible (for a review, please refer to
Ilmoniemi and Kicic [Ilmoniemi and Kicic, 2010]). Furthermore, the EEG electrodes used
during TMS-EEG should satisfy the physical requirements to operate within the harsh TMS
environment. The electrodes must be designed with a small enough diameter to avoid
overheating or to be affected by the forces that result from the induced TMS currents. Also,
the electrodes must be coated with suitable surface material to ensure a proper interface with
skin contact [Ilmoniemi and Kicic, 2010]. It is suggested that the optimal electrodes to record
TMS-EEG are small silver/silver chloride pellet electrodes (e.g., to allow the measurement of
the electrical potential on the skin [Ives et al., 2006; Roth et al., 1992; Virtanen et al., 1999]).
There are several post-processing procedures for removal of TMS-induced artifact from the
EEG recording as extensively reviewed by Ilmoniemi and Kicic [Ilmoniemi and Kicic,
2010]. EEG amplitudes greater than 100 µV, and containing large artifacts from
electromagnetic residuals, eye blinks, eye movement or muscle activity should be rejected.
Alternatively, there are more superior methods that enable the separation of brain signals
from artifacts such as signal-space projection [Ilmoniemi and Kicic, 2010], independent
component analysis (ICA) [Hamidi et al., 2010; Korhonen et al., 2011], modeling of sources
and artifacts [Ilmoniemi and Kicic, 2010] and principal component analysis [Levit-Binnun et
30
al., 2010; Litvak et al., 2007]. Offline procedures such as the use of filters to eliminate TMS-
related artifact from EEG have also been proposed; these procedures require further
investigation [Morbidi et al., 2007]. A major limitation in using these post-processing
techniques is the fact that there is no way of verifying that neuronal activity is not being
removed along with the artifact components.
2.4.3 Advantages of TMS and EEG There are four main advantages of using TMS combined with EEG in research studies. First,
by using TMS-EEG, investigators can study the mechanisms through which MEPs are
generated and modulated. Second, online EEG recording allows for the possibility to
evaluate the effects of electromagnetic induction on cortical oscillatory activity to
appropriately identify the cortical oscillations that are closely associated with the TMS-
induced MEP generation and modulation. A third major advantage of combined TMS-EMG
and EEG is the possibility to evaluate the cortico-cortical connectivity between motor
cortices. Functional connectivity between cortical regions (e.g. left and right motor cortices)
is easily probed by measuring the propagation of TMS-induced cortical responses. TMS-
EEG methodologies permit the investigation of the frontal brain areas that are more proximal
to the underlying phenotype (non-motor regions of the cortex). For example, examining LICI
in the DLPFC enhances our understanding of the inhibitory mechanisms that underlie a
cortical area that is more closely associated with the pathophysiology of psychiatric
disorders.
2.4.4 Rhythms of the Brain as Measured by EEG Network oscillations are generated from the rhythmic and synchronized firing of output
neurons in the cortex. Oscillations can be recorded from the surface of the cortex through
EEG and are represented as five frequency bands. These bands include: delta (1 to 3.5 Hz);
theta (4 to 7 Hz); alpha (8 to 12 Hz); beta (12.5 to 28 Hz); and gamma (30 to 50 Hz). Each
frequency band is related to different states. For example, the delta and theta bands are
greatest during deep sleep and are demonstrated during wakefulness in various pathological
conditions (e.g., tumors, Alzheimer’s disease) [Babiloni et al., 2004; Babiloni et al., 2006;
31
Huang et al., 2000; Montez et al., 2009]. Alpha bands are greatest in the incipient stages of
sleep, during low arousal periods and when individuals close their eyes. Beta oscillations
show greatest activity during resting wakefulness. Gamma oscillations are associated with
the most complex cognitive demands, including information encoding, feature binding as
well as information storage and recall [Meltzer et al., 2008; Tallon-Baudry et al., 1998].
Several studies have suggested that frontal cortical gamma oscillations are necessary for WM
[Barr et al., 2011; Barr et al., 2009; Barr et al., 2010; Basar-Eroglu et al., 2007; Cho et al.,
2006; Howard et al., 2003]. Functionally, gamma oscillatory activity has been suggested to
provide the temporal dimension in information encoding [Fries et al., 2007], whereby the
successful encoding of information depends on its arrival time relative to the gamma cycle
[Fries et al., 2007]. Interneuron activity mediated by GABA then shapes the time course for
prefrontal pyramidal activation [Constantinidis et al., 2002] that is maximally activated when
the fast-spiking interneurons are not firing [Wilson et al., 1994]. GABA receptor activity is
also responsible for the generation (GABAA) and modulation (GABAB) of gamma
oscillations [Bartos et al., 2007; Brown et al., 2007; Leung and Shen, 2007; Traub et al.,
1996; Wang and Buzsaki, 1996; Whittington et al., 1995]. Thus, GABA plays a critical role
in the generation and modulation of gamma oscillations, which are vital in cognitive tasks.
2.4.5 Single-Pulse TMS Combined With EEG Ilmoniemi and colleagues were one of the first research groups to use interleaved TMS and
EEG to investigate the effect of TMS on cortical excitability [Ilmoniemi et al., 1997]. It was
demonstrated that TMS applied to the hand representation area of the human motor cortex
elicited a cortical response that spread to the adjacent ipsilateral area and to the homologous
regions in the opposite hemisphere. It was further shown that the application of TMS to the
visual cortex resulted in a similar pattern of signal propagation to the contralateral areas,
therefore providing evidence that the cortical potentials following motor cortex stimulation
were less likely to be a result of peripheral sensory activation. This original experiment
resulted in a series of studies that further characterized the EEG substrate of cortical
excitability, inhibition, plasticity, and connectivity in healthy participants [Esser et al., 2006;
32
Kahkonen et al., 2001; Kahkonen et al., 2003; Komssi et al., 2002; Komssi and Kahkonen,
2006; Nikulin et al., 2003; Paus et al., 2001; Thut et al., 2003].
2.4.6 Paired-Pulse TMS Combined With EEG Daskalakis et al. and Fitzgerald et al. were the first to demonstrate that recording LICI
(paired- pulse technique) through interleaved TMS-EEG was feasible [Daskalakis et al.,
2008c; Fitzgerald et al., 2008] in both the motor cortex and DLPFC in healthy subjects. In
the motor cortex, EEG measures of LICI were represented by the reduction of cortical
evoked activity in the electrode C3, which best represents evoked activity in the hand area of
motor cortex closest to the optimal site of abductor pollicis brevis activation through TMS
[Cui et al., 1999]. LICI was defined using the area under rectified unconditioned and
conditioned waveforms for averaged EEG recordings between 50 and 150 msec post-TS.
This interval was chosen because it represents the earliest artifact-free data (i.e., 50 msec
post-TS) and reflects the duration of GABAB receptor-mediated IPSPs (i.e., 250 msec post
conditioning stimulus) [Deisz, 1999a]. There was a significant inhibition in mean cortical
evoked activity through LICI compared to the TS alone in both the motor cortex and DLPFC
(targeted through cortical co-registration methods [Rusjan et al., 2010]. Farzan et al. has
demonstrated the validity, replicability, and test re-test reliability (Cronbach’s alpha > 0.7) of
LICI using the TMS-EEG method in both the motor cortex and DLPFC [Farzan et al.,
2010b]. In this study, a significant correlation was found between MEP suppression and
suppression of cortical evoked EEG activity [Farzan et al., 2010b]. These results provide
compelling evidence to suggest that TMS-induced EEG suppression is related to GABAergic
processes (i.e., GABAB inhibition), which mediate EMG measures of LICI [Sanger et al.,
2001; Siebner et al., 1998]. Similar research was also developed through experiments by
Fitzgerald and colleagues who used equivalent methods of assessing LICI and reported
maximal inhibition from 50 to 250 msec in the DLPFC, and from 50 to 175 msec in the
parietal lobe. They concluded that LICI can be recorded from several cortical regions with a
time course similar to GABAB receptor-mediated inhibition [Fitzgerald et al., 2009a].
33
More recently, Ferreri et al. also investigated the ability to record SICI and ICF using TMS-
EEG [Ferreri et al., 2011]. In these experiments, SICI was recorded using an ISI of 3 msec
while ICF was recorded using an ISI of 11 msec. These authors demonstrated that significant
inhibition could be reliably recorded in the motor cortex through both EMG and EEG
(recorded from the Cz electrode) and that these recordings were correlated suggesting that
such measures are mechanistically related to those recorded from peripheral hand muscles
through EMG.
2.4.7 Application of TMS-EEG in Sleep Studies Massimini and colleagues investigated cortical effective connectivity during wakefulness and
sleep using TMS with high-density EEG, evaluating the premotor cortex [Massimini et al.,
2005]. They found that during wakefulness, TMS-induced a sustained response made of
recurrent waves of activity; time-locked high frequency (20 to 35 Hz) oscillations followed
by a few slower (8 to 12 Hz) components that persisted until 300 msec. During stage 1 sleep,
this TMS-evoked response grew stronger and became shorter in duration. With the onset of
non rapid-eye movement (NREM), the TMS-induced brain response changed markedly. The
initial wave doubled in amplitude and lasted longer; however, no further TMS-locked
activity could be detected following this large wave. Based on these findings, they concluded
a breakdown of long-range effective connectivity during NREM sleep. Recently, Massimini
et al. used TMS with high-density EEG over the premotor cortex and found that during REM
sleep, the TMS-evoked brain response consisted of a sequence of fast oscillations during the
first 150 msec similar to wakefulness [Massimini et al., 2010]. They also found that activity
during stage 1 sleep replicated previous findings [Massimini et al., 2005]. Using TMS-EEG
in sleeping participants, Massimini et al. demonstrated that TMS evoked a high-amplitude
slow wave that originated under the coil and spread over the cortex; this triggered slow
waves during sleep that were state-dependent [Massimini et al., 2007]. Regardless of
stimulation site and intensity, TMS pulses that evoked slow waves during NREM and could
not do so during wakefulness. Taken together, these findings suggest that the effects of TMS-
EEG are strongly dependent on the state of the activated brain region (i.e. initial level of
34
underlying cortical activity) [Silvanto et al., 2008a; Silvanto et al., 2008b; Silvanto and
Muggleton, 2008; Silvanto et al., 2007; Silvanto and Pascual-Leone, 2008].
2.4.8 Application of TMS-EEG in Loss of Consciousness Studies Similarly, Ferrarelli and colleagues indexed TMS-evoked EEG responses in wakefulness
compared to induced loss of consciousness using midazolam [Ferrarelli et al., 2010]. Before
the injection of the anesthetic, TMS pulses to the premotor cortex evoked a complex
spatiotemporal pattern of low-amplitude high frequency activity. Conversely, following
midazolam-induced loss of consciousness, TMS pulses gave rise to high amplitude low-
frequency EEG potentials that faded shortly after the stimulation. They concluded that a
breakdown of effective cortical connectivity was a key mechanism mediating midazolam-
induced loss of consciousness. More recently, Rosonova et al. evaluated cortical effective
connectivity in patients emerging from a coma after a severe brain injury using TMS-EEG
[Rosanova et al., 2012]. They found that patients in a vegetative state who were behaviorally
awake (open-eyed) but unresponsive. TMS triggered a simple, local response, indicating a
breakdown of effective connectivity similar to unconscious, sleeping or anaesthetized
participants. In contrast, in minimally conscious patients (who were non-reflexive), TMS
triggered complex long-range activation in distant cortical areas. Taken together, the
literature indicates that TMS-EEG can evaluate effective connectivity in sleep, wakefulness,
anethetheized and vegetative states.
2.5 How Can We Apply Combined TMS and EEG in
Psychiatric Disorders? Several studies have used combined TMS and EEG to examine the pathophysiology of
psychiatric disorders. For example, Ferrarelli et al. stimulated the premotor cortex using
combined TMS-EEG and reported that reduced TMS-evoked gamma oscillations within the
first 100 msec post-stimulus in patients with SCZ. Gamma oscillations were significantly
attenuated in amplitude and demonstrated less synchrony in the fronto-central regions
[Ferrarelli et al., 2008]. The authors concluded that there was an intrinsic dysfunction in
35
frontal thalamocortical circuits in SCZ. Similarly, Farzan et al. demonstrated assessed
patients with SCZ, BD and healthy controls using the TMS-EEG paired-pulse technique (i.e.,
LICI) in both the motor cortex and DLPFC [Farzan et al., 2010a]. They found that overall
LICI (1 to 50 Hz) in SCZ patients did not differ significantly in any region when compared
with BD patients and healthy controls. However, when the evoked EEG response was filtered
into different frequency bands, they found a significant deficit in the inhibition of gamma (30
to 50 Hz) oscillations in the DLPFC of SCZ patients relative to patients with BD and healthy
controls (figure 3).
Figure 3. This illustration demonstrates that patients with SCZ have selective deficits in the
inhibition of gamma (30 - 50 Hz) oscillations in the dorsolateral prefrontal cortex compared
to healthy controls using interleaved TMS and EEG.
They also found no differences in the inhibition of other oscillatory frequencies in the
DLPFC or in the motor cortex between the three groups. The authors concluded that this
selective deficit in the inhibition of gamma oscillations demonstrates that the DLPFC is a
region in the brain that is closely related to the pathophysiology of SCZ. Furthermore,
36
Frantseva and colleagues demonstrated an increased TMS-induced cortical activation (in the
gamma frequency range) that spread across the cortex as measured by TMS-EEG in SCZ,
however, in healthy controls this activation faded away soon after stimulation [Frantseva et
al., 2012]. Recently, Hoppenbrouwers et al. showed that psychopathic offenders suffer from
dysfunctional inhibitory neurotransmission in the DLPFC as measured through combined
TMS and EEG assessing LICI [Hoppenbrouwers et al., 2013]. The authors concluded that the
impairments demonstrated in the study might render the psychopath unable to regulate
impulses, in turn, subjecting them to a disinhibited, antisocial life. Casarotto et al.
investigated frontal cortex excitability in healthy young and elderly individuals compared to
patients with Alzheimer's disease [Casarotto et al., 2011]. They found that TMS-evoked
potentials were not affected by physiological aging, unless an abnormal cognitive decline
(Alzheimer's disease) was associated. They demonstrated that frontal cortex excitability
identified as early and local cortical response to TMS was reduced in elderly patients with
Alzheimer's disease; however, this was not significantly different between healthy young and
elderly individuals. Lastly, Casarotto et al. were the first to evaluate MDD patients using
TMS-EEG in order to assess neuroplastic responses before and after their last administration
of ECT [Casarotto et al., 2013]. They demonstrated that there was a significant increase of
frontal cortical excitability (in every patient) after a course of ECT when compared to
baseline, suggesting that ECT produces synaptic potentiation. These above mentioned studies
illustrate that TMS-EEG can be used as a clinical tool to characterize the underlying
neurobiological dysfunction and to evaluate the neurophysiological effects of treatments over
time.
37
2.6 Limitations Advances in cortical stimulation and cortical recording techniques over the past few decades
have allowed for the systematic and non-invasive investigation of the neurophysiological
processes from the cortex in humans. TMS is a cutting-edge technique that allows for the
investigation of the cortical phenomena in both motor and non-motor regions to further
elucidate the pathophysiology of psychiatric disorders. Among such advancements,
concurrent TMS and EMG recordings have been instrumental in identifying and probing
cortical processes that underlie the generation and modulation of MEPs. Although the
evidence is still limited, research to date suggests that disorders such as SCZ, MDD, OCD
and BD are characterized by specific deficits in CI and abnormalities in cortical excitability.
However, the published studies are not entirely consistent. Factors that may play a role in the
discrepant results include small sample sizes, differences in TMS parameters used, the use of
heterogeneous populations, and presence of comorbid illness. Further, medications may
affect outcomes of TMS measures and it is likely that different classes of psychotropic
medications may do this in unique ways. As such, the inclusion of medicated individuals on
various classes of psychotropic agents in these studies is a significant confounder of results.
Addressing these issues systematically in future research would allow greater confidence in
results and provide a more stable evidence base for elucidating biological markers and
mechanisms involved in psychiatric illnesses. The ability to evaluate physiological response
profiles of different oscillatory frequencies in response to TMS combined with EEG may
ultimately serve as a key technique for evaluating biological markers in psychiatric illnesses.
Combined TMS and EEG will continue to provide a deeper insight into the neurobiological
underpinnings of psychiatric disorders.
38
Chapter 3
Schizophrenia and Their First-Degree Relatives
39
3.1 Abstract Several lines of evidence suggest that deficits in inhibitory neurotransmission are implicated
in the pathophysiology of SCZ. Despite more than 100 years of research in psychiatry, an
objective laboratory-based biological marker has not been identified. Biomarkers facilitate
the development of etiologic rather than symptom-based diagnostic methods, foster early
identification and treatment, and advance our understanding of the complex genetic and
neurobiological mechanisms. True biomarkers should also facilitate advanced treatment of a
population at high risk for the disorder, which may ultimately translate into prevention of
developing an illness in a subset of individuals. Endophenotypes can be used in psychiatry to
provide a way to dissect the underlying neural circuit abnormalities of complex disorders.
This chapter will first discuss the mechanisms underlying oscillatory activity in the brain
with application to the neurobiology of SCZ. The concluding portion will proceed to discuss
potential candidate neurophysiological endophenotypes currently investigated in SCZ and the
challenges involved in identifying an adequate biological marker for this disorder.
40
3.2 Introduction SCZ is a debilitating disorder that exacts enormous personal, social and economic costs [van
Os and Kapur, 2009]. SCZ affects 0.3-0.7% of the population, whose pathophysiology
remains poorly understood [McGrath et al., 2008]. SCZ is characterized by a constellation of
clinical symptoms such as delusions, hallucinations, disorganized thinking, social
withdrawal, motivational impairments and poor cognitive functioning [van Os and Kapur,
2009]. It is estimated that patients with SCZ occupy 10% of all hospital beds and despite
treatment efforts, as many as 15% of those diagnosed with SCZ eventually commit suicide
[Kaplan et al., 1994]. Despite some treatment successes, up to 45% of patients remain
treatment resistant. After 100 years of research, the neural framework of the underlying risk
of developing SCZ remains unclear and a biological marker of SCZ has not been effectively
identified. Thus, there is a great need to identify markers of disease in order to facilitate early
identification and treatment to better understand the neurobiological underpinnings of this
devastating disorder.
3.3 The Relationship between GABA-Mediated Inhibition
and Gamma Oscillations Gamma oscillations in the cortex are generated as a result of rapid firing of output pyramidal
neurons. Several lines of evidence suggest that pyramidal neuron firing is governed by
GABA inhibitory interneurons (i.e., basket and chandelier cells). GABA interneurons are
located throughout the uppermost layers of the cortex and form extensive synaptic networks
of connectivity, though limited in number (i.e., GABA interneurons only represent 20-30
percent of neurons in the cortex), one GABA interneuron typically connects extensively with
several pyramidal neurons [Bartos et al., 2007] forming neuronal networks that fire
contemporaneously. IPSPs are generated by GABAergic interneurons terminating on
pyramidal cells. As a result of this pattern of connectivity, inhibitory interneurons exert fine
control over the firing of pyramidal neuron networks, which translates into high frequency
gamma oscillatory activity on EEG [Sohal et al., 2009]. This is a crucial property of the
GABAergic system because pyramidal neurons are roughly three times as numerous in the
41
central nervous system [Bartos et al., 2007]. It has been shown that certain forms of
electrical or chemical stimulation can produce highly synchronous rhythmic IPSPs across
multiple pyramidal neurons suggesting that synchronized IPSP waves propagate throughout
cellular networks. If this synchronized activity is sufficiently large, then the amplitude of
these signals will rise above the electrophysiological noise and result in observable
oscillatory rhythms [Buzsaki, 2006]. In this way, GABA-mediated synaptic inhibition plays a
critical role in the production of neuronal synchronization in cortical circuits.
In the cortex, GABAergic interneurons have several important physiological functions, such
as the down-regulation of excessive cortical excitability (e.g. seizures) and neuroplastic
generativity, as well as serving discriminative (e.g. top down modulation) and cognitive
processes (e.g. memory). GABA and gamma oscillations mediate several cognitive functions,
including WM and attention [Daskalakis et al., 2008d]. Previous evidence has demonstrated
an association between GABAergic inhibitory neurotransmission and gamma oscillations
[Bartos et al., 2007; Bragin et al., 1995; Brown et al., 2007; Jefferys et al., 1996; Marrosu et
al., 2006; Scanziani, 2000; Traub et al., 1997; Traub et al., 1996; Wang and Buzsaki, 1996;
Whittington et al., 1995]. GABAA receptor-mediated IPSPs contribute to the generation of
gamma oscillations [Bartos et al., 2007; Wang and Buzsaki, 1996; Whittington et al., 1995].
GABAA receptors typically discharge at 30 to 50 Hz, resulting in a high-frequency on-off
oscillatory pattern of pyramidal cell discharge, recorded as gamma oscillations via EEG.
Thus, GABAA receptor-mediated IPSPs act as a switch which allows pyramidal neurons to
fire at high-frequencies. By contrast, GABAB receptor-mediated IPSPs play a critical role in
the modulation of gamma oscillations. These effects are essential to higher order cognitive
processes including information processing and WM [Benes and Berretta, 2001; Deco and
Rolls, 2003]. Furthermore, the modulation of gamma oscillations represent an important
neurophysiological process that may, in part, be responsible for optimal cognitive
functioning in the DLPFC.
42
3.3.1 Gamma Oscillations, GABAergic Inhibition and Working
Memory WM is a higher order cognitive process governed by the prefrontal cortex, which acts to
retain information (short-term) while other cortical regions organize the continuous
perceptual information processed [Baddeley, 1986; Deco and Rolls, 2003]. WM refers to the
ability to encode, manipulate, and retrieve ordered information online and over a limited
period of time, shown to be critical to language comprehension, learning and reasoning
[Baddeley, 1986; Baddeley, 1992].
GABAergic inhibitory interneurons of the prefrontal cortex are the major subtype subserving
WM [Deco and Rolls, 2003; Lewis et al., 2005]. In particular, gamma band synchrony
appears to be the neurophysiological analogue of inhibitory functioning in the prefrontal
cortex, a process critical to neurologic regulation and thus often referred to as a ‘building
block’ of brain function [Basar-Eroglu et al., 1996]. Oscillations in the gamma frequency
range are associated with WM and cognitive tasks involving prefrontal cortical function
[Palva et al., 2005]. For example, Palva and colleagues [Palva et al., 2005] demonstrated that
in healthy subjects, gamma band oscillations can be enhanced through engaging subjects in a
progressively increasing WM task load. The gamma band oscillation has been observed to be
synchronized in functionally and spatially distinct neural networks, as observed by EEG.
Synchronization of the gamma frequency spectra is seen as primarily dependent on
horizontally oriented GABAergic interneurons in the DLPFC using EEG. As shown above,
Chen et al., [Chen et al., 2014] demonstrated a positive relationship between WM-induced
gamma oscillatory activity (via EEG) and left DLPFC GABA levels. The orientation of these
neurons generates the electrical current that forms the basis of this frequency band [Tallon-
Baudry and Bertrand, 1999]. Taken together, GABAergic inhibitory interneurons may
synchronize pyramidal cell discharge from multiple cortical regions while suppressing
irrelevant or extraneous cortical activation, allowing for the integration of cortical activity
and optimized execution of cognitive information (e.g. WM).
43
3.4 Inhibitory Deficits in Schizophrenia The widespread theory in SCZ has focused on abnormal dopamine signaling due to the
primary form of treatment for SCZ (dopamine antagonists) in treatment of the positive
symptoms. However, these medications are not successful in targeting and relieving the
cognitive and negative symptoms. Since these are also core symptoms of the illness, one such
hypothesis may be due to the elevation in the ratio of cortical cellular excitation to inhibition
[Yizhar et al., 2011]. As a result, other neurotransmitter systems such as the glutamate and
the GABA systems have been implicated in the disease. For example, the increased activity
in excitatory neurons or reduction in inhibitory neuron function, may contribute to the social
and cognitive deficits observed in SCZ [Yizhar et al., 2011]. Several lines of evidence
suggest that SCZ is associated with the dysfunction of GABAergic inhibitory interneurons
and may be a mechanism through which to develop a biomarker for this disorder. The
evidence is not clear in how these GABAergic abnormalities relate to symptoms of the
illness, this section will highlight these findings.
3.4.1 NMDA-receptor Hypofunction Hypothesis The NMDA-receptor (NMDAR) hypofunction hypothesis has been one mechanism
underlying the dysfunctional GABAergic system in SCZ [Moreau and Kullmann, 2013]. It
has been shown that a blockade of the glutamate-mediated excitatory neurotransmission by
NMDAR antagonists mimics positive and negative symptoms as well as cognitive deficits in
SCZ [Krystal et al., 1994]. This hypothesis proposes a specific deficit in NMDAR signaling,
leading to a decrease in parvalbumin (PV) positive GABAergic interneuron activity and
consequent pyramidal cell disinhibition, diminishing GABA synthesis and release [Gonzalez-
Burgos and Lewis; Moreau and Kullmann, 2013; Olney et al., 1999]. The decreased
glutamatergic transmission owing to the lack of NMDAR signalling in pyramidal neurons
could decrease PV and may widely contribute to neuronal circuit hypoactivity. This has been
shown to cause a decrease in glutamic acid decarboxylase (GAD) 67 levels and pruning of
perisomatic inhibition mediated by PV basket cells [Jiao et al., 2006]. At the pyramidal cell
level a reduction in NMDAR activity results in decreased spine density, associated with
44
impairments in cognition. Lastly, disinhibition decreases the power of gamma oscillations
and may contribute to negative and cognitive symptoms in SCZ [Lisman et al., 2008].
3.4.2 Neuroanatomic Findings A potential physiological substrate for abnormal and inefficient DLPFC function is the
deficit in GABAergic neurons. Abnormalities of GABAergic interneurons are some of the
most consistent findings in SCZ [Lewis et al., 2004]. The major determinant of GABA in the
neocortex, glutamic acid decarboxylase, is consistently downregulated in postmortem studies
of SCZ patients [Torrey et al., 2005]. Benes et al. first reported that patients with SCZ have
morphologic changes in cortical GABA interneuron’s, by demonstrating a decreased density
of non-pyramidal cells (i.e., interneuron’s) in anterior cingulate layers II-VI and in prefrontal
cortex layer II [Benes et al., 1991]. Akbarian et al. also reported that in SCZ messenger
RNA, which encodes the 67-kilo-Dalton isoform of GAD, a key enzyme in the synthesis of
GABA was reduced in the DLPFC [Akbarian et al., 1995]. The decrease is specific to layers
III – V [Akbarian et al., 1995; Volk et al., 2000] and is accompanied by reduced expression
of the GABA membrane transporter 1 (also known as SLC6A1) [Volk et al., 2001]. This
reduction has also been localized to PV positive GABAergic interneurons [Hashimoto et al.,
2003] in the DLPFC [Akbarian et al., 1995], anterior cingulate cortex [Woo et al., 2004],
motor cortex [Hashimoto et al., 2008], visual cortex [Hashimoto et al., 2008] and
hippocampus [Benes et al., 2007; Knable et al., 2004]. Taken together, these studies imply
that there is impaired density, synthesis and reuptake of GABA in SCZ. It was also reported
that patients with schizoaffective disorder show a 30 percent reduction in non-pyramidal cells
whereas patients with SCZ demonstrate a 16 percent decrease. A related study of tyrosine-
hydroxylase immunoreactive cells [Benes et al., 1997] indicated a reduction of 18 percent in
the density of non-pyramidal neurons in layer II of the anterior cingulate cortex in patients
with SCZ. Together, these anatomical data suggest that both the structure and function of
GABA inhibitory interneurons are impaired in post-mortem studies of SCZ.
45
3.4.3 Magnetic Resonance Spectroscopy Studies Magnetic resonance spectroscopy (MRS) is an imaging technique, which can assay GABA
concentrations in vivo. Measurement of GABA via MRS is relatively new for SCZ research,
with the first study published in 2009 [Goto et al., 2009]. The MRS results to date are mixed,
a few studies suggest reduced GABA in SCZ [Goto et al., 2009; Rowland et al., 2013; Yoon
et al., 2010], some have shown increased GABA [Kegeles et al., 2012; Ongur et al., 2010]
and no differences between healthy subjects and SCZ have also been found [Tayoshi et al.,
2010]. An approach to exploring these mixed findings may be to link MRS and
neurophysiology. For example, Chen et al. [Chen et al., 2014] recently assessed WM
performance, baseline GABA levels (using MRS) in the left DLPFC and gamma oscillations
at baseline and during a WM task in SCZ and healthy subjects. They showed that, as a whole
(patients and healthy subjects combined), gamma amplitudes during both rest and a working
memory task were positively correlated with left DLPFC GABA levels. Despite gamma band
amplitude deficits in patients across working memory stages, both baseline and working
memory-induced gamma oscillations showed strong dependence on baseline GABA levels.
These findings suggest a critical role for GABA function in gamma band oscillations, even
under conditions of impairment. In summary, more studies are needed to confirm reduced
GABA levels in patients with SCZ via MRS as the power of this technique has yet to be fully
explored.
3.4.4 The N100 Response Evoked by TMS A prominent long-latency negative peak has been commonly observed when TMS is
delivered over the motor cortex in many studies, the N100 is the most pronounced and
reproducible TMS-evoked potential (TEP) component and has been related to the slow IPSP
[Bonato et al., 2006; Bonnard et al., 2009; Ferreri et al., 2011; Kicic et al., 2008; Komssi et
al., 2004; Lioumis et al., 2009; Nikulin et al., 2003; Paus et al., 2001; Rogasch et al., 2013].
This reproducible large negative peak occurs at about 100 ms after the TMS pulse is named
the N100 response [Yamanaka et al., 2013]. Previous studies have suggested that the N100 of
TEPs in the motor cortex may be associated with cortical inhibitory processes and very
sensitive to cortical excitability. Farzan et al., [Farzan et al., 2013] found that the amplitude
46
of the N100 component is strongly associated with the CSP duration, demonstrating that the
N100 component evoked by TMS-EEG may be related to inhibitory mechanisms. The N100
component may reflect GABAB inhibitory mechanisms due to the time course and its
relationship with CSP. Two studies have also shown that the N100 amplitude measured by
TEPs has been enhanced after administration of Baclofen (a GABAB receptor agonist),
suggesting that the N100 reflects activity of the GABAB receptor [Premoli et al., 2014a;
Premoli et al., 2014b]. More work needs to be done with the N100 response and SCZ due to
the similar GABAergic inhibitory properties comparable to LICI.
3.4.5 Main Findings The current use of methodologies available implicate a wide range of functions and brain
regions associated with GABAergic abnormalities in SCZ, providing a platform for future
work and theoretical models to build biomarkers. More enhanced research is needed,
however, combinations of existing methodologies, such as pharmacologic challenges with
neuroimaging and neurophysiology studies, have the potential to yield new information that
may refine our understanding of treatment targets and, ultimately, benefit those patients
suffering from this devastating disorder
3.5 Endophenotypes in Schizophrenia Endophenotypes are laboratory measures that are heritable quantitative traits found in
patients with a disease, and in their unaffected first-degree relatives. Endophenotypes may be
useful because they are presumably upstream in the pathophysiology of the illness and closer
to the genetic variation which underlies complex disorders such as SCZ [Braff et al., 2008].
Endophenotypes are useful for genetic studies in identifying a potentially more homogenous
subgroup that shares a common genetic etiology for both the endophenotype and the disease.
Recently, specific guidelines have been published which have identified criteria that must be
met before a biological marker can be considered an endophenotype of a neuropsychiatric
disorder [Braff et al., 2008]. Endophenotype criteria include: (1) heritability (i.e., the
proportion of genotypic variance that contributes to endophenotypic variance); (2) trait
stability (e.g., an endophenotype that is unrelated to illness duration or pharmacological
47
treatments), this implies relatively insensitive to modest fluctuations in clinical symptoms;
(3) test-retest reliability, suitable for use as repeated measures; (4) diagnostic specificity (i.e.,
an endophenotype is present in the disease of interest but not present in other disorders); and
(5) moderate to large effect size differences between patients and controls [Gottesman and
Gould, 2003]. This last condition suggests that the measure is associated with the illness and
exhibits significant deficits in patients. Future research studies need to identify measures that
demonstrate sensitive, robust, reliable, state-independent, heritable and specific biomarkers
to predict disease, track disease progression and monitor treatment. The following sections
will discuss the heritability of SCZ; the challenges involved in identifying an adequate
endophenotype for this disorder and provide evidence to suggest that frontal inhibitory
deficits may represent a candidate endophenotype.
3.5.1 Heritability of Schizophrenia Heritability is the extent to which a trait is genetically determined. Heritability is calculated
by measuring how much of the variance in a particular trait is accounted for by genetic
variance. The heritability of SCZ has been estimated to be as high as 80 percent [Cardno et
al., 1999a; Cardno et al., 1999b]. However, it is now clear that the genetics of SCZ are
complex: with many susceptibility genes and epigenetic, epistatic, stochastic and non-genetic
(i.e., environmental) influences. In this context, endophenotypes may help to group subjects
into genetic and physiological subtypes to increase the power of genetic association and
linkage analyses, as well as focus investigations into specific neurobiological pathways.
3.5.2 Applications of Endophenotypes Several lines of evidence have demonstrated that SCZ involves neurophysiological, cognitive
and genetic abnormalities. The Consortium of Genetics in Schizophrenia (COGS), a large
multi-site, and government sponsored-collaboration has investigated several
neurophysiological and cognitive markers as potential endophenotypes [Calkins et al., 2007].
Their strategy has been to acquire neurophysiological measures such as P50 suppression,
antisaccade task for eye movements and prepulse inhibition and apply the main
endophenotype criteria to each tool. Cognitive markers evaluated include the Continuous
48
Performance Test as a test of attention, the California Verbal Learning Test, as a test of
verbal declarative memory, and the Letter-Number Sequencing test as a test of WM. Results
to date suggest modest rates of heritability for the above mentioned neurophysiological and
cognitive tests [Greenwood et al., 2007]. For example, the antisaccade task for eye
movements showed a relatively moderate neurophysiological heritability of 0.42 in SCZ. The
authors suggest that factors which may account for such modest heritability include poor trait
stability (e.g., change in relation to antipsychotic treatments), low test-retest reliability and
poor diagnostic specificity [Greenwood et al., 2007]. Consequently, through this work, it has
been shown that identification of better endophenotypes for SCZ is needed, which will be
outlined in this section.
3.5.3 P50 Suppression Accumulating evidence suggests that patients with SCZ have impaired ability to filter
extraneous sensory information, predisposing them to misperceiving environmental stimuli.
These neurophysiological deficits in sensory gating can be formally evaluated through event-
related potential paradigms and indexed as P50 suppression. In P50 suppression an auditory
click generates an evoked potential within 40-80 ms of the click that is attenuated in healthy
subjects by another click which precedes it by 500 ms. Patients with SCZ consistently
demonstrate deficits in P50 suppression compared to healthy subjects. Freedman et al.
[Freedman et al., 2000] have extensively investigated the biological basis for these P50
suppression abnormalities and have concluded that P50 suppression occurs through
activation of GABAB inhibitory interneurons which, in turn, attenuates pyramidal neuron
firing. Furthermore, it has been demonstrated that P50 suppression deficits in SCZ are not
normalized with the addition of selective dopamine D2 antagonists (e.g., haloperidol) but
potentiated with clozapine [Olincy and Martin, 2005]. P50 suppression is currently being
investigated as an endophenotype in SCZ. Results to date demonstrate that P50 suppression
demonstrates low heritability of 0.10 that was not significant in a sample of 183 nuclear
families [Greenwood et al., 2007]. This may be due to the low test-retest reliability of P50
suppression (e.g., intraclass correlation coefficient <0.5)[Boutros et al., 1991], lack of
diagnostic specificity (i.e., P50 suppression abnormalities demonstrated in BD [Olincy and
49
Martin, 2005] and Alzheimer’s disease [Thomas et al., 2010]) and lack of trait stability (i.e.,
P50 suppression is altered with medications).
3.5.4 Anti Saccade Paradigm The antisaccade task requires the participant to make a saccade to an unmarked location
opposite to a flashed stimulus. Interest for this paradigm surged after the discovery that
frontal lobe lesions specifically and severely affect human performance of antisaccades while
prosaccades (i.e., saccades directed to the visual stimulus) are facilitated [Amador et al.,
1998; Campanella and Guerit, 2009]. For example, the antisaccade task for eye movements
showed a moderate to strong neurophysiological heritability of 0.42 in SCZ. The authors
suggest that factors which may account for the moderate heritability include poor trait
stability (e.g., change in relation to antipsychotic treatments), low test-retest reliability and
poor diagnostic specificity [Greenwood et al., 2007].
3.5.5 Prepulse Inhibition Gating deficits, as assessed by prepulse inhibition (PPI) of the acoustic startle has been
shown to be a candidate for an endophenotype in SCZ, reflecting the brains’ ability to filter
or gate sensory information [Braff et al., 1992]. In PPI, a nonstartling stimulus (prepulse
tone) is presented shortly before a startling stimulus [Campanella and Guerit, 2009; Graham,
1975]. When the interval between the prepulse tone and the startle stimulus is 250 ms, the
magnitude of the startle eye blink response is reduced compared with the one evoked in
response to the startle stimulus alone. However, if the interval is longer (e.g. 2000 ms), the
startle eye blink reflex is enhanced, this ‘‘prepulse facilitation’’ reflects a combination of
arousal and sustained attention elicited by the prepulse. Many published studies to date have
demonstrated PPI deficits in SCZ [Swerdlow et al., 2014]. PPI deficits have also been
demonstrated in their relatives [Braff et al., 2001; Kumari et al., 2005]. However, these
findings are not consistent as recently Ivleva et al. [Ivleva et al., 2014] found no differences
between SCZ patients, BD patients, healthy controls and their first degree relatives similar to
other studies which showed no deficits in SCZ compared to healthy controls [Ford et al.,
1999]. The inconsistency in results and modest heritability of PPI demonstrated between
50
0.32-0.45 [Hasenkamp et al., 2010] [Greenwood et al., 2007] in SCZ may deem it as an
undesirable endophenotype candidate.
3.5.6 Auditory Event-Related Potentials In this paradigm identical auditory “clicks” are presented in close succession (500 ms), with
stimulus pairs separated by long intervals (6 to 10 s) [Hamm et al., 2013; Hamm et al., 2012].
Evoked brain responses to the first stimulus (S1) and second stimulus (S2) are measured with
simultaneous EEG recording. There is a larger difference between S1 and S2 responses for
healthy subjects than for those with SCZ [Brockhaus-Dumke et al., 2008], caused by either
larger ERPs to S2 [Sanchez-Morla et al., 2009] and/or an attenuated response to S1
[Blumenfeld and Clementz, 2001] among those with SCZ. Ivleva et al., found [Ivleva et al.,
2014] an attenuated response to S1 in patients with SCZ, BD and their first-degree relatives
with a significantly lower magnitude in the theta, alpha and beta frequency bands. The lack
of diagnostic specificity in the results does not fit the criteria of a desirable endophenotype.
Furthermore, oscillatory abnormalities have been identified among those with SCZ in the
gamma band [Clementz et al., 1997; Johannesen et al., 2005], low frequency oscillations to
S1 [Brockhaus-Dumke et al., 2008; Clementz and Blumenfeld, 2001; Johannesen et al.,
2005] and late (200 to 300 ms) beta band oscillations to S1 [Brenner et al., 2009; Hong et al.,
2004].
Growing evidence has shown that abnormalities of high-frequency oscillations in the
gamma-range (30 to 100 Hz) via EEG are heritable, as demonstrated by assessing unaffected
first-degree relatives of SCZ patients. For example, Hall et al. [Hall et al., 2011] examined
the early auditory gamma-band response in SCZ and their unaffected co-twins during an
auditory oddball target detection task in 194 individuals. They found that both evoked power
and phase-locking phenotypes were reduced in unaffected co-twins of patients with SCZ and
both were shown to be heritable traits. The heritability estimates were high and found to be
0.65 for evoked power and 0.63 for phase-locking. As a follow-up to this work, Leicht et al.,
[Leicht et al., 2010] investigated the early auditory gamma band response in first-degree
relatives of SCZ patients. Both patients and unaffected siblings showed a significant
51
reduction of evoked power and phase locking in the early auditory gamma band response
compared to healthy subjects. These findings suggest that the gamma band response may also
have a heritable component in SCZ.
3.5.7 Auditory-Related N100 Findings The N100 is also defined as a response arising from the supratemporal auditory cortex
approximately 100 ms after presentation of an auditory stimulus [Hari et al., 1982; Hari et al.,
1987]. Reductions in the auditory N100 amplitude have been consistently reported in SCZ
and may reflect specific elements of the pathophysiology of SCZ [Rosburg et al., 2008].
Although this has been an observation repeatedly shown in SCZ, the N100 amplitude has
received little attention as a potential endophenotype. Evidence suggests that reduced N100
amplitude is a stable deficit found in both recent-onset [Sumich et al., 2006] and medication-
free patients [Ogura et al., 1991]. However, heritability of the N100 in unaffected first-degree
relatives is understudied. Four studies failed to find any significant differences in N100
amplitude between first degree-relatives of SCZ patients and healthy subjects [Frangou et al.,
1997; Karoumi et al., 2000; Waldo et al., 1988; Winterer et al., 2001]. Based on a twin study,
the N100 appears to be moderately heritable [Anokhin et al., 2007]. As part of the COGS,
Turetsky et al., [Turetsky et al., 2008] found reduced N100 amplitude in SCZ patients and
their unaffected first-degree relatives using EEG via an auditory paired-click paradigm. The
N100 amplitude was found to be heritable measure, as heritability estimates were for click 1
amplitude (heritability of 0.40) and click 2 amplitude (heritability of 0.29) and for the ratio
(heritability of 0.22). Taken together, the N100 is abnormal in patients and their relatives,
however, the inconsistency in these results and low heritability may deem it as an undesirable
endophenotype candidate. It is important to emphasize that, beyond its utility as a
quantitative endophenotype; the auditory N100 may provide a method to investigate the
neural substrates of SCZ.
3.5.8 Mismatch Negativity There is compelling evidence that sensory processing impairments contribute to the cognitive
and psychosocial dysfunction affecting the majority of SCZ patients [Braff and Light, 2004;
52
Kirihara et al., 2012; Light et al., 2006; Light et al.]}. Mismatch negativity is an event-related
potential of the negative component of a waveform obtained by subtracting event-related
responses to a frequency stimulus (standard) from those to a rare stimulus (deviant) with an
ISI of 500 to 1000 ms [Light et al.]. This response has been shown to reflect the function of
the auditory sensory memory system and to reflect a predominantly automatic or pre-
conscious process of detecting a “mismatch” between the deviant stimulus and a sensory–
memory trace [Naatanen et al., 1989]. Mismatch negativity has been shown to be reduced in
patients with SCZ, reduced in their first-degree relatives, heritable, reliability, trait-like
stability. These qualities suggest that the measure may be a potential endophenotype
candidate requiring further investigation [Light et al.; Michie et al., 2002].
3.6 The Bipolar-Schizophrenia Network on Intermediate
Phenotypes Research Multi-site studies increase the likelihood that findings will be generalizable by testing larger
and more heterogeneous samples. Another consortium study specializing in multi-sensor
recordings is the Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP)
which provides a platform for investigating markers of disease. The B-SNIP trial has focused
on a single clinical phenotype which is psychosis within the SCZ-BD spectrum. The main
aim was to characterize an intermediate endophenotype with probands and their first-degree
relatives. Similar clinical characteristics observed across SCZ and psychotic BD including
overlapping diagnoses and shared risk genes highlight the importance of evaluating
endophenotypes across these diagnoses. Within the B-SNIP, Tamminga et al. [Tamminga et
al., 2013] found that SCZ and schizoaffective probands had a lower proportion of
Caucasians, were less educated, had higher positive and negative syndrome scale scores and
scored lower on social functioning scales when compared to patients with BD. This study
also found that among relatives of the psychosis probands, only 33% - 38% were free of any
axis I or II diagnosis, suggesting a high burden of psychiatric morbidity in families with
psychosis. Data from biological relatives of probands are important tools for biomarker work.
This section will summarize the B-SNIP imaging and EEG studies that included first-degree
relatives of SCZ and BD probands.
53
3.6.1 Neuroimaging and EEG B-SNIP Studies
Studies of biological relatives of psychosis probands have found variable gray matter
alterations in the basal ganglia, parahippocampal gyrus, and prefrontal cortex in the relatives
of SCZ probands [Palaniyappan et al., 2012] and in the frontotemporal regions in relatives of
bipolar probands [Hajek et al., 2013]. The B-SNIP trial aimed to further understand and
expand upon these imaging results with a larger sample. For example, Ivleva et al. [Ivleva et
al., 2013] found gray matter volume reductions in psychosis probands and in relatives with
psychosis spectrum disorders in overlapping cortical regions in the frontotemporal, anterior
cinguate and parietal regions compared with nonpsychotic relatives. Furthermore, pairwise
comparisons reveal substantial overlapping gray matter reduction in SCZ and schizoaffective
probands in the frontal, anterior and posterior cingular, insular, temporal, parietal and
occipital cortices as well as in the basal ganglia, thalamus and cerebellum relative to healthy
controls. Probands with BD showed gray matter reduction in the frontal, anterior and
posterior cingulate, insular, temporal, and parietal cortices relative to healthy controls,
regionally overlapping with those in the SCZ and schizoaffective proband. No increases in
gray matter volume were found in any proband group relative to healthy controls. Many
overlapping brain regions showing structural gray matter reductions in both relatives and
probands of SCZ and BD, may demonstrate this measure to be an undesirable endophenotype
candidate. An additional B-SNIP imaging study evaluated the integrity of the white matter
connections and integrity by diffusion tensor imaging through fractional anisotropy
[Skudlarski et al., 2013]. They found decreases in functional anisotropy in the genu and body
of the corpus collosum in both SCZ and BD patients and no significant differences between
the two proband groups. They also showed that SCZ and BD relatives showed fractional
anisotropy deviations similar to their probands, however, most pronounced in the relatives of
SCZ. Lastly, Arnold et al. [Arnold et al., 2015] examined hippocampal volume as putative
biomarker and found that SCZ patients showed specific hippocampal volume reductions.
Relatives of SCZ patients were not significantly different than healthy controls, however, had
significantly higher volumes than probands, suggesting an intermediate deficit. The findings
from this large B-SNIP trial support further exploration of functional anisotropy in the genu
and body of the corpus collosum and hippocampal volumes in SCZ and their relatives.
54
Ethbridge et al. [Ethridge et al., 2015] recently examined an auditory oddball task, whereby
EEG responses were measured to deviant auditory (“oddball”) stimuli interleaved in a train
of standard tones assessing SCZ, BD and their first-degree relatives. Spatial principal
components analysis derived data-driven frequency waveforms for each subject. Some of the
biomarkers of familial risk (N100, P3b) were more specific to SCZ, whereas another
response abnormality (P2) was specific to BD, and another (N2) was common to both
psychoses. The novel methods presented in this work quantify neural oscillatory information
based on spatial and time-frequency components may provide new targets for investigating
the alterations unique to SCZ and BD or shared across the psychosis spectrum.
3.7 Transcranial Magnetic Stimulation Studies with First-
Degree Relatives of Schizophrenia Patients To date, limited studies have evaluated first-degree relatives of SCZ patients using TMS
paradigms. Saka et al., [Saka et al., 2005] evaluated TMS measures of inhibition in
unaffected first-degree relatives of SCZ patients compared to healthy subjects (no proband
group was assessed). They found that 25% of first-degree relatives lacked transcallosal
inhibition and showed psychosis-proneness relative to healthy controls. No differences were
found in MEP amplitude (excitability measure) or the cortical silent period (inhibitory
measure). Furthermore, Hasan et al. evaluated MEP amplitude in patients with SCZ, their
unaffected first-degree relatives and healthy controls after cathodal transcranial direct current
stimulation (tDCS). It was hypothesized that cathodal tDCS would induce long-term
depression of the MEP in healthy subjects. They found that cathodal tDCS reduced the MEP
of the stimulated hemisphere in healthy subjects as predicted, however, had no effect on first-
degree relatives and SCZ patients. Within the non-stimulated hemisphere, MEPs were
facilitated in first-degree relatives and SCZ patients. This study provides preliminary
evidence for impaired plasticity in SCZ patients and their first-degree relatives. Taken
together, more research needs to be done using TMS in biological relatives of SCZ patients.
55
3.8 Outline of the Dissertation In summary, evaluating tools using the endophenotype approach will continue to provide a
deeper insight into the neurobiological underpinnings of SCZ. Future studies investigating
brain structure and function characteristics in individuals with severe mental illness using
multimodal investigative approaches such as neuroimaging coupled with genetic, molecular
and TMS measures may help to elucidate disease-related mechanisms and biomarkers. As a
summary of the dissertation, chapters 1, 2 and 3 have provided the background for this
current project. Chapter 4 provides the objectives and hypotheses for each of the three
original research studies. Chapters 5, 6 and 7 represent studies one, two, and three,
respectively. The dissertation will conclude with chapter 8, devoted to the general discussion,
conclusions and future aims of this doctoral project.
56
Chapter 4
Research Aims and Hypotheses of the Experiments
57
4.1 Introduction In the introduction chapters we reviewed that the severity of inhibitory dysfunction has been
shown to be linked to functional outcome in SCZ [van Os and Kapur, 2009]. As previously
mentioned, the primary brain region responsible for processing of executive functions is the
DLPFC, implicated in the pathophysiology of SCZ [Chen et al., 2014]. In this regard, two
major DLPFC abnormalities have been shown in SCZ patients: deficits in GABAergic
inhibitory neurotransmission and impairments in the inhibition of cortical oscillations.
In vivo human studies examining the modulatory effect of GABAB receptors on cortical
gamma oscillations in DLPFC are lacking. LICI is a paired-pulse TMS paradigm that is
suggested to reflect the modulatory effect of GABAB receptors in the cortex. Several
additional experiments were designed to confirm the possibility that deficits in frontal LICI
may represent a candidate endophenotype for SCZ. In this regard, several questions were
raised, (1) would inhibitory deficits be a ubiquitous finding in psychiatric disorders as
measured via TMS; (2) would patients with SCZ have specific impairments of frontal
inhibition (previously demonstrated in this patient population); (3) if patients with SCZ have
inhibitory impairments, whether or not such impairments could represent a candidate
endophenotype for this illness; (4) are frontal inhibition deficits a replicable finding in SCZ;
(5) are frontal inhibitory deficits in SCZ independent of antipsychotic medication; (6) are
frontal inhibition deficits disease-specific to SCZ; (7) would inhibitory deficits be
demonstrated in unaffected first-degree relatives of SCZ.
These seven questions raised above gave rise to a series of three studies that were conducted
during the course of my PhD program. This study was designed by my supervisor Dr.
Daskalakis. Upon joining the Temerty Centre for Therapeutic Brain Intervention, I was given
the great opportunity of undertaking this large PhD project, as I contributed extensively to
recruitment of participants, study design, data collection, development of analytical methods,
data analysis, conference presentations as well as publication of manuscripts.
58
4.2 A Meta-Analysis of Cortical Inhibition and Excitability
Using Transcranial Magnetic Stimulation in Psychiatric
Disorders (Chapter 5) Objectives and Hypotheses
4.2.1 Objective 1, Inhibitory Deficits Present in Severe Psychiatric
Disorders The objective of the meta-analysis was to quantitatively assess all studies that used TMS
motor cortex measures of inhibition and excitation in OCD, MDD and SCZ. The effect size
statistic we used was a Hedge’s G to calculate mean differences in each measure per group.
The Hedge’s G analysis was used as it had the benefit to control for sample size. The main
aim of the study was to determine the exact pattern of deficit with inhibition and excitation
measures in each psychiatric disorder based on the statistically significant (p < 0.05) Hedge’s
G value.
4.2.2 Hypothesis 1, Inhibitory Deficits Present in Severe Psychiatric
Disorders We hypothesized that motor cortex GABAergic inhibitory deficits would be a ubiquitous
finding across OCD, MDD and SCZ patients. Specifically, OCD and MDD would have
similar significant profiles of inhibitory deficit in both GABAB and and GABAA receptor-
mediated inhibition. SCZ patients would show a pattern of deficit specific to GABAA
receptor-mediated inhibition.
4.3 Evidence for Inhibitory Deficits in the Prefrontal Cortex
in Schizophrenia (Chapter 6) Objectives and Hypotheses
59
4.3.1 Objective 1, Replication of Frontal Inhibitory Deficits The first objective was to evaluate GABAB inhibition via LICI in both the DLPFC and motor
cortex using TMS combined with EEG. The main aim was to conduct neurophysiological
assessments with a large sample size of SCZ, OCD and healthy controls. The main premise
was to demonstrate frontal inhibitory deficits in a large sample size of SCZ patients, based on
previous findings.
4.3.2 Hypothesis 1, Replication of Frontal Inhibitory Deficits The first hypothesis was that SCZ patients would show frontal LICI deficits compared to
healthy subjects. The core study hypothesis was that frontal LICI deficits would be
significantly greater in patients with SCZ than OCD patients. No significant differences
would be found in the motor cortex between all three groups.
4.3.3 Objective 2, Diagnostic Specificity of Frontal Inhibitory
Deficits The second objective was to evaluate the diagnostic specificity of frontal cortex LICI in SCZ
and OCD, due to the comparable levels of psychopathology severity and similar
pharmacology that are used to treat these two disorders.
4.3.4 Hypothesis 2, Diagnostic Specificity of Frontal Inhibitory
Deficits This study hypothesized that frontal inhibition deficits would be specific to patients with
SCZ, while not shown in OCD and healthy participants. The absence of such abnormalities in
OCD would confirm that frontal LICI deficits are not a generalized pattern of severe
psychopathology and may be specific to patients with SCZ. This would be demonstrated by
showing a significant negative relationship between clinical severity scores in SCZ and LICI
deficits.
60
4.3.5 Objective 3, Large Effect Size Differences via Cohen’s D The third objective was to demonstrate effect sizes differences in frontal LICI between SCZ
patients and healthy controls using the Cohen’s D measure.
4.3.6 Hypothesis 3, Large Effect Size Differences via Cohen’s D It was hypothesized that large effect size differences would be shown in frontal LICI when
comparing SCZ patients to healthy subjects using the Cohen’s D measure.
4.3.7 Objective 4, Trait Stability of Frontal Inhibition The final objective was to demonstrate that frontal inhibition deficits found in SCZ patients
were independent of antipsychotic treatment.
4.3.8 Hypothesis 4, Trait Stability of Frontal Inhibition We hypothesized that there would be no effect of antipsychotic treatment on frontal LICI in
SCZ patients. This would be demonstrated with two main approaches. First, medicated SCZ
patients would show inhibitory deficits when compared to similarly treated OCD patients (on
antipsychotic medications) who are age and sex matched. Second, this study hypothesized no
significant relationship between LICI and chlorpromazine equivalent dosages in SCZ
patients, thereby, demonstrating no effect of medication.
4.4 Investigating the Heritability of Cortical Inhibition in
First-Degree Relatives and Probands in Schizophrenia
(Chapter 7) Objectives and Hypotheses
4.4.1 Objective 1, Assessing Inhibition in Unaffected First-Degree
Relatives
61
The main objective of this study was to evaluate inhibition of the DLPFC and motor cortex
using the LICI paradigm (via TMS-EEG) in SCZ patients, OCD patients and both of their
unaffected first-degree relatives compared to healthy controls. We aimed to assess unaffected
first-degree relatives of SCZ patients as they share degrees of genetic vulnerability with the
proband; however, they are free from confounds related to medication and symptomatology.
4.4.2 Hypothesis 1, Frontal Inhibition in Unaffected First-Degree
Relatives This study hypothesized that frontal inhibition deficits would be demonstrated in SCZ and
would also show the greatest LICI impairment. Furthermore, we hypothesized that frontal
inhibition in first-degree relatives of SCZ would be intermediate of their related probands
and healthy subjects. We hypothesized a moderate heritability for frontal inhibition in SCZ.
Lastly, we hypothesized no significant frontal inhibition differences would be found in OCD
patients when compared to their unaffected first-degree relatives and healthy subjects.
62
Chapter 5
A Meta-Analysis of Cortical Inhibition and Excitability Using
Transcranial Magnetic Stimulation in Psychiatric Disorders.
Contents of this chapter have been reprinted by permission from Elsevier Ireland Ltd
Radhu N, de Jesus DR, Ravindran LN, Zanjani A, Fitzgerald PB, Daskalakis ZJ. (2013). A
Meta-Analysis of Cortical Inhibition and Excitability Using Transcranial Magnetic
Stimulation in Psychiatric Disorders. Clinical Neurophysiology. 124 (7): 1309-1320.
A link of this paper can be found at:
http://www.clinph-journal.com/article/S1388-2457(13)00056-4/abstract
63
5.1 Abstract Objective
To evaluate TMS measures of inhibition and excitation in OCD, MDD and SCZ.
Methods
Paradigms included: SICI, CSP, RMT, ICF, and MEP amplitude. A literature search was
performed using PubMed, Ovid Medline, Embase Psychiatry and PsycINFO 1990 through
April 2012. Motor cortex LICI was not included as there were no studies found.
Results
A significant Hedge's g was found for decreased SICI (g = 0.572, 95% confidence interval
[0.179, 0.966], p = 0.004), enhanced intracortical facilitation (g = 0.446, 95% confidence
interval [0.042, 0.849], p = 0.030) and decreased CSP (g = -0.466, 95% confidence interval [-
0.881,-0.052], p = 0.027) within the OCD population. For MDD, significant effect sizes were
demonstrated for decreased SICI (g = 0.641, 95% confidence interval [0.384, 0.898],
p=0.000) and shortened CSP (g = -1.232, 95% confidence interval [-1.530, -0.933], p =
0.000). In SCZ, a significant Hedge's g was shown for decreased SICI (g = 0.476, 95%
confidence interval [0.331, 0.620], p = 0.000).
Conclusions
Inhibitory deficits are a ubiquitous finding across OCD, MDD, SCZ and enhancement of
intracortical facilitation is specific to OCD.
Significance: Provides a clear platform from which diagnostic procedures can be developed.
64
5.2 Introduction GABA is the main inhibitory neurotransmitter in the brain, critical for the modulation of
cortical excitability and neuroplasticity [DeFelipe et al., 1986; Schieber and Hibbard, 1993].
GABAergic neurons constitute 25% to 30% of the neuronal population in the motor cortex
and their horizontal connections can extend up to 6 mm or more [Gilbert and Wiesel, 1992;
Jones, 1993]. Pyramidal cell activity is synchronized through a balance of inhibitory
postsynaptic potentials and excitatory postsynaptic potentials [Krnjevic, 1997]. IPSPs are
generated by GABAergic interneurons terminating on the pyramidal cell [Krnjevic, 1997].
Cortical inhibition is a neurophysiological mechanism whereby GABA inhibitory
interneurons attenuate the activity of other neurons (e.g. pyramidal neurons) in the cortex
[Daskalakis et al., 2007].
TMS is a non-invasive method used to assess inhibitory and excitatory mechanisms. TMS
was first introduced in 1985 by Barker et al. for investigating the state of motor pathways in
patients with neurological disorders and in healthy participants [Barker et al., 1985]. They
showed that a single TMS pulse applied to the motor cortex could activate cortical tissues
associated with the hand or leg muscles and elicit motor evoked potentials.
5.3 Inhibitory TMS Paradigms TMS has been used to assess inhibitory processes, these paradigms are referred to as the CSP
[Cantello et al., 1992], LICI [Valls-Sole et al., 1992], and SICI [Kujirai et al., 1993]. The
CSP duration is measured from the motor evoked potential onset to the return of
electromyography activity [Cantello et al., 1992]. LICI involves the pairing of a
suprathreshold CS followed by a suprathreshold TS at long ISIs, resulting in inhibition of the
motor evoked potential [Valls-Sole et al., 1992]. CSP and LICI appear to be assessing
GABAB receptor-mediated inhibitory neurotransmission as evidenced by pharmacological
studies [McDonnell et al., 2006; Siebner et al., 1998], the time course of the GABAB
inhibitory postsynaptic potential [McCormick, 1989; Siebner et al., 1998; Werhahn et al.,
1999b] and the high intensity suprathreshold CS [Sanger et al., 2001]. By contrast, SICI is
measured by applying a subthreshold CS before the suprathreshold TS at short ISIs, resulting
65
in inhibition of the motor evoked potential response by 50% to 90% [Kujirai et al., 1993].
SICI has been associated with the GABAA receptor-mediated inhibitory neurotransmission as
demonstrated by the pharmacological effects on this measure [Ziemann et al., 1996a], the
time course of the GABAA inhibitory postsynaptic potential [Wang and Buzsaki, 1996] and
the low intensity subthreshold CS [Sanger et al., 2001].
5.3.1 Excitatory TMS Paradigms TMS has also been used to examine cortical excitability, these paradigms include: MEP
amplitude, RMT, and ICF. The MEP amplitude is measured as the average response to a
series of pulses applied at a consistent TMS intensity [Zaaroor et al., 2003]. The RMTis
defined as the minimal intensity that produces a motor evoked potential > 50 μV in 5 of 10
trials in a relaxed muscle [Rossini et al., 1994]. Finally, ICF is a paired-pulse paradigm
whereby a CS is applied to the motor cortex before the TS, resulting in an enhanced motor
evoked potential [Kujirai et al., 1993; Nakamura et al., 1997]. ICF originates from excitatory
postsynaptic potentials transmitted by N-methyl-D-aspartate glutamate receptors [Nakamura
et al., 1997]. For a review of the pharmacological effects on inhibitory and excitatory TMS
paradigms, please see [Paulus et al., 2008b].
5.4 Applications within Psychiatric Disorders Numerous studies have implicated GABA in the pathophysiology of neuropsychiatric
disorders, notably OCD, MDD, SCZ, and BD. Several lines of evidence suggest that cortical
inhibition is impaired in these disorders. For example, previous TMS studies have
demonstrated deficits in cortical inhibition assessed from the motor cortex in patients with
OCD [Greenberg et al., 2000; Greenberg et al., 1998; Richter et al., 2012], MDD [Bajbouj et
al., 2006b; Fitzgerald et al., 2004a; Lefaucheur et al., 2008; Levinson et al., 2010], SCZ
[Daskalakis et al., 2002a; Daskalakis et al., 2008b; Fitzgerald et al., 2002a; Fitzgerald et al.,
2002b; Fitzgerald et al., 2003; Liu et al., 2009; Wobrock et al., 2010; Wobrock et al., 2009;
Wobrock et al., 2008] and bipolar disorder [Levinson et al., 2007]. An overall deficit of
GABAergic inhibition has been associated with these psychiatric disorders; however, each
may have a distinct illness profile and response to treatment. This meta-analysis aims to
66
quantitatively assess TMS evoked measures of inhibitory and excitatory paradigms in OCD,
MDD and SCZ.
5.5 Methods 5.5.1 Data Sources A literature search was performed using PubMed, Ovid Medline, Embase Psychiatry and
PsycINFO 1990 through April 2012.
A description of the exact search terms used:
motor cortex tms and psychiatry, motor cortex tms and mental disorder, motor cortex tms
and psychiatric disorder, motor cortex tms and anxiety disorder, motor cortex tms and
bipolar disorder, motor cortex tms and mania, motor cortex tms and depression, motor
cortex tms and obsessive-compulsive disorder, motor cortex tms and posttraumatic stress
disorder, motor cortex tms and schizophrenia, motor cortex tms and major depressive
disorder, short-interval cortical inhibition and schizophrenia, short-interval cortical
inhibition and depression, short-interval cortical inhibition and ocd, intracortical faciliation
and schizophrenia, intracortical facilitation and depression, intracortical facilitation and
ocd, cortical silent period and schizophrenia, cortical silent period and depression, cortical
silent period and ocd, resting motor threshold and schizophrenia, resting motor threshold
and depression, resting motor threshold and ocd, motor evoked potential amplitude and
schizophrenia, motor evoked potential amplitude and depression, motor evoked potential
amplitude and ocd.
5.5.2 Study Selection Studies were included if the following criteria were fulfilled:
1. Cortical inhibition or cortical excitability motor cortex measurements were assessed using
TMS.
2. Psychiatric disorders were diagnosed in accordance with DSM criteria.
3. The study had no specific “narrow” diagnosis or subgroup, such as depression after stroke
or vascular depression.
4. The study included a healthy unaffected comparison group.
67
5. The data were sufficient to compute Hedges’ G (sample size, means, and standard
deviations).
6. At least 2 studies per psychiatric disorder/symptom cluster.
7. More than 3 participants per study.
8. Articles written in English.
9. In the case of articles with overlapping samples, the article with the largest sample size
was included.
5.5.3 Data Extraction The following data were acquired: number patients, number of healthy controls, mean and
standard deviation of the outcome measure at baseline. When publications contained
insufficient or incomplete data, the authors in question were contacted and invited to send
additional data so that their study could be included in the meta-analysis.
5.5.4 Hedge's g Calculation for the Meta-Analysis We employed standardized meta-analytic techniques used in the literature. A Hedge's g, 95%
confidence interval and p-value were calculated (patients versus healthy controls) for each
psychiatric disorder for measures of cortical inhibition (SICI, CSP) or excitability (resting
motor threshold, intracortical facilitation) and the motor evoked potential amplitude for
MDD and SCZ. This was analyzed using Comprehensive Meta Analysis Version 2.0
(Biostat, Englewood, New Jersey) in a fixed effects model. The means and standard
deviations of separate studies were weighted according to sample size.
5.5.5 Test of Heterogeneity We evaluated heterogeneity among studies by calculating a Cochran Q, p-value and I2.
Heterogeneity in a meta-analysis refers to the variation in study outcomes between studies
[Higgins and Thompson, 2002]. The Q statistic is a value that demonstrates how the
independent studies varied in terms of their findings. The I² statistic is a percentage of
variation across studies that is due to heterogeneity rather than chance [Higgins and
Thompson, 2002; Higgins et al., 2003]. The I2 ranges from 0% to 100%, a value of 0%
68
means no heterogeneity and 100% means a high level of heterogeneity. A meta-regression
was implemented to control for variables such as age and medication status; this allowed for
the comparison of multiple sources of heterogeneity. Three or more studies were needed for
each variable to complete a meta-regression.
5.5.6 N Fail-Safe To examine publication bias, an N fail-safe value was calculated. This value is defined as the
number of non-significant unpublished studies needed to make the obtained effect size
calculations non-significant. Three or more studies were needed to complete this analysis.
We adopted a significance level of p = 0.05, 2-tailed for all of the analyses.
5.6 Results Table 3 provides the total number of studies that fulfilled the 9 stated criteria for inclusion
(described in the methods) and the total number of studies excluded based upon specified
reasons. The search was completed by N.R. and the studies were checked for reliability by
D.R.J. Studies met the checklist for assessing the methodological quality of studies using
TMS [Chipchase et al., 2012].
Psychiatric Disorder Number of Studies Included in
Meta-Analysis
Reasons for Exclusion and
Number of Studies Excluded
OCD Resting Motor Threshold (2)
Short Interval Cortical Inhibition (3)
Intracortical Facilitation (2)
Cortical Silent Period (2)
Motor Evoked Potential Amplitude
(0)
Insufficient Data (1)
No Healthy Comparison Group
(1)
MDD Resting Motor Threshold (8)
Short Interval Cortical Inhibition (3)
Intracortical Facilitation (3)
Cortical Silent Period (4)
Motor Evoked Potential Amplitude
(3)
Insufficient Data (2)
No Healthy Comparison Group
(9)
Epileptic Patients with Major
Depression (1)
69
Table 3. Number of Included and Excluded Studies.
5.6.1 Patients with OCD
5.6.2 OCD - Resting Motor Threshold Figure 4A illustrates the summary of the Hedge's g analysis as a forest plot based on 2
studies [Greenberg et al., 2000; Richter et al., 2012] that met inclusion criteria. The analysis
comprised a total of 50 patients with OCD compared to 45 healthy controls. No significant
differences were found in resting motor threshold in OCD. The Hedge's g was g = -0.251,
95% confidence interval [-0.658, 0.156], p = 0.227. The test of heterogeneity was found to be
significant (Q= 7.822, df(q) = 1, p = 0.005, I2 = 87.216). Meta-regression and publication bias
analyses were not possible due to the fact that only 2 published studies were available.
5.6.3 OCD - SICI Figure 4B displays the summary of the Hedge's g analysis as a forest plot based on 3 studies
[Greenberg et al., 2000; Greenberg et al., 1998; Richter et al., 2012] that met inclusion
criteria. This analysis consisted of 62 OCD patients compared to 57 healthy controls. SICI
was significantly reduced in OCD. The Hedge's g was found to be g = 0.572, 95% confidence
interval [0.179,0.966], p = 0.004. The test of heterogeneity was found to be significant (Q =
SCZ Resting Motor Threshold (21)
Short Interval Cortical Inhibition
(12)
Intracortical Facilitation (11)
Cortical Silent Period (11)
Motor Evoked Potential Amplitude
(4)
Insufficient Data (3)
No Healthy Comparison Group
(2)
Not in English (1)
BD 0 Insufficient Data (2)
Post-Traumatic Stress
Disorder
0 Less than 2 studies for this
disorder (1)
Social Anxiety
Disorder
0 Less than 2 studies for this
disorder (1)
70
36.366, df(q) = 2, p = 0.000, I2 = 94.5). The n-failsafe value was found to be 3 unpublished
studies. A meta-regression was not possible due to 2 studies publishing the values for age.
5.6.4 OCD - Intracortical Facilitation Figure 4C illustrates the summary of the Hedge's g analysis as a forest plot based on 2 studies
[Greenberg et al., 2000; Richter et al., 2012] that fit the inclusion criteria. The analysis
included 50 patients with OCD compared to 45 healthy controls. Intracortical facilitation was
significantly enhanced in OCD. The Hedge's g was found to be g = 0.446, 95% confidence
interval [0.042, 0.849], p = 0.030. The test of heterogeneity was not significant (Q = 1.162,
df(q) = 1, p = 0.281, I2 = 13.912). A meta-regression and publication bias analyses were not
possible due to only 2 published studies available.
5.6.5 OCD - CSP Figure 4D illustrates the summary of the Hedge's g analysis as a forest plot based on 2
studies [Greenberg et al., 2000; Richter et al., 2012] that fit the inclusion criteria. This
analysis contained 50 patients with OCD compared to 45 healthy controls. CSP was
significantly reduced in OCD. The Hedge's g was found to be g = -0.466, 95% confidence
interval [-0.881,-0.052], p = 0.027. The test of heterogeneity was significant (Q = 10.435,
df(q) = 1, p = 0.001, I2 = 90.417). A meta-regression and publication bias analyses were not
possible due to only 2 published studies available.
71
72
Figure 4. Forest Plot of the Hedge's g Analysis For All Studies That Included Patients With
Obsessive-Compulsive Disorder Compared to Healthy Controls.
(A) Resting Motor Threshold
(B) Short-interval Cortical Inhibition
(C) Intracortical Facilitation
(D) Cortical Silent Period
5.7 Patients with MDD 5.7.1 MDD - Resting Motor Threshold Figure 5A illustrates the summary of the Hedge's g analysis as a forest plot based on 8
studies [Abarbanel et al., 1996; Bajbouj et al., 2006b; Chroni et al., 2002; Grunhaus et al.,
2003; Lefaucheur et al., 2008; Levinson et al., 2010; Maeda et al., 2000; Reid et al., 2002]
that fit the inclusion criteria. This analysis comprised of 176 patients with MDD compared
to 188 healthy controls. No significant differences were found in resting motor threshold in
MDD. The Hedge's g was g = -0.043, 95% confidence interval [-0.248, 0.161], p = 0.677.
The test of heterogeneity was not significant (Q =16.034, df(q) = 9, p = 0.066, I2 = 43.87).
The n-failsafe value was found to be 10 unpublished studies. Controlling for age, the meta
regression yielded a correlation of r = 0.04891 and p = 0.01399. Controlling for medications,
the meta regression yielded a correlation of r = 0.40970, p = 0.08217.
5.7.2 MDD - SICI Figure 5B illustrates the summary of the Hedge's g analysis as a forest plot based on 3 studies
[Bajbouj et al., 2006b; Lefaucheur et al., 2008; Levinson et al., 2010] that fit the inclusion
criteria. The analysis included 115 patients with MDD compared to 130 healthy controls.
SICI was significantly reduced in MDD. The Hedge's g was found to be g = 0.641, 95%
confidence interval [0.384, 0.898], p = 0.000. The test of heterogeneity was significant (Q =
10.362, df(q) = 4 , p = 0.035, I2 =61.398) and the n-failsafe value was found to be 5
unpublished studies. Controlling for age, the meta regression yielded a correlation of r =
0.03221 and p = 0.23094. Controlling for medications, the meta regression was found to be r
= 0.21356, p = 0.44282.
73
5.7.3 MDD - Intracortical Facilitation Figure 5C illustrates the summary of the Hedge's g analysis as a forest plot based on 3 studies
that fit the inclusion criteria [Bajbouj et al., 2006b; Lefaucheur et al., 2008; Levinson et al.,
2010]. The analysis consisted of 115 patients with MDD compared to 130 healthy controls.
No significant differences were found in intracortical facilitation in MDD. The Hedge's g
was g = -0.062, 95% confidence interval [-0.311, 0.188], p = 0.628. The test of heterogeneity
was not significant (Q = 7.465, df(q) = 4, p = 0.113, I2 = 46.413). The n-failsafe value was 5
unpublished studies. Controlling for age, the meta regression yielded a correlation of r = -
0.06835 and p = 0.00855. Controlling for medications, the meta regression was found to be r
= -0.16181 , p = 0.54986 .
5.7.4 MDD - CSP Figure 5D illustrates the summary of the Hedge's g analysis as a forest plot based on 4
studies that fit the inclusion criteria [Bajbouj et al., 2006b; Lefaucheur et al., 2008; Levinson
et al., 2010; Steele et al., 2000]. The analysis comprised of 131 patients with MDD compared
to 149 healthy controls. CSP was significantly reduced in MDD. The Hedge's g was found to
be g = -1.232, 95% confidence interval [-1.530, -0.933], p = 0.000. The test of heterogeneity
was significant (Q = 158.857, df(q) = 5, p = 0.000, I2 = 96.853). The n-failsafe value was 6
unpublished studies. Controlling for age, the meta regression correlation was r = 0.01035 and
p = 0.68408. Controlling for medications, the meta regression was found to be r = 0.69466, p
= 0.04121.
5.7.5 MDD - Motor Evoked Potential Amplitude Three studies [Chroni et al., 2002; Reid et al., 2002; Shajahan et al., 1999] that fit the
inclusion criteria yielded a Hedge's g of g = 0.162, 95% confidence interval [-0.300, 0.623], p
= 0.492. No significant differences were found in the motor evoked potential amplitude in
MDD. The test of heterogeneity was significant (Q = 6.586, df(q) = 2, p = 0.037, I2 =
69.633). This analysis included 34 patients with MDD compared to 37 healthy controls. The
n-failsafe value was 3 unpublished studies. Controlling for age, the meta regression yielded a
74
correlation of r = 0.09093 and p = 0.21375. All studies included medicated patients and a
meta-regression for medication was not possible.
75
Figure 5. Forest Plot of the Hedge's g Analysis For All studies That Included Patients With
Major Depressive Disorder Compared to Healthy Controls.
(A) Resting Motor Threshold
(B) Short-interval Cortical Inhibition
(C) Intracortical Facilitation
(D) Cortical Silent Period
5.8 Patients with SCZ 5.8.1 SCZ - Resting Motor Threshold Figure 6 displays the Hedge's g as a forest plot based on 21 studies [Abarbanel et al., 1996;
Bajbouj et al., 2004; Boroojerdi et al., 1999; Chroni et al., 2002; Daskalakis et al., 2002a;
Daskalakis et al., 2008b; Eichhammer et al., 2004; Fitzgerald et al., 2002a; Fitzgerald et al.,
2002b; c; Fitzgerald et al., 2004b; Fitzgerald et al., 2003; Herbsman et al., 2009; Hoy et al.,
2007; Liu et al., 2009; Oxley et al., 2004; Pascual-Leone et al., 2002; Reid et al., 2002;
Soubasi et al., 2010; Wobrock et al., 2009; Wobrock et al., 2008] that met inclusion criteria.
No significant differences were found in resting motor threshold in SCZ. The Hedges G was
g = 0.067, 95% confidence interval [-0.053, 0.186], p = 0.274. The test of heterogeneity was
significant (Q = 83.977, df(q) = 30, p = 0.000, I2 = 64.276). This analysis included 500 SCZ
patients and 617 healthy controls. The n-failsafe value was 31 unpublished studies. After
controlling for age, the meta regression yielded a correlation of r = 0.02696, p = 0.05239.
After controlling for medications, the meta regression demonstrated a correlation of r =
0.20309 and p = 0.19117.
76
Figure 6. Forest Plot of Resting Motor Threshold Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls.
5.8.2 SCZ - SICI Figure 7 displays the Hedge's g as a forest plot based on 12 studies [Daskalakis et al., 2002a;
Daskalakis et al., 2008b; Eichhammer et al., 2004; Fitzgerald et al., 2002b; c; Fitzgerald et
al., 2004b; Hasan et al., 2012; Liu et al., 2009; Oxley et al., 2004; Pascual-Leone et al., 2002;
Wobrock et al., 2009; Wobrock et al., 2008] that met inclusion criteria. SICI was
significantly reduced in SCZ. The Hedge's g was found to be g = 0.476, 95% confidence
77
interval [0.331, 0.620], p = 0.000. The test of heterogeneity was not significant (Q = 19.170,
df(q) = 19, p = 0.446, I2 = 0.887). The analysis included 335 SCZ compared to 440 healthy
controls. The n-failsafe was 20 unpublished studies. After controlling for age, the meta
regression was found to be r = 0.01029, p = 0.56518 . After controlling for medications, the
meta regression demonstrated a correlation of r = -0.08425, p = 0.6429.
Figure 7. Forest Plot of Short-Interval Cortical Inhibition Hedge's g Analysis For All Studies
That Included Patients With Schizophrenia Compared to Healthy Controls.
78
5.8.3 SCZ - Intracortical Facilitation Figure 8 displays the Hedge's g as a forest plot based on 11 studies [Daskalakis et al., 2002a;
Daskalakis et al., 2008b; Eichhammer et al., 2004; Fitzgerald et al., 2002b; c; Fitzgerald et
al., 2004b; Hasan et al., 2012; Liu et al., 2009; Pascual-Leone et al., 2002; Wobrock et al.,
2009; Wobrock et al., 2008] that met inclusion criteria. No significant differences were found
in intracortical facilitation in SCZ. The Hedge's g was g = 0.015 , 95% confidence interval [-
0.130,0.160], p = 0.841. The test of heterogeneity was not significant (Q = 17.236, df(q) =
18, p = 0.507, I2 = 0). The analysis incorporated 323 patients with SCZ compared to 428
healthy controls. The n-failsafe value was 19 unpublished studies. After controlling for age,
the meta regression correlation was r = 0.00200, p = 0.91120. After controlling for
medications, the meta regression was found to be r = -0.11468, p = 0.52264.
79
Figure 8. Forest Plot of Intracortical Facilitation Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls.
5.8.4 SCZ - CSP Eleven studies yielded a Hedge's g of g = -0.093, 95% confidence interval [-0.241, 0.055], p
= 0.218 [Bajbouj et al., 2004; Daskalakis et al., 2002a; Daskalakis et al., 2008b; Fitzgerald et
al., 2002b; c; Fitzgerald et al., 2004b; Hasan et al., 2012; Herbsman et al., 2009; Liu et al.,
2009; Soubasi et al., 2010; Wobrock et al., 2009] (figure 9). No significant differences were
found in CSP in SCZ. The test of heterogeneity was significant (Q = 161.499, df(q) = 18, p =
0.000, I2 = 88.854). The analysis consisted of 334 SCZ patients compared to 457 healthy
controls. The n-failsafe was 19 unpublished studies. After controlling for age, the meta
regression yielded a correlation of r = 0.01088, p = 0.58550. After controlling for
medications, the meta regression was found to be r = 0.53667, p = 0.00855.
80
Figure 9. Forest Plot of Cortical Silent Period Hedge's g Analysis For All Studies That
Included Patients With Schizophrenia Compared to Healthy Controls.
5.8.5 SCZ - Motor Evoked Potential Amplitude Four studies yielded a Hedge's g of g = -0.102, 95% confidence interval [-0.391, 0.187], p =
0.489 [Chroni et al., 2002; Enticott et al., 2008; Reid et al., 2002; Soubasi et al., 2010]. No
significant differences were found in the motor evoked potential amplitude in SCZ. The test
of heterogeneity was significant (Q = 12.134, df(q) = 3, p = 0.007, I2 = 75.276). The analysis
included 91 SCZ patients compared to 93 healthy controls. The n-failsafe was 4 unpublished
studies. After controlling for age, the meta regression yielded a correlation of r = -0.07118, p
= 0.04532. It was not possible to conduct an analysis to control for medication status (meta-
regression) as all patients were medicated.
5.9 Discussion To our knowledge, this is the first study to provide a quantitative summary of TMS studies
evaluating inhibition and excitatory paradigms in severe psychiatric disorders. The literature
included ample high-quality studies with effect sizes in the low to moderate and moderate to
high range. We found decreased SICI, enhanced intracortical facilitation and reduced CSP
within the OCD population. For MDD, decreases in CSP and SICI were demonstrated.
Lastly, reductions in SICI were shown in SCZ. These findings suggest that impairments in
GABAergic inhibition are a ubiquitous finding in severe psychiatric illnesses.
The greatest significant effect size was found in patients with OCD for decreased SICI.
Furthermore, enhanced intracortical facilitation and shortened CSP were also significant.
This finding held strong in spite of the small number of studies. This is in line with the
literature which has shown decreased SICI [Greenberg et al., 2000; Greenberg et al., 1998],
shortened CSP [Richter et al., 2012] and enhanced intracortical facilitation [Richter et al.,
2012], independent of medication status [Richter et al., 2012]. OCD may be associated with a
dysregulation of both GABAA and GABAB receptor-mediated inhibitory neurotransmission
81
and N-methyl-D-aspartate receptor-mediated excitatory neurotransmission, consistent with
genetic findings [Arnold et al., 2006; Dickel et al., 2006; Samuels et al., 2011; Stewart et al.,
2007; Voyiaziakis et al., 2011; Zai et al., 2005]. Compared to MDD and SCZ, these results
provide further evidence to demonstrate that inhibitory deficits in combination with enhanced
intracortical facilitation may be specific to OCD.
The greatest significant effect size found in patients with MDD was for shortened CSP. Also,
SICI was significantly reduced in patients with MDD. These findings show that a decrease in
SICI and shortened CSP may be unique to MDD. For example, Levinson et al. [Levinson et
al., 2010] demonstrated that all patients with MDD, regardless of symptom or medication
state, demonstrated significant CSP deficits compared with healthy participants. By contrast,
only treatment resistant MDD patients demonstrated SICI deficits. Taken together, this data
suggests that MDD is associated with deficits in neurophysiological indexes of GABAB
receptor-mediated inhibitory neurotransmission; whereas treatment resistant MDD patients
demonstrated deficits in neurophysiological indexes of both GABAB and GABAA receptor-
mediated inhibition. Previous evidence has suggested that the altered function of the
GABAergic system may contribute significantly to the pathophysiology and potential
successful treatment of this disorder [Sanacora and Saricicek, 2007].
With regards to patients with SCZ, studies showed significant deficits in SICI, after
controlling for age and medications using a meta-regression. This finding shows specificity
of decreased SICI as a characteristic of SCZ. Previous research suggests that dysfunctional
cortical inhibition may be a mechanism through which symptoms of SCZ are mediated.
Altered markers of cortical GABAergic neurotransmission are consistently observed
abnormalities in postmortem studies of SCZ [Benes and Berretta, 2001; Lewis et al., 1999;
Stan and Lewis, 2012]. Similarly, several neurophysiological studies have found a reduction
in SICI and CSP duration in both medicated [Daskalakis et al., 2002a; Daskalakis et al.,
2008b; Liu et al., 2009] and unmedicated patients with SCZ [Daskalakis et al., 2002a;
Daskalakis et al., 2008b; Liu et al., 2009] suggesting deficits in cortical inhibition of the
motor cortex. Taken together, SICI may be a specific attribute when characterizing SCZ.
82
5.9.1 Clinical Implications This study provides compelling evidence to suggest that impairments in GABAergic
inhibition are involved in the pathophysiology of OCD, MDD and SCZ, nevertheless, the
overall pattern of these deficits differs. For example, in OCD, research has found inhibitory
deficits and enhanced intracortical facilitation, independent of medication status [Greenberg
et al., 2000; Greenberg et al., 1998; Richter et al., 2012]. By contrast, Levinson et al.
[Levinson et al., 2010] found that all MDD patients showed CSP abnormalities but only
treatment-resistant depressed patients demonstrated SICI reductions. MDD is associated with
deficits in neurophysiological indexes of GABAB receptor-mediated inhibitory
neurotransmission, whereas treatment-resistant patients demonstrated deficits in both
GABAB and and GABAA receptor-mediated inhibition. Treatment with antidepressants had
no apparent effects on either measure though other research has shown that selective
serotonin reuptake antidepressants normalize GABAergic deficits in depression through
enhanced SICI and decreased intracortical facilitation [Manganotti et al., 2001; Minelli et al.,
2010]. Serotonin is able to modulate excitatory and inhibitory effects, respectively mediated
by glutamate and GABA [Ciranna, 2006]. The serotonin receptor (5-HT) induces a decrease
of glutamate transmission and a parallel increase in GABA transmission evident in the
hippocoampus, frontal cortex and the cerebellum [Ciranna, 2006]. Previous studies have
shown that selective serotonin reuptake inhibitors (SSRIs) increase GABA by magnetic
resonance spectroscopy [Bhagwagar et al., 2004] and TMS [Robol et al., 2004]. The
modulatory action of the serotonin receptor (5-HT) may serve as a "brake" on neuronal
excitability. Given this inconsistency, replication is warranted to disentangle the effects of
medication. Finally, unmedicated SCZ patients have demonstrated impairments in SICI and
CSP [Daskalakis et al., 2002a]. Two studies have showed that clozapine-treated SCZ patients
demonstrated significantly longer CSP durations, implicating the role of the GABAB receptor
in clozapine [Daskalakis et al., 2008b; Liu et al., 2009]. Enhancing inhibition or decreasing
facilitation in the cortex through pharmacological or non-pharmacological means (i.e.,
electroconvulsive therapy, repetitive TMS, magnetic seizure therapy, cognitive behavioural
therapy) represent an important approach to targeted treatment. Further investigation is
83
needed to develop these TMS measures as neurophysiological markers of both diagnosis and
treatment.
5.9.2 Limitations This study is limited in several ways. First, studies assessing patients with OCD compared to
healthy controls had small sample sizes with limited amount of studies published in this field,
more work needs to be done in this population. Also, there is an overall lack of diagnostic
specificity of these neurophysiological deficits due to the overlap in results. It has been
shown that pharmacological treatment can have an effect on cortical inhibition in healthy
participants, [Langguth et al., 2008; Robol et al., 2004; Ziemann et al., 1998; Ziemann et al.,
1996a; b; Ziemann et al., 1997b] SCZ [Daskalakis et al., 2008b; Liu et al., 2009] and MDD
[Manganotti et al., 2001; Minelli et al., 2010]. No such medication effect has been reported in
OCD as inhibitory deficits in OCD have been found independent of medication status,
suggesting that these neurophysiological abnormalities may be trait related. However, more
studies are needed to investigate the impact of medications on cortical inhibition in
psychiatric disorders. Furthermore, these measures are traditionally limited to the motor
cortex which is a significant limitation since non-motor neurophysiological processes are of
primary interest. Other brain areas such as the dorsolateral prefrontal cortex may be more
proximal to the pathophysiology of these illnesses and can be measured by combining TMS
with electroencephalography [Daskalakis et al., 2008c; Farzan et al., 2010a; b; Fitzgerald et
al., 2008]. Lastly, there are differences in the TMS methodologies between studies. The
following approaches need to be implemented to have consistent measurements, for example,
CSP should be measured by stimulating an active contralateral muscle (i.e., 20% of
maximum contraction) at 140% of the resting motor threshold [Cantello et al., 1992] (Figure
1B). LICI should be evaluated by using a suprathreshold conditioning stimulus that precedes
a suprathreshold test stimulus at a 100 ms interstimulus interval [Valls-Sole et al., 1992]
(Figure 1C). SICI and intracortical facilitation should be assessed by using a subthreshold
conditioning stimulus set at 80% of the resting motor threshold that precedes a
suprathreshold test stimulus [Kujirai et al., 1993]. SICI is measured at interstimulus intervals
of 2 ms and 4 ms and intracortical facilitation is evaluated at interstimulus intervals of 10 ms,
84
15 ms and 20 ms [Kujirai et al., 1993; Nakamura et al., 1997] (Figures 1D and 1E).
Following these exact TMS guidelines can ensure rigorous methods across research groups.
5.9.3 Main Findings Summarized This meta-analytic review of motor cortex TMS paradigms in OCD, MDD and SCZ
(summarized in table 4) has revealed promising findings for objective clinical applications.
This study provides a meaningful summary of research in this field demonstrating a clear
platform from which further studies and diagnostic procedures can be developed.
Table 4. Summary of Significant Hedge's g Results in Psychiatric Populations.
Psychiatric
Disorder
Summary of Significant Hedge's g Results of TMS Paradigms
OCD Decreased cortical silent period (2 studies)
(g = -0.466, 95% confidence interval [-0.881,-0.052], p =
0.027)
Deficits in short-interval cortical inhibition (3 studies)
(g = 0.572, 95% confidence interval [0.179, 0.966], p = 0.004)
Enhanced intracortical facilitation (2 studies)
(g = 0.446, 95% confidence interval [0.042, 0.849], p = 0.030)
MDD Shortened cortical silent period (4 studies)
(g = -1.232, 95% confidence interval [-1.530, -0.933], p =
0.000)
Deficits in short-interval cortical inhibition (3 studies)
(g = 0.641, 95% confidence interval [0.384, 0.898], p=0.000)
SCZ Impairments in short-interval cortical inhibition (12 studies)
(g = 0.476, 95% confidence interval [0.331, 0.620], p = 0.000)
85
Chapter 6
Evidence for Inhibitory Deficits in the Prefrontal Cortex in
Schizophrenia
Contents of this chapter have been reprinted by permission from Oxford University Press
Radhu N, Garcia Dominguez L, Farzan F, Richter MA, Semeralul MO, Chen R, Fitzgerald
PB, Daskalakis ZJ. (2015). Evidence for Inhibitory Deficits in the Prefrontal Cortex in
Schizophrenia. Brain. 138 (Pt 2): 483-497.
A link to the published paper can be found at:
http://brain.oxfordjournals.org/content/138/2/483
86
6.1 Abstract Abnormal gamma-aminobutyric acid (GABA) inhibitory neurotransmission is a key
pathophysiological mechanism underlying SCZ. Transcranial magnetic stimulation (TMS)
can be combined with electroencephalography (EEG) to index long-interval cortical
inhibition (LICI), a measure of GABAergic receptor-mediated inhibitory neurotransmission
from the frontal and motor cortex. In previous studies we have reported that SCZ is
associated with inhibitory deficits in the DLPFC compared to healthy subjects and patients
with bipolar disorder. The main objective of the current study was to replicate and extend
these initial findings by evaluating LICI from the DLPFC in patients with SCZ compared to
obsessive-compulsive disorder (OCD). A total of 111 participants were assessed: 38 patients
with SCZ (average age: 35.71, 25 males, 13 females), 27 patients with OCD (average age:
36.15, 11 males, 16 females) and 46 healthy subjects (average age: 33.63, 23 females, 23
males)]. LICI was measured from the DLPFC and motor cortex through TMS-EEG. In the
DLPFC, LICI was significantly reduced in SCZ patients compared to healthy subjects (p =
0.004) and not significantly different between patients with OCD and healthy subjects (p =
0.5445). LICI deficits in the DLPFC were also significantly greater in patients with SCZ
compared to patients with OCD (p = 0.0465). There were no significant differences in LICI
across all three groups in the motor cortex. These results demonstrate that LICI deficits in the
DLPFC are specific to patients with SCZ and are not a generalized deficit that is shared by
disorders of severe psychopathology.
87
6.2 Introduction SCZ and OCD are psychiatric disorders associated with significant psychopathology and
personal suffering. SCZ is characterized by hallucinations, delusions, cognitive deficits and
negative symptoms [American Psychiatric Association, 2000]. OCD is associated with the
occurrence of unwanted and disturbing intrusive thoughts, images or impulses (obsessions),
followed by repetitive ritualistic behaviors (compulsions) completed in stereotyped
succession [American Psychiatric Association, 2000]. There are substantial areas of overlap
between SCZ and OCD [Lee et al., 2009]. Both are lifelong chronic conditions sharing a
similar distribution for age at onset and affect both men and women equally [Lee et al.,
2009]. Studies show that the rate of co-morbidity between SCZ and OCD is approximately
7% to 26% [Eisen et al., 1997; Fabisch et al., 1997; Porto et al., 1997; Poyurovsky et al.,
1999; Tibbo et al., 2000]. Abnormalities of the prefrontal cortex, anterior cingulate, caudate
nucleus, the basal ganglia, the thalamus, and the cerebellum have been implicated in both
SCZ and OCD [Gross-Isseroff et al., 2003; Tibbo and Warneke, 1999; Venkatasubramanian
et al., 2009]. Both disorders have been shown to have poor global functional performance
[Cavedini et al., 2002; Light and Braff, 2005a; b; van den Heuvel et al., 2005]. Finally, both
disorders also respond to dopaminergic antagonists and serotonin reuptake inhibitors,
suggesting pathophysiological overlap [Poyurovsky and Koran, 2005].
Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the brain,
critical in modulating cortical excitability and neuroplasticity [DeFelipe et al., 1986; Schieber
and Hibbard, 1993]. Cortical inhibition (CI) represents a neurophysiological index of
GABAergic inhibitory neurotransmission in the human cortex [Daskalakis et al., 2007;
Krnjevic, 1997]. Several lines of evidence suggest that CI is impaired in severe psychiatric
disorders. For example, previous TMS studies have demonstrated deficits in CI in patients
with SCZ [Daskalakis et al., 2002a; Daskalakis et al., 2008b; Fitzgerald et al., 2002a;
Fitzgerald et al., 2002b; Fitzgerald et al., 2003; Liu et al., 2009; Wobrock et al., 2010;
Wobrock et al., 2009; Wobrock et al., 2008], OCD [Greenberg et al., 2000; Greenberg et al.,
1998; Richter et al., 2012], major depressive disorder [Bajbouj et al., 2006b; Fitzgerald et al.,
88
2004a; Lefaucheur et al., 2008; Levinson et al., 2010], and bipolar disorder [Levinson et al.,
2007]. These findings suggest that many severe psychiatric disorders are associated with
deficits in GABAergic receptor-mediated inhibitory neurotransmission. However, these
findings were demonstrated in the motor cortex, a cortical region of limited interest in the
pathophysiology of psychiatric disorders.
This past decade has seen significant developments in the concurrent use of TMS with
electroencephalography (EEG) to directly assess the neurophysiology of the motor cortex
and DLPFC [Daskalakis et al., 2008c; Farzan et al., 2010b; Fitzgerald et al., 2008; Miniussi
and Thut, 2010; Taylor et al., 2008]. TMS-EEG allows for the measurement of a paired-pulse
paradigm known as LICI, whereby a suprathreshold CS is followed by a suprathreshold TS at
long ISIs (e.g. 50 ms - 200 ms), associated with the suppression of neuronal activity [Claus et
al., 1992; Valls-Sole et al., 1992]. A relationship between EMG and EEG measures of LICI
has been found in the motor cortex [Daskalakis et al., 2008c; Farzan et al., 2010b].
Additionally, LICI has demonstrated high test-retest reliability in the motor cortex and
DLPFC [Farzan et al., 2010b]. Evidence suggests that LICI is mediated by slow inhibitory
post-synaptic potentials via activation of GABAB receptors. For example, LICI is potentiated
by baclofen (GABAB receptor agonist) [McDonnell et al., 2006]. Furthermore, LICI is
optimal at 100 - 150 ms [Sanger et al., 2001] comparable to the time course of the GABAB
inhibitory post-synaptic potential, peaking at 150-200 ms post-stimulus [McCormick, 1989].
Lastly, LICI is evoked by a suprathreshold CS [Valls-Sole et al., 1992] as GABAB receptor-
mediated responses have higher activation thresholds and their inhibitory influence is longer
[Deisz, 1999a; Sanger et al., 2001].
GABA plays a pivotal role in the generation and inhibition of gamma oscillations in the
cortex [Bartos et al., 2007; Brown et al., 2007; Leung and Shen, 2007; Traub et al., 1996;
Wang and Buzsaki, 1996; Whittington et al., 1995; Whittington et al., 2000]. Research has
shown that GABAA receptor-mediated inhibitory post-synaptic potentials contribute to the
generation of gamma oscillations [Bartos et al., 2007; Wang and Buzsaki, 1996; Whittington
et al., 1995] and GABAB receptor-mediated inhibitory post-synaptic potentials are associated
with the inhibition of gamma oscillations [Brown et al., 2007; Leung and Shen, 2007;
89
Whittington et al., 1995]. Many studies involving neuropsychiatric disorders have focused on
abnormalities in gamma oscillations within the DLPFC due to its association with higher
order cognitive processes including sensory processing, attention, working memory, and
executive functioning, all domains in which SCZ patients are impaired [Uhlhaas et al., 2008].
For example, patients with SCZ have impairments in gamma oscillatory activity in response
to 40 Hz auditory stimulation [Light et al., 2006], during perception of gestalt objects
[Spencer et al., 2003] and during working memory performance [Barr et al., 2010; Cho et al.,
2006]. Disrupted gamma oscillatory activity has also been demonstrated in SCZ using TMS
and EEG [Farzan et al., 2010a; Ferrarelli et al., 2008; Frantseva et al., 2012]. One of the aims
of the study was to replicate findings by Farzan et al. [Farzan et al., 2010a] in which TMS-
EEG was used to assess LICI in the motor cortex and DLPFC in patients with SCZ compared
with bipolar patients and healthy subjects. Bipolar patients were included as these patients
are often treated with dopamine antagonists and can have comparable levels of
psychopathology. It was demonstrated by Farzan et al. that only patients with SCZ had
significant deficits in the inhibition of gamma oscillations that was specific to the DLPFC.
These results suggest that impairments in the inhibition of gamma oscillations in SCZ may be
closely associated with the pathophysiology of this disorder.
Thus, this study had two main objectives. The first was to assess LICI in patients with SCZ
and OCD in both the DLPFC and motor cortex using TMS-EEG. The second aim was to
evaluate the diagnostic specificity of LICI in SCZ and OCD, due to the comparable levels of
psychopathology severity and the similar pharmacology that are used to treat these two
disorders. The core study hypotheses were that both patients with SCZ and OCD would show
LICI deficits in the DLPFC relative to healthy subjects and deficits would be significantly
greater in patients with SCZ.
90
6.3 Materials and Methods A total of 111 participants were included [38 patients with SCZ (average age: 35.71, 25
males, 13 females), 27 patients with OCD (average age: 36.15, 11 males, 16 females) and 46
healthy subjects (average age: 33.63, 23 females, 23 males)] at the Centre for Addiction and
Mental Health in Toronto, Canada. The handedness of the participants were: SCZ patients
(33 right-handed, 3 left-handed, 2 ambidextrous), OCD patients (25 right-handed, 2 left-
handed), healthy subjects (39 right-handed, 4 left-handed, 3 ambidextrous). All subjects gave
their written informed consent and the protocol was approved by the Centre for Addiction
and Mental Health in accordance with the Declaration of Helsinki. The Structured Clinical
Interview for the Diagnostic and Statistical Manual for Mental Disorders (DSM)-IV
confirmed the diagnosis of SCZ or OCD. The exact medication dosage of SCZ and OCD
patients are included in table 5. Diagnostic information of the SCZ and OCD patients are
included in table 6. In healthy subjects, psychopathology was ruled out by the Structured
Clinical Interview for DSM-IV and subjects were only included in the study if they had no
first-degree relative diagnosed with a psychiatric disorder. Exclusion criteria for both patients
and healthy subjects included: (1) individuals meeting DSM-IV criteria for substance abuse
or dependence in the last 6 months, with the exception of nicotine; (2) concomitant major and
unstable medical or neurological illness; (3) experiencing suicidal ideation; (4) pregnant; (5)
positive urine toxicology screen for drugs of abuse; (6) any magnetic material or any other
conditions that would preclude the magnetic resonance image (MRI) scan or TMS-EEG
measures; (7) clinically significant claustrophobia. The exclusion criteria established by
international safety standards for TMS were followed [Rossi et al., 2009]. The TMS Adult
Safety Screen [Keel et al., 2001] was administered to all subjects.
91
Table 5A. Patients with Schizophrenia Medication Details
CLASS MEDICATIONS # OF SUBJECTS/DOSE(S) in
mg
ANTIPSYCHOTICS
Second Generation
Clozapine
n=11: 200 (2), 250 (2), 300 (5),
350 (1), 475 (1)
Olanzapine n=4: 7.5, 12.5, 15, 22.5
Paliperidone n=1: 150/4weeks
Quetiapine n=3: 300 (2), 400
Quetiapine Fumarate n=3: 200, 800, 900
Risperidone n=7: 2 (2), 3 (2), 5, 6, 8
Risperidone Injection n=2: 37.5/2 weeks, 75/4weeks
Ziprasidone n=2: 60, 120
Thioxanthenes Flupenthixol injection n=1: 60/3 weeks
Flupenthixol tabs n=1: 1
Zuclopenthixol injection n=2: 100/2 weeks, 280/2 weeks
Phenothiazines Fluphenazine Decanoate n=1: 37.5/2 weeks
Perphenazine n=1: 8
Dibenzoxazepines Loxapine n=2: 25, 30
Diphenylbutylpiperidines Pimozide n=1: 6
Third Generation Aripiprazole n=2: 15, 20
ANTIDEPRESSANTS
Selective serotonin re-uptake
inhibitors (SSRIs)
Citalopram n=3: 10, 40(2)
Escitalopram n=1: 20
92
Fluoxetine n=2: 10, 60
Paroxetine n=1: 20
Sertraline n=4: 75, 150, 200(2)
Serotonin–norepinephrine
reuptake inhibitors (SNRIs) Desvenlafaxine n=1: 30
norepinephrine-dopamine
reuptake inhibitor (NDRIs) Bupropion SR n=1: 200
MOOD STABILIZERS
Carbamazepine n=1: 400
Divalproex Sodium n=1: 1250
Lamotrigine n=1: 100
Lithium n=4: 900(2), 1050, 1200
Topiramate n=1: 200
BENZODIAZEPINES
Clonazepam n=3: 0.5 (2), 1
Clonazepam prn n=1: 0.25
Lorazepam n=1: 2
Lorazepam prn n=4: 1(2), 2(2)
OTHERS
Benzatropine n=3: 1(2), 4
93
Table 5B. Patients with Obsessive-Compulsive Disorder Medication Details
CLASS Medication # OF SUBJECTS/DOSE(S) in mg
ANTIDEPRESSANTS
Selective serotonin re-uptake
inhibitors (SSRIs)
Citalopram n=2: 20, 220
Escitalopram n=8: 15, 20(3), 30, 40(2), 50
Paroxetine n=3: 10, 25, 60
Fluoxetine n=2: 80(2)
Sertraline n=2: 100, 250
Serotonin–norepinephrine
reuptake inhibitors (SNRIs)
Duloxetine n=1: 120
Venlafaxine n=2: 100, 187.5
Tricyclic antidepressants
(TCAs) Clomipramine n=5: 50(2), 175, 250(2)
norepinephrine-dopamine
reuptake inhibitor (NDRIs) Bupropion n=1: 300
ANTIPSYCHOTICS
Loxapine n=1: 25
Olanzapine n=1: 20
Quetiapine n=4: 100, 150, 200, 600
Risperidone n=2: 0.125, 0.25
MOOD STABILIZERS
Divalproex Sodium n=1: 750
Lithium n=1: 1200
BENZODIAZEPINES
Clonazepam n=2: 0.5, 2
94
Table 5. Description of the Psychotropic Medications Displayed as Number of
Subjects/Dose(s).
Table 5A. Patients with Schizophrenia Medication Details.
Table 5B. Patients with Obsessive-Compulsive Disorder Medication Details.
Clonazepam prn n=1: 0.5
Diazepam prn n=2: 2.5, 4
Lorazepam prn n=4: 0.5 (2), 1, 2
Oxazepam n=1: 60
Temazepam n=1: 30
Temazepam prn n=1: 15
OTHERS
Buspar n=1: 10
L-Tryptophan n=1: 1500
Trazodone n=1: 25
Trazodone prn n=1: 25
Zopiclone n=2: 7.5, 15
Zopiclone prn n=1: 7.5
95
Schizophrenia (N = 38) Number of
Subjects %
Schizophrenia (paranoid type) 22 57.89 Schizoaffective (bipolar type) 13 34.21
Schizoaffective (depressive type) 3 7.89
Current Comorbidities Number of Subjects
%
Major Depressive Disorder 2 5.26 Obsessive-Compulsive Disorder 3 7.89
Panic Disorder without Agoraphobia
2 5.26
Panic Disorder with Agoraphobia
2 5.26
Social Phobia 2 5.26 Post-Traumatic Stress Disorder 1 2.63 Generalized Anxiety Disorder 1 2.63
Obsessive-Compulsive Disorder (N = 27)
Current Comorbidities Number of Subjects
%
Bipolar I Disorder 1 3.70 Major Depressive Disorder 4 14.81
Psychotic Disorder Not Otherwise Specified
1 3.70
Panic Disorder without Agoraphobia
3 11.11
Panic Disorder with Agoraphobia
2 7.41
Agoraphobia without Panic Disorder
1 3.70
Social Phobia 8 29.63 Specific Phobia 2 7.41
Generalized Anxiety Disorder 10 37.04 Body Dysmorphic Disorder 2 7.41
Table 6. Diagnostic Information for Schizophrenia and Obsessive-Compulsive Disorder
patients.
96
6.3.1 Clinical Severity The 24-construct Brief Psychiatric Rating Scale (BPRS) was used for evaluating
psychopathology in patients with SCZ [Overall and Gorham, 1962].
6.3.2 Transcranial Magnetic Stimulation Data Recording Monophasic TMS pulses were administered using a 7 cm figure-of-eight coil, and two
Magstim 200 stimulators (Magstim Company Ltd, UK) connected via a Bistim module. TMS
was administered over the left motor cortex and DLPFC. Inhibition was measured through
LICI and indexed through electromyography and EEG at the optimal 100 ms ISI [Sanger et
al., 2001]. One hundred TMS stimuli were delivered per-condition (paired and single-pulse)
every 5 seconds. The intensity of TMS pulses was determined at the beginning of each
experiment and it was set such that it elicited an average motor evoked potential of 1mV
peak-to-peak upon delivery of 20 pulses over the motor cortex. Both TMS pulses were
delivered at the same suprathreshold intensity. No significant between-group differences
were found for the 1mV peak-to-peak TMS intensity (F (2,108) = 0.794, p = 0.455) (healthy
controls = 69.46% ± 13.23%, SCZ = 73.03% ± 13.49%, OCD = 69.85% ± 14.53%).
6.3.3 Localization of the Motor Cortex The TMS coil was placed at the optimal position for eliciting motor evoked potentials from
the right abductor pollicis brevis muscle, which corresponded to a region between the
electrodes FC3 and C3.
6.3.4 Localization of the DLPFC Localization of the DLPFC was achieved through neuronavigation techniques using the
MINIBIRD system (Ascension Technologies) and MRIcro/registration software using a T1-
weighted MRI scan obtained for each subject with seven fiducial markers in place
[Daskalakis et al., 2008c; Farzan et al., 2010a]. Stimulation was directed at the junction of
the middle and anterior one-third of the middle frontal gyrus (Talairach coordinates (x, y, z)
97
= -50, 30, 36) corresponding with posterior regions of Brodmann area 9, which overlap with
the superior section of Brodmann area 46.
6.3.5 Electromyography Recording Electromyography was captured by placing two disposable disc electrodes over the right
abductor pollicis brevis muscle in a tendon-belly arrangement and motor evoked potentials
were filtered (band-pass 2 Hz to 5 kHz), digitized at 5 kHz (Micro 1401, Cambridge
Electronics Design, Cambridge, UK).
6.3.6 EEG Recording and Pre-Processing To evaluate TMS-induced cortical evoked potentials, EEG was recorded concurrently with
electromyography. EEG was acquired through a 64-channel Synamps 2 EEG system. A 64-
channel EEG cap was used to record the cortical signals, and four electrodes were placed on
the outer side of each eye, and above and below the left eye to closely monitor eye movement
artifacts. All electrodes were referenced to an electrode positioned posterior to Cz electrode.
EEG signals were recorded DC and with a low pass filter of 100 Hz at a 20 kHz sampling
rate, shown to avoid saturation of the amplifiers and minimize the TMS-related artifact
[Daskalakis et al., 2008c; Daskalakis et al., 2012].
EEG recordings were down-sampled to 1000 Hz and epoched from -1000 ms to 2000 ms
after the test TMS pulse. In both, the single and paired-pulse conditions, the segment from -
100 ms to 10 ms was removed (where 0 correspond to the test TMS pulse). This step
removes not only the test-pulse TMS in the single-pulse and paired-pulse conditions but also
the conditioning TMS pulse in the paired-pulse condition. Traces were visually inspected for
artifacts in order to eliminate trials and channels highly contaminated by noise (muscle
activity, 60Hz noise, movement-related activity as well as electrode artifacts). Two rounds of
ICA were subsequently applied. The first found was to minimize and remove the typical
TMS-related decay artifact that appears in some subjects at specific locations. In each
subject, the number of components that needed to be removed to eliminate this kind of
artifact varied from 0 to 6. Following this, a bandpass FIR filter was applied from 1 to 55 Hz
98
and a second round of ICA was computed to remove eye-related artifacts (blinks and
movements) and remaining muscle components. During this analysis, the subject identity and
group was hidden to the researcher using randomly generated files. No actual association of
each recording was made to a particular group (healthy, SCZ or OCD) when completing the
processing steps.
The time frequency decomposition was obtained using the Event-Related Spectral
Perturbation (ERSP) analysis in EEGLab. Specifically the analysis was wavelet based, using
a cycle of the complex Morlet wavelet across all frequencies. The ERSP was computed
independently for the single-pulse and paired-pulse conditions. The analysis is expressed in
decibels of spectral power (µV2/Hz) after subtracting the log baseline to the whole trial. For
comparison, a similar analysis was carried on, in a reduced sample of healthy subjects but
using the Discrete Fourier Transform instead of the wavelet [Garcia Dominguez et al., 2014].
The resulting time-frequency decompositions, for single-pulse and paired-pulse conditions,
were then subtracted (single-pulse minus paired-pulse), to obtain an index of inhibition.
Thus, a value of inhibition was obtained for each subject (46 healthy subjects, 38 SCZ and 26
OCD, 111 in total), site of stimulation (DLPFC, Motor), electrode (1…60), time (400 values,
-442 to 1442 ms) and frequency (50 values, 1…50Hz), that is 266,400,000 data points.
Based on [Garcia Dominguez et al., 2014], LICI was assessed with all electrodes and all
frequencies from 1-50 Hz. We proceed with a cluster-based analysis that is a development of
the one proposed in [Maris and Oostenveld, 2007], which provides corrections for multiple
comparisons due to the large amount of multidimensional data. In this paper, the clusters are
defined in the time-frequency space and also include the two extra spatial dimensions of the
electrode grid. Using this approach, we effectively reduced the number of total comparisons
to only group comparisons by site of stimulation. Since, in the original proposal of the cluster
test, the relative volume of the cluster depends on the threshold used to define the cluster; we
carried out, a fairly intuitive correction which consists of calculating the volumes for a set of
many discrete thresholds exhausting all the volumes to produce a single statistic: the average
volume t-score across all thresholds.
99
Before proceeding to the group comparisons, we assessed CI following the same
methodology within each group independently. To achieve this we applied a one-tailed,
paired t-test to every voxel or 3-tuple (i.e. time, frequency and channel) in the single-pulse
vs. the paired-pulse protocol. In this case the null hypothesis was that the single-pulse
condition is not larger than the paired-pulse, suggesting no inhibition. The method proceeds
as follows:
Step 1: Select all samples whose t-scores are larger than 1.6 and identify all the
clusters to which these voxels belong. Clusters are connected sets on the basis of
time, frequency and spatial adjacency.
Step 2: In each cluster sum all the t-scores and select the maximum of the obtained
values. This value corresponds to the size (or volume) of the major cluster of
inhibition and will be denoted by S.
Step 3: Repeat steps 2 and 3 with a new threshold that is 0.2 larger than the previous
one, obtaining successively new values for the maximum cluster indexed by the
threshold. Until a threshold is reached that does not contain any t-score.
Step 4: Repeat steps 2 to 4 with random reallocation of conditions across all subjects.
The number of random reallocations was chosen to be 10,000 for this specific
analysis.
Step 5: Proceed to calculate the proportion of the permuted values that are larger than
the original one. This proportion is the p-value.
In step 3 multiple cluster sizes are calculated, each for a specific threshold. This is
represented by a vector. In step 5 the original vector has to be compared to all the others
resulting from the randomization. This calculation proceeds as follows:
100
Let us denote the values of S as where k refers to the threshold index and the
superscript i to the randomization (where 0 is the original un-randomized data). One solution
is to compute the following ranks for each i and k:
Also, H counts the number of
times a positive difference arises.
That is, for each particular threshold, sum all the randomizations that have a score S higher
than randomization a, including the original case a = 0. After obtaining this matrix R, a sum
can be performed across the thresholds to obtain a single rank per randomization:
The actual p-value can be expressed as the proportion of R-values larger than the original R-
values.
The results of these analyses are shown in figures 1-3 for three different levels of alpha.
The comparison between groups proceeds the same way except that a) in step 4 the
randomization proceeds not by reallocating conditions but subjects between the two
populations compared preserving the number of subjects in each and b) the S-values are
calculated as the difference between the S values for each population at the same threshold.
The hypothesis is that if sample A is larger than B, then S=SA-SB, otherwise the order of the
factors does not matter and the p-values are computed from the population of absolute S
values is step 5.
Since t-scores are compared across different populations, differences in S values (i.e. SA-SB)
may be confounded by differences in the standard deviation between these two groups. We
101
consider that this effect is not detrimental to the analysis, as it may indicate intrinsic
population differences that should be taken into account. However, in order to observe
inhibitory differences attributed to the average (and not standard deviation), a secondary
analysis was performed in which the t-scores results as a pooled standard deviation from the
two groups compared. The p-values resulting from this analysis are displayed as the second
line of table 7.
Brain
Region
Group Channels 1-50 Hz Delta Theta Alpha Beta Gamma
DLPFC HCL>
SCZ
All 0.0040
0.0390
0.0555
0.1860
0.0175
0.0630
0.0025
0.0175
0.0005
0.0375
0.0405
0.3415
Local 0.0015
0.0140
0.0205
0.2210
0.0065
0.0305
0.0010
0.0100
0.0000
0.0230
0.0080
0.0805
HCL>
OCD
All 0.5445
0.8035
0.6585
0.7205
0.4290
0.6965
0.3295
0.8925
0.4445
0.7000
0.5950
0.7040
Local 0.4910
0.7070
0.6955
0.7165
0.5720
0.7185
0.4280
0.9415
0.2970
0.6140
0.1495
0.5075
SCZ<>OCD All 0.0465
0.0105
0.0710
0.1965
0.0295
0.0250
0.0170
0.0035
0.0565
0.0155
0.0565
0.5930
Local 0.0345
0.0110
0.0550
0.0835
0.0275
0.0175
0.0180
0.0010
0.0320
0.0280
0.1900
0.4895
Motor
HCL>
SCZ
All 0.9220
0.7380
0.7425
0.6965
0.6870
0.5630
0.6170
0.4550
0.9660
0.8545
0.6695
0.9590
Local 0.6870
0.9010
0.6845
0.8650
0.4980
0.4610
0.1785
0.2665
0.8535
0.9540
0.5370
0.4200
HCL>
OCD
All 0.8240
0.6845
0.5765
0.5950
0.5200
0.4180
102
0.7010 0.7020 0.5050 0.5290 0.5220 0.3255
Local 0.9870
0.8945
0.8295
0.8720
0.6850
0.5295
0.4970
0.3525
0.7855
0.5065
0.5565
0.2420
SCZ<>OCD All 0.4805
0.3755
0.4605
0.3000
0.3720
0.2590
0.4100
0.1940
0.5835
0.5665
0.4165
0.6685
Local 0.5700
0.3100
0.3095
0.2600
0.3770
0.2205
0.5150
0.2165
0.7540
0.5400
0.5325
0.6430
Table 7. All p-values of the between-group comparisons by site of stimulation, frequency
band, and electrode grids. The primary analysis is the first line; the second line displays the
pooled variance analysis.
This method was also applied to subsets of the original 4-dimensional space in order to
obtain additional information about the contribution of specific frequencies and electrodes to
the overall group differences. This time only 2000 random permutations were used. For these
purposes the time-frequency space was divided into the five common frequency bands [delta,
theta, alpha, beta, and gamma]. The local grid of electrodes for the DLPFC stimulation used
the following frontal electrodes: FP1, FPZ, FP2, AF3, AF4, F7, F5, F3, F1, FZ, F2, F4, F6,
F8, FT7, FC5, FC3, FC1, FCZ, FC2, FC4, FC6, FT8, while for the motor stimulation the
electrodes were: T7, C5, C3, C1, CZ, C2, C4, C6, T8, TP7, CP5, CP3, CP1, CPZ, CP2, CP4,
CP6, TP8.
In addition to the cluster-based analysis described, a Spearman's rho correlation analysis was
performed between the BPRS (total score) and the size of the larger cluster of DLPFC
inhibition for each SCZ subject. This correlation was also conducted with chlorpromazine
equivalents [American Psychiatric Association, 1997; Bezchlibnyk-Butler et al., 2014; Chue
et al., 2005; Woods, 2003]. The size was determined over the same 4D-space previously
illustrated by counting significant values of the larger cluster using a single threshold of
alpha level: 0.05. The size of the larger cluster of significant values is a way to capture the
103
degree of inhibition at the subject level, since a larger inhibition correlates with the extent of
significant voxels of inhibition in the time-frequency-spatial domain.
The methodological approach we follow in this study is a development of the one presented
in [Garcia Dominguez et al., 2014]. We have now added two dimensions to the
characterization of the cluster, the electrode grid. In this sense, we are better able to capture
and describe all the related responses into a unifying variable. The analysis presented here
also overcomes the dependence of the result on the threshold chosen to define the cluster by
applying the same analysis over a multiple of different thresholds and expressing the results
as a normalized average. This procedure may be considered as virtually threshold free, as
long as a sufficiently small step in the sequence of thresholds is considered.
6.4 Results 6.4.1 Comparing Single and Paired-Pulse Conditions (Within-Group
Analysis) A within-group cluster-based test was conducted to compare single-pulse and paired-pulse
TMS paradigms in order to assess LICI in the DLPFC and motor cortex for each 3-tuple:
electrode, time and frequency. These tests were single-tailed since the null hypothesis (no
inhibition) is when the paired-pulse amplitude is larger or equal to the amplitude of the
single-pulse. Figures 10 to 12 show the time-frequency map of LICI in the DLPFC (the
corresponding for motor cortex are shown in figures 15-17). Significant values imply that the
single-pulse induced response is higher than the paired-pulse condition. The areas of
significant inhibition across the time-frequency plots are designated as three shades of blue
corresponding to three different alpha levels: 0.05, 0.01 and 0.001. Figures 10-12 showed
that there was significant inhibition in most channels across many samples of the time-
frequency domain in all 3 groups. Lower frequencies tend to show extended inhibition up to
around 400ms after the test-pulse stimulation, while higher frequencies show inhibition over
narrower or specific temporal regions. Inhibition is particularly strong over the central,
midline channels.
104
In the DLPFC, all groups demonstrated significant within-group inhibition, healthy subjects
(p < 0.0001), SCZ (p = 0.0018), OCD, (p = 0.0002). That is, the single-pulse condition was
significantly greater than the paired-pulse condition. For the within-group analysis, the SCZ
group (figure 11) showed a reduced, but not absent, degree of inhibition in relation to healthy
subjects (figure 10). In fact, in all three groups, the pattern of inhibition in the time-frequency
space was similar (figures 10-12), as well as the topology (figures 13 and 14).
In the motor cortex (see figures 15-17), all groups demonstrated significant within-group
inhibition, healthy subjects (p < 0.0001), SCZ (p = 0.0002) and OCD (p = 0.01).
6.4.2 Between-Group Results for DLPFC Stimulation The between-group results were single-tailed to match the core study hypotheses
where healthy controls are expected to show more inhibition than SCZ and OCD. In addition,
the comparisons between the two patient groups were two-tailed, as we had no a priori
hypothesis in this case. Overall inhibition (1-50 Hz), assessed through the cluster mass test,
was significantly larger in healthy subjects than in patients with SCZ (p=0.004, i.e. only 40
random permutations, out of 10,000, showed a value for the difference in inhibition larger
than the one from the original samples). No significant differences were found between OCD
and healthy subjects. Significant differences were found between SCZ and OCD in overall
inhibition (p = 0.0465). Using the same approach, we investigated the contribution of
different frequency bands by partitioning the time-frequency space into 5 frequency bands
corresponding to delta, theta, alpha, beta and gamma. LICI was significantly different
between healthy subjects and SCZ in theta (p = 0.0175), alpha (p = 0.0025), beta (p =
0.0005) and gamma frequency bands (p=0.0405). Significant differences were found between
SCZ and OCD in the theta (p = 0.0295), and alpha frequency bands (p = 0.017).
6.4.3 Local Grid of Electrodes Analysis for DLPFC Stimulation Overall inhibition (1-50 Hz) was significantly larger in healthy subjects than in patients with
SCZ (p=0.0015) as well as in the: delta (p = 0.0205), theta (p = 0.0065), alpha (p = 0.001),
beta (p < 0.0001), and gamma frequency bands (p = 0.008). No significant differences were
105
found between OCD and healthy subjects. Significant differences were found between SCZ
and OCD in overall inhibition (p = 0.0345), theta (p = 0.0275), alpha (p = 0.018) and beta
frequency bands (p = 0.032).
6.4.4 Effect Size for DLPFC Stimulation Figure 18 shows the Cohen's d for each channel in the time-frequency space (DLPFC). Large
effect sizes are displayed between patients with SCZ and healthy subjects as demonstrated in
the frontal and midline regions of the time-frequency plot.
6.4.5 Motor Cortex Stimulation No significant differences were found between patients with SCZ and healthy subjects in
LICI across all frequency bands. No significant differences were found between OCD
patients and healthy subjects in LICI across all frequency bands. No significant differences
were found between the two patient groups.
6.4.6 Local Grid of Electrodes Analysis for Motor Cortex Stimulation No significant differences were found between patients with SCZ and healthy subjects in
LICI across all frequency bands. No significant differences were found between OCD
patients and healthy subjects in LICI across all frequency bands. No significant differences
were found between patients with SCZ and OCD.
6.4.7 Sum of T-Score Topology The sum of t-scores over the largest cluster is an index of the strength of inhibition at each
channel by band and by region as presented in figure 13 (DLPFC) and figure 14 (motor
cortex). In this case, red indicates greater inhibition over the specific electrodes in terms of a
wider and/or stronger magnitude of effect of inhibition over the time-frequency space.
106
6.4.8 Effect of Medication Treatments A between-group cluster-based analysis was conducted for the subset of antipsychotic-treated
OCD patients (n = 8) compared to SCZ patients (n = 8) (age and sex matched) on LICI
(DLPFC). SCZ patients showed deficits in LICI (p = 0.0298) with 10,000 randomizations.
Additionally, we evaluated the relationship between LICI (DLPFC) and antipsychotic
medications (converted chlorpromazine equivalents) [American Psychiatric Association,
1997; Bezchlibnyk-Butler et al., 2014; Chue et al., 2005; Woods, 2003] in patients with SCZ
treated with antipsychotic medications (n = 38). This analysis revealed no significant
correlation between LICI and chlorpromazine equivalents (Spearman’s rho = -0.0096, p =
0.9542). These results demonstrate that LICI deficits were not related to antipsychotic
treatment. Lastly, in the DLPFC, no significant differences were found between SSRI-
treated OCD patients (n = 17) and SSRI-treated SCZ patients (n = 11) in overall inhibition (p
= 0.52) and gamma inhibition (p = 0.49).
6.4.9 Clinical Severity Correlation Analysis We found a trending correlation between the BPRS and the largest cluster of inhibition in 38
SCZ patients (Spearman's rho = -0.2464, p = 0.0680). The greater the BPRS score is
indicative of increased severity of SCZ symptoms. This negative correlation signifies that the
higher BPRS score is related to a lower degree of inhibition. We ran an outlier detection
algorithm using criteria based on Cook’s distance and removed data points whose distance
were larger than 4/n (n = number of data points) [Bollen and Jackman, 1990]. Two outliers
were identified, after their removal the correlation was statistically significant (Spearman's
rho = -0.2855, p = 0.0457).
107
Figure 10. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in healthy subjects when
stimulating the dorsolateral prefrontal cortex.
Figure 11. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in schizophrenia patients when
stimulating the dorsolateral prefrontal cortex.
108
Figure 12. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in obsessive-compulsive disorder
patients when stimulating the dorsolateral prefrontal cortex.
109
Figure 13. Strength of inhibition by electrode. Each value consists of the sum of all t-scores
of inhibition in the major cluster of inhibition in the time-frequency maps for each electrode.
These plots show the three groups by frequency bands in the dorsolateral prefrontal cortex.
Values have been normalized within each frequency band. The color bar is omitted since
only the pattern matters, as the actual sum is dependent on the resolution of the time-
frequency-spatial domain.
Figure 14. Strength of inhibition by electrode. Each value consists of the sum of all t-scores
of inhibition in the major cluster of inhibition in the time-frequency maps for each electrode.
These plots show the three groups by frequency bands in the motor cortex. Values have been
normalized within each frequency band. The color bar is omitted since only the pattern
matters, as the actual sum is dependent on the resolution of the time-frequency-spatial
domain.
110
Figure 15. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in healthy subjects when
stimulating the motor cortex.
111
Figure 16. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in schizophrenia patients when
stimulating the motor cortex.
Figure 17. Statistical significance of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode in obsessive-compulsive disorder
patients when stimulating the motor cortex.
112
Figure 18. Effect size (Cohen’s d) of each voxel (single pulse vs. paired pulse) in the time-
frequency domain corresponding to each specific electrode when comparing schizophrenia
and healthy subjects in the dorsolateral prefrontal cortex. Red corresponds to when healthy
subjects show greater inhibition, blue corresponds to when SCZ show greater inhibition.
113
6.5 Discussion To summarize, in the DLPFC we found significant deficits in LICI in patients with SCZ
compared to healthy subjects but there were no significant LICI deficits in patients with
OCD. LICI deficits in the DLPFC were also significantly greater in patients with SCZ
compared to patients with OCD. Finally, there were no significant LICI differences across all
three groups in the motor cortex.
6.6 Frontal LICI Deficits in SCZ Two key lines of evidence support our findings of frontal LICI deficits in patients with SCZ.
First, previous studies have demonstrated reduced GABA inhibitory interneurons in the
DLPFC in SCZ [Akbarian et al., 1995; Benes et al., 1991]. For example, in SCZ, Benes et al.
reported morphologic changes in cortical GABA interneurons, by demonstrating a decreased
density of non-pyramidal cells in anterior cingulate layers II-VI and in prefrontal cortex layer
II [Benes et al., 1991]. Akbarian et al. found reduced messenger RNA in the DLPFC of SCZ
patients (a key enzyme involved in the synthesis of GABA) [Akbarian et al., 1995]. Impaired
GABAergic inhibitory neurotransmission in SCZ may be responsible for several of its key
phenotypic features. Dysfunctional GABAergic inhibitory neurotransmission may be related
to an imbalance between cortical excitation and inhibition [Yizhar et al., 2011]. Excessive
excitability in the cortex may result in discoordinated neuronal activation that may lead to the
disorganized behaviour and impulsivity that is commonly found in SCZ [Uhlhaas et al.,
2006; Uhlhaas and Singer, 2010]. Abnormal GABAergic inhibitory neurotransmission may
also lead to altered neural plasticity and aberrant neuronal wiring [Gaiarsa et al., 2002]. The
phenotypic manifestation of such dysfunction includes cognitive dysfunction, behavioural
disorganization, delusions and hallucinations [Constantinidis et al., 2002; Kapur, 2003;
Lewis et al., 2005].
Additional support for our findings of frontal LICI deficits in patients with SCZ relates to the
fact that LICI also plays a key role in modulating plasticity and in working memory
performance [Akerman and Cline, 2007; Butefisch et al., 2000; Deisz, 1999c;
Hoppenbrouwers et al., 2013]. For example, our group has previously shown a strong
114
positive correlation between frontal LICI and working memory [Daskalakis et al., 2008d;
Hoppenbrouwers et al., 2013]. Working memory impairment is considered a core cognitive
deficit in SCZ [Barrantes-Vidal et al., 2007; de Leeuw et al.]. The DLPFC is a functional
brain region critical for higher-order cognitive tasks such as executive functioning, attention,
and working memory performance [Barbey et al., 2012]. In SCZ, dysfunctional activation of
the DLPFC may underlie the working memory deficits present in this disorder [Weinberger
et al., 1986]. This has been demonstrated in both functional MRI (fMRI) [Jansma et al.,
2004; Karlsgodt et al., 2007; Karlsgodt et al., 2009; Potkin et al., 2009] and
neurophysiological studies [Barr et al., 2010; Cho et al., 2006]. Lastly, we demonstrated a
significant negative correlation between LICI and the clinical severity scores of SCZ. Thus,
significant impairments in frontal inhibitory neurotransmission may represent an important
mechanism underlying some of the key phenotypic features of SCZ.
6.6.1 Frontal Gamma LICI Deficits in SCZ Our findings replicated a previous study demonstrating impaired frontal gamma LICI in
SCZ. These results are a larger replication of a previous TMS-EEG study [Farzan et al.,
2010a], emphasizing a deficit in the inhibition of frontal gamma oscillations in SCZ. In this
current study, the findings are extended by showing overall inhibitory deficits in the DLPFC
is selective to patients with SCZ. Frontal gamma inhibitory deficits in SCZ may be due to the
hypofunction of the N-methyl-D-aspartate-receptor (NMDAR). It has been shown that a
blockade of the glutamate-mediated excitatory neurotransmission by NMDAR antagonists
mimics positive and negative symptoms as well as cognitive deficits in SCZ [Krystal et al.,
1994]. This hypothesis proposes a specific deficit in NMDAR signaling, leading to a
decrease in parvalbumin-positive GABAergic interneuron activity and consequent pyramidal
cell disinhibition, diminishing GABA synthesis and release [Gonzalez-Burgos and Lewis;
Moreau and Kullmann, 2013; Olney et al., 1999]. As reviewed above, there have been
several reports suggesting a relationship between GABAergic inhibitory neurotransmission
and gamma oscillations in the cortex [Bartos et al., 2007; Bragin et al., 1995; Brown et al.,
2007; Jefferys et al., 1996; Marrosu et al., 2006; Scanziani, 2000; Traub et al., 1997; Traub et
al., 1996; Wang and Buzsaki, 1996; Whittington et al., 1995]. Gamma oscillations appear to
115
be dependent on inhibitory neurotransmission from parvalbumin-containing GABA
interneurons. The lack of GABAergic neurotransmission in SCZ may translate into excessive
gamma oscillations leading to a pathophysiological plasticity (long-term potentiation) and
ultimately translate into aberrant learning and inflexible thinking that over time may lead to
delusions – a manifestation of erroneous information that is learned and reinforced.
6.6.2 Neurophysiology of OCD We found no significant LICI deficits in patients with OCD relative to healthy subjects in the
DLPFC and motor cortex. Previous research has shown that OCD has been associated with
motor cortex impairments in GABAA receptor-mediated inhibition [Greenberg et al., 2000;
Greenberg et al., 1998], GABAB receptor-medicated inhibition [Richter et al., 2012] and
NMDAR-mediated excitation [Richter et al., 2012]. Our findings could be accounted for by
the large medicated OCD sample, as the recent Richter et al., study mentioned above
included a majority of unmedicated patients with OCD (68%). SSRIs are the established
pharmacologic first-line treatment for OCD [Decloedt and Stein, 2010; Kellner, 2010]. In the
current study, 63% of the OCD patients (17/27) were medicated with SSRI's. Serotonin is
able to modulate excitatory and inhibitory effects, respectively mediated by glutamate and
GABA [Ciranna, 2006]. The serotonin receptor (5-HT) induces a decrease of glutamate
transmission and a parallel increase in GABA transmission evident in the hippocoampus,
frontal cortex and the cerebellum [Ciranna, 2006]. Previous studies have shown that SSRI’s
increase GABA by magnetic resonance spectroscopy [Bhagwagar et al., 2004] and TMS
[Robol et al., 2004], thus concealing any potential LICI deficits in the present study. The
modulatory action of the serotonin receptor (5-HT) may serve as a "brake" on neuronal
excitability. Given this inconsistency, replication is warranted to disentangle the effects of
medication. Future directions of this work may be to evaluate LICI (TMS-EEG) before and
after SSRI treatment for OCD to establish a relationship between inhibition and therapeutic
response.
116
6.7 Advancements in Analyses In this paper, we have improved our methodology over previous LICI analyses. Our
advanced analyses provide two main benefits: a concise characterization of LICI, including
all relevant dimensions in the data (time, frequency, space) that may have been omitted in
previous studies and addresses the issue of multiple comparisons. Specifically, the analysis
allows for the assessment of LICI over the whole brain cortical network by means of an
extended time and spatial domain [Garcia Dominguez et al., 2014]. The analysis can also be
applied to a subset of the original grid to assess the contribution of specific electrodes.
Inhibition is characterized as a continuous response over a 4-dimensional space and is not
linked to a particular fixed pre-conceived window in this space. Past results have attempted a
variety of methodologies because there is not an immediate, easy to recognize, feature that
indexes LICI. Previous analyses have revealed evidence for inhibition over a restricted
domain, by sacrificing either, the temporal, the spatial or the frequency component.
Examples are: the fixed window analysis [Daskalakis et al., 2008c; Farzan et al., 2010b], the
measure of peak amplitude [Rogasch et al., 2013] and the analysis over a 25ms sliding
window [Fitzgerald et al., 2009a] at a single 1-40Hz band-pass. A second major advantage of
the analysis is that it tackles the problem of multiple comparisons. We applied a cluster-
based permutation analysis, allowing for the use of a single statistic from the whole
multidimensional space [Maris and Oostenveld, 2007; Premoli et al., 2014a]. This global
analysis contains the adequate correction for all of the sub-analyses between the
corresponding groups. In the context of TMS-EEG, limited studies have examined the entire
time-frequency and spatial domain since presenting a multidimensional analysis increases the
likelihood of committing a Type I error due to the problem of multiple comparisons [Maris
and Oostenveld, 2007]. Taken together, the presented methodology is parameter free, while
at the same time, avoids the multiple comparison issue without the need to discard vital
information as done in previous TMS-EEG analyses.
117
6.8 Limitations This study is limited in several ways. First, SCZ patients were treated with a variety of
antipsychotic medications and other psychotropic medications and were chronically ill,
which may have effects on neural oscillations and may explain the lack of motor inhibitory
deficits found in this disorder. Future studies should recruit unmedicated SCZ and OCD
patients. Second, while pharmacological findings suggest that LICI is mediated by slow
inhibitory post-synaptic potentials via activation of GABAB receptors [McDonnell et al.,
2006], the effect of other neurotransmitter systems cannot be completely ruled out (dopamine
and serotonin). Third, when interpreting these results, inferences have been made for the role
of LICI vis à vis cognition. We did not measure cognition in this study; therefore, there is
currently no direct evidence to support these implications. In order to better link
neurophysiological findings to the symptoms of SCZ and OCD, future studies should
establish better relationships between LICI, cognition and behavior. More research should be
done to demonstrate the behavioural impacts of decreased inhibition in SCZ. Lastly, the TMS
evoked-potential after 40 ms contains the afferent component in the motor cortex which may
affect inhibition; however, any potential artifact is expected to be similar between groups and
should not account for group differences.
6.9 Clinical Implications This study shows that impairments in frontal GABAB receptor-mediated inhibitory
neurotransmission are associated with pathophysiology specific to SCZ. Conceivably TMS
measures of GABAergic and NMDAR functioning could be used as biological markers of
novel treatments that are aimed at enhancing inhibition or decreasing excitation in the cortex.
Several lines of evidence have suggested that the potentiation in GABAB receptor-mediated
inhibition are associated with clinical improvements as demonstrated by clinical
interventions such as meditation [Guglietti et al., 2013], cognitive behavioral therapy [Radhu
et al., 2012], repetitive TMS [Daskalakis et al., 2006], electroconvulsive therapy [Bajbouj et
al., 2006a] and clozapine treatment for SCZ [Daskalakis et al., 2008b; Liu et al., 2009; Wu et
al., 2011]. These results are promising and suggest the potential of using TMS-EEG in
neurophysiological research and in clinical settings.
118
Chapter 7
Investigating the Heritability of Cortical Inhibition in First-Degree
Relatives and Probands in Schizophrenia Radhu N, Garcia Dominguez L, Greenwood TA, Farzan F, Semeralul MO, Richter MA,
Kennedy JL, Blumberger DM, Chen R, Fitzgerald PB, Daskalakis ZJ.
Manuscript Submitted
119
7.1 Abstract Deficits in GABAergic inhibitory neurotransmission are a reliable finding in patients with
SCZ as demonstrated through multiple investigational approaches. Previous studies have also
reported that unaffected first-degree relatives of patients with SCZ demonstrate
neurophysiological abnormalities that are intermediate between probands and healthy
controls. In this study, first-degree relatives of patients with SCZ and their related probands
were investigated to assess frontal cortical inhibition – a neurophysiological index of
GABAergic inhibitory neurotransmission - as a potential endophenotype of SCZ. LICI was
measured from the DLPFC using combined TMS and EEG. The study presents an extended
sample of 129 subjects (66 subjects have been previously reported): 19 patients with SCZ (9
females, 10 males, average age = 30.2 years), 30 unaffected first-degree relatives of these
SCZ patients (17 females, 13 males, average age = 53.8 years), 13 obsessive-compulsive
disorder (OCD) patients (9 females, 4 males, average age = 28.9 years), 18 unaffected first-
degree relatives of these OCD patients (12 females, 6 males, average age = 41.9 years), and
49 healthy subjects (25 females, 24 males, average age = 33.4 years). In the DLPFC, cortical
inhibition was significantly less in patients with SCZ (t = 1.76, df = 66, p = 0.041) compared
to healthy subjects. First-degree relatives of patients with SCZ showed significantly more
cortical inhibition than their SCZ probands (t = 2.14, df = 47, p = 0.038). No differences
were demonstrated between first-degree relatives of SCZ patients and healthy subjects (t = -
0.44, df = 77, p = 0.66). While not significant, family members of SCZ were shown to be
intermediate between their related probands and healthy controls. No differences were found
between healthy subjects compared to OCD patients (t = -1.08, df = 60, p = 0.29) and their
first-degree relatives (t = -0.33, df = 65, p = 0.75). Lastly, no frontal inhibition differences
were shown between OCD patients and their first-degree relatives (t = -0.64, df = 29, p =
0.52). Altered frontal inhibition was specific to SCZ and their first-degree relatives, as no
inhibitory differences were found in OCD and their first-degree relatives, demonstrating
further insight into the biological mechanisms of SCZ. Larger family-based studies are
needed to establish frontal inhibition as an endophenotype of SCZ.
120
7.2 Introduction SCZ is a severe psychotic disorder characterized by positive symptoms (i.e. hallucinations
and delusions), negative symptoms (i.e. affective flattening and motivational deficits), and
cognitive impairments [American Psychiatric Association, 2000]. OCD typically manifests in
compulsive urges to perform irrational behaviors associated with the occurrence of
obsessions (disturbing intrusive thoughts or impulses) [Abramowitz et al., 2009; American
Psychiatric Association, 2000; Heyman et al., 2006; Stein, 2002]. It has been shown that
there is considerable overlap between SCZ and OCD in their pathophysiology, clinical
symptom profile, and treatment [Poyurovsky and Koran, 2005].
In psychiatry, there are no objective laboratory tests to inform diagnoses and monitor
response to interventions. Biomarkers facilitate the development of etiologic rather than
symptom-based diagnostic methods and also help to advance our understanding of the
genetic mechanisms underlying psychiatric disorders [Daskalakis, 2012; Turetsky et al.,
2007]. Genetic load is known to be one of the strongest determinants for the development of
SCZ with heritability estimates as high as 80% [Cardno et al., 1999a; Cardno et al., 1999b].
Endophenotypes are a way to biologically assess psychiatric illnesses. In order to validate an
endophenotype, specific criteria of the biomarker must be fulfilled, e.g., high test-retest
reliability of the biomarker, trait stability and large effect size differences between patients
and healthy subjects. Lastly, it is important to determine the genetic component in the
marker; a true biomarker shows that unaffected first-degree relatives have an intermediate
deficit in the trait when compared to their related probands and healthy subjects [Gottesman
and Gould, 2003]. Unaffected first-degree relatives of SCZ patients are ideal candidates for
biomarker development as they share degrees of genetic vulnerability with probands,
however, are free from confounding variables such as antipsychotic treatment and
psychopathology, helpful in evaluating the neurobiological mechanisms of the disease [Hall
et al., 2011].
Deficits in GABAergic inhibitory neurotransmission have been a reliable finding in SCZ
across multi-modal approaches. These deficits may be due to an imbalance between cortical
121
excitation and inhibition of the cortex [Yizhar et al., 2011]. For example, Benes et al.
reported a decreased density of non-pyramidal cells in anterior cingulate layers II-VI and in
prefrontal cortex layer II in SCZ [Benes et al., 1991]. Akbarian et al. found reduced
messenger RNA (involved in the synthesis of GABA) in the DLPFC of SCZ patients
[Akbarian et al., 1995]. Additional studies have shown that SCZ patients exhibit deficits in
GABAergic inhibition using transcranial magnetic stimulation (TMS) [Daskalakis et al.,
2002a; Daskalakis et al., 2008b; Fitzgerald et al., 2002a; Fitzgerald et al., 2002b; Fitzgerald
et al., 2003; Liu et al., 2009; Wobrock et al., 2010; Wobrock et al., 2009; Wobrock et al.,
2008], limited to the motor cortex.
TMS combined with EEG is a powerful tool for investigating cortical mechanisms and
networks of frontal brain areas. Using this approach, GABAB receptor-mediated inhibitory
neurophysiological mechanisms can be measured through a paired-pulse paradigm, LICI. We
have demonstrated using TMS-EEG that LICI of gamma oscillations were selectively
impaired in the DLPFC of patients with SCZ compared to both healthy subjects and similarly
treated patients with bipolar disorder [Farzan et al., 2010a]. Patients with bipolar disorder
were similar to patients with SCZ in relation to severity of symptoms, illness duration, and
history of psychosis. In a recent study, we found that frontal LICI was significantly reduced
in SCZ patients, compared to OCD patients and healthy subjects, also showing no effect of
antipsychotic medication [Radhu et al., 2015]. These findings suggest that LICI
abnormalities may be specific to SCZ and are not part of a generalized deficit associated with
severe psychopathology.
Limited studies have investigated first-degree relatives of SCZ patients using TMS
paradigms. Saka et al., [Saka et al., 2005] evaluated TMS measures of inhibition in
unaffected first-degree relatives of SCZ patients compared to healthy subjects (no proband
group was assessed). They found that 25% of first-degree relatives lacked transcallosal
inhibition and showed psychosis-proneness relative to healthy controls. No differences were
found in MEP amplitude (excitability measure) or the cortical silent period (inhibitory
measure). The above findings provide evidence for the genetic liability of these TMS
markers, highlighting the need for further research in this area. Several lines of evidence have
122
found a relationship between GABAergic inhibitory neurotransmission and gamma
oscillations [Bartos et al., 2007; Brown et al., 2007; Leung and Shen, 2007; Traub et al.,
1996; Wang and Buzsaki, 1996; Whittington et al., 1995; Whittington et al., 2000].
Specifically, GABAA receptor-mediated inhibitory post-synaptic potentials contribute to the
generation of gamma oscillations [Bartos et al., 2007; Wang and Buzsaki, 1996; Whittington
et al., 1995], whereas, GABAB receptor-mediated inhibitory post-synaptic potentials have
been associated with the modulation of gamma oscillations (Whittington et al., 1995; Brown
et al., 2007b; Leung and Shen, 2007). Growing evidence has shown that abnormalities of
high-frequency oscillations in the gamma-range (30-50 Hz) via EEG are heritable. For
example, Hall et al. [Hall et al., 2011] examined the early auditory gamma-band response in
SCZ and their unaffected co-twins during an auditory oddball target detection task in 194
individuals. They found that both evoked power and phase-locking phenotypes were reduced
in unaffected co-twins of patients with SCZ, with heritability of 0.65 for evoked power and
0.63 for phase-locking. This work suggests that the gamma band response may have a
heritable component in SCZ.
The main objective of this study was to evaluate DLPFC and motor cortex overall and
gamma inhibition using the LICI paradigm in patients with SCZ, patients with OCD, their
unaffected first-degree relatives and compared these groups to healthy subjects. We
hypothesized that frontal inhibition deficits would be demonstrated in SCZ and that this
group would also show the greatest LICI impairment. Furthermore, we hypothesized that
frontal inhibition in first-degree relatives of SCZ would be intermediate of healthy subjects
and their related probands. Lastly, inhibitory deficits would not be shown in OCD patients
and their unaffected first-degree relatives.
123
7.3 Materials and Methods The study assessed 129 subjects: 19 patients with SCZ (9 females, 10 males, average age =
30.2 years, age range: 22 - 42 years); 30 first-degree relatives of SCZ patients, including 23
parents (13 mothers, 10 fathers) and 7 full siblings (4 sisters, 3 brothers), with a total of: 17
females, 13 males, average age = 53.8 years, age range: 23 - 69 years. There were a total of
18 families (related to SCZ patients) with an average of 1.61 members (range: 1 - 4
members). 13 OCD patients (9 females, 4 males, average age = 28.9 years, age range: 19 - 40
years); 18 first-degree relatives of OCD patients, including 9 parents (5 mothers, 4 fathers)
and 9 full siblings (7 sisters, 2 brothers), with a total of: 12 females, 6 males, average age =
41.9 years, age range: 20 – 71 years and 49 healthy subjects (25 females, 24 males, average
age = 33.4 years, age range: 20 – 56 years). There were a total of 13 families (related to OCD
patients) with an average of 1.38 members (range: 1-3 members). The results from a subset
of the subjects studies has been published (13 SCZ patients, 7 OCD patients and 43 healthy
subjects) [Radhu et al., 2015].
All subjects gave their written informed consent and the protocol was approved by the Centre
for Addiction and Mental Health in accordance with the Declaration of Helsinki. The
Structured Clinical Interview for the Diagnostic and Statistical Manual for Mental Disorders
(DSM)-IV confirmed the diagnosis of SCZ or OCD. Medications of SCZ and OCD patients
are shown in table 1. Diagnostic information of the SCZ and OCD patients are included in
table 2. In healthy subjects, psychopathology was ruled out by the Structured Clinical
Interview for DSM-IV and healthy subjects were only included in the study if they had no
first-degree relative diagnosed with a psychiatric disorder. Healthy subjects and all first-
degree relatives of probands were administered the Family Interview for Genetic Studies
[Calkins et al., 2007]. Relatives of probands had no psychopathology in the last 2 years as
ruled out through the Structured Clinical Interview for DSM-IV. First-degree relatives were
recruited through advertisements as well as from referrals from their related probands that
were enrolled in the study. Additionally, recruitment methods included advertisements on
public transportation (Toronto subway cars) for a period of one month. At least one first-
degree relative of a proband was a requirement for this study; either one biological parent of
124
a proband or full sibling of a proband and the proband had to be available for
neurophysiological assessments. Exclusion criteria for all research subjects included: (1)
individuals meeting DSM-IV criteria for substance abuse or dependence in the last 6 months,
with the exception of nicotine; (2) concomitant major and unstable medical or neurological
illness; (3) experiencing suicidal ideation; (4) pregnant; (5) positive urine toxicology screen
for drugs of abuse; (6) any magnetic material or any other conditions that would preclude the
magnetic resonance image (MRI) scan or TMS-EEG measures; (7) clinically significant
claustrophobia. The exclusion criteria established by international safety standards for TMS
were followed [Rossi et al., 2009]. The TMS Adult Safety Screen [Keel et al., 2001] was
administered to all subjects.
125
Table 8A. Patients with Schizophrenia Medication Details
CLASS MEDICATION # OF SUBJECTS/DOSE(S) in mg
ANTIPSYCHOTICS
Second Generation Clozapine n=9: 150 (1), 200 (1), 250 (2), 300 (2), 350 (1), 400 (1) 475 (1)
Olanzapine n=2: 7.5, 22.5
Paliperidone n=1: 150/4 weeks
Quetiapine n=1: 300
Risperidone n=3: 2 (2), 3
Risperidone
Injection n=2: 50/2 weeks, 75/4 weeks
Ziprasidone n=1: 60
Dibenzoxazepines Loxapine n=1: 30
Third Generation Aripiprazole n=2: 20, 30
ANTIDEPRESSANTS
Selective serotonin re-
uptake inhibitors (SSRIs) Citalopram n=2: 40 (2)
Serotonin–norepinephrine
reuptake inhibitors (SNRIs) Desvenlafaxine n=1: 50
Norepinephrine-dopamine
reuptake inhibitor (NDRIs) Bupropion SR n=1: 150
MOOD STABILIZERS
Divalproex Sodium n=1: 500
Lamotrigine n=1: 100
Topiramate n=1: 200
BENZODIAZEPINES
Clonazepam n=2: 0.5, 1
Clonazepam prn n=1: 0.25
Lorazepam prn n=1: 2
OTHERS
Benzatropine n=1: 2
126
Table 8B. Patients with Obsessive-Compulsive Disorder Medication Details
CLASS MEDICATION # OF SUBJECTS/DOSE(S) in mg
ANTIDEPRESSANTS
Selective serotonin re-uptake
inhibitors (SSRIs)
Escitalopram n=1: 50
Fluoxetine n=2: 20, 80
Sertraline n=1: 250
Serotonin–norepinephrine
reuptake inhibitors (SNRIs) Duloxetine n=1: 60
Tricyclic antidepressants
(TCAs) Clomipramine n=4: 50, 250 (3)
Norepinephrine reuptake
inhibitor (NRIs) Atomoxetine n=1: 80
ANTIPSYCHOTICS
Loxapine n=1: 25
Olanzapine n=1: 20
Aripiprazole n=1: 2
MOOD STABILIZERS
Divalproex Sodium n=1: 750
BENZODIAZEPINES
Clonazepam n=1: 0.5
Clonazepam prn n=1: 0.5
Diazepam prn n=1: 4
Temazepam n=1: 30
Table 8. Description of the Psychotropic Medications Displayed as Number of
Subjects/Dose(s).
Table 8A. Patients with Schizophrenia Medication Details.
Table 8B. Patients with Obsessive-Compulsive Disorder Medication Details.
127
Schizophrenia (N = 19) Number of
Subjects %
Schizophrenia (paranoid type) 14 73.68 Schizoaffective (bipolar type) 5 26.32
Current Comorbidities Number of
Subjects %
Major Depressive Disorder 1 5.26 Post-Traumatic Stress Disorder 1 5.26
Panic Disorder without Agoraphobia
1 5.26
Obsessive-Compulsive Disorder (N = 13)
Current Comorbidities Number of Subjects
%
Social Phobia 4 30.77 Panic Disorder without
Agoraphobia 2 15.38
Generalized Anxiety Disorder 4 30.77 Major Depressive Disorder 1 7.69
Attention Deficit Hyperactivity Disorder
1 7.69
Table 9. Diagnostic Information for Schizophrenia and Obsessive-Compulsive Disorder
patients.
7.3.1 Clinical Severity The Schizotypal Personality Questionnaire (SPQ) [Raine, 1991] was used for evaluating
psychopathology in first-degree relatives of SCZ patients. The 24-construct Brief Psychiatric
Rating Scale (BPRS) was used for evaluating psychopathology in SCZ patients [Overall and
Gorham, 1962].
128
7.3.2 Data Recording
7.3.3 Transcranial Magnetic Stimulation Monophasic TMS pulses were administered using a 7 cm figure-of-eight coil, and two
Magstim 200 stimulators (Magstim Company Ltd, UK) connected via a Bistim module. TMS
was administered over the left motor cortex and DLPFC. Inhibition was measured through
LICI and indexed through electromyography and EEG at the optimal 100 ms interstimulus
interval [Sanger et al., 2001]. One hundred TMS stimuli were delivered per-condition (paired
and single-pulse) every 5 seconds. The intensity of TMS pulses was determined at the
beginning of each experiment and it was set such that it elicited an average motor evoked
potential of 1mV peak-to-peak upon delivery of 20 pulses over the motor cortex. Both the
conditioning stimulus and test stimulus were delivered at the same suprathreshold intensity.
7.3.4 Localization of the Motor Cortex The TMS coil was placed at the optimal position for eliciting motor evoked potentials from
the right abductor pollicis brevis muscle, which corresponded to a region between the
electrodes FC3 and C3.
7.3.5 Localization of the DLPFC Localization of DLPFC was achieved through neuronavigation techniques using the
MINIBIRD system (Ascension Technologies) and MRIcro/registration software using a T1-
weighted MRI scan obtained for each subject with seven fiducial markers in place
[Daskalakis et al., 2008c; Farzan et al., 2010a]. Stimulation was directed at the junction of
the middle and anterior one-third of the middle frontal gyrus (Talairach coordinates (x, y, z)
= -50, 30, 36) corresponding with posterior regions of Brodmann area 9, which overlap with
the superior section of Brodmann area 46.
129
7.3.6 Electromyography Recording Electromyography was captured by placing two disposable disc electrodes over the right
abductor pollicis brevis muscle in a tendon-belly arrangement and motor evoked potentials
were filtered (band-pass 2 to 5 kHz), digitized at 5 kHz (Micro 1401, Cambridge Electronics
Design, Cambridge, UK).
7.3.7 EEG Recording and Pre-Processing To evaluate TMS-induced cortical evoked potentials, EEG was recorded concurrently with
electromyography. EEG was acquired through a 64-channel Synamps 2 EEG system. A 64-
channel EEG cap was used to record the cortical signals, and four electrodes were placed on
the outer side of each eye, and above and below the left eye to closely monitor eye movement
artifacts. All electrodes were referenced to an electrode positioned posterior to Cz electrode.
EEG signals were recorded DC and with a low pass filter of 100 Hz at a 20 kHz sampling
rate, shown to avoid saturation of the amplifiers and minimize the TMS-related artifact
[Daskalakis et al., 2008c; Daskalakis et al., 2012].
EEG recordings were down-sampled to 1000 Hz and epoched from -1000 ms to 2000 ms
after the test TMS pulse. In both, the single and paired-pulse conditions, the segment from -
100 ms to 10 ms was removed (where 0 correspond to the test TMS pulse). This step
removes not only the test-pulse TMS in the single-pulse and paired-pulse conditions but also
the conditioning TMS pulse in the paired-pulse condition. Traces were visually inspected for
artifacts in order to eliminate trials and channels highly contaminated by noise (muscle
activity, 60Hz noise, and movement-related activity as well as electrode artifacts). Two
rounds of ICA were subsequently applied. The first round was to minimize and remove the
typical TMS-related decay artifact that appears in some subjects at specific locations. In each
subject, the number of components that needed to be removed to eliminate this kind of
artifact varied from 0 to 6. Following this, a bandpass FIR filter was applied from 1 to 55 Hz
and a second round of ICA was computed to remove eye movement-related artifacts (blinks
and movements) and remaining muscle components.
130
7.4 Post-Processing Analyses A time frequency decomposition was obtained using the Event-Related Spectral Perturbation
(ERSP) analysis in EEGLab. Specifically the analysis was wavelet based, using a cycle of the
complex Morlet wavelet across frequencies 2 - 50Hz. The ERSP was computed
independently for the single-pulse and paired-pulse conditions. The analysis is expressed in
decibels of spectral power (µV2/Hz) after subtracting the log baseline to the whole trial. In
previous studies, we have shown inhibition can be evaluated as the difference, of some
suitable measure of amplitude, of the evoked activity between the single and paired-pulse
stimulation. In this study, the measure of amplitude is the power of a wavelet decomposition.
Nine electrodes were retained for the analysis of inhibition for both the DLPFC and motor
cortex stimulation. This subset of electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2) was
located midline over the frontal-central regions and were chosen for two main reasons. First,
these nine electrodes were the least influenced by muscle activity and TMS-related artifacts,
thus, these electrodes were not excluded due to artifacts. Second, these electrodes show the
greatest and most consistent inhibitory response for both the motor cortex and DLPFC site of
stimulation [Garcia Dominguez et al., 2014; Radhu et al., 2015]. These two reasons are
related since the large signal-to-noise-ratio is mainly due to low noise from muscle combined
with a relatively good proximity and connectivity to the site of stimulation. This was a region
of interest of analysis; the artificat removal was completed soley on these nine electrodes.
7.4.1 Calculating Inhibition by Subject In this study we computed the difference in the evoked power of the two conditions, we also
computed a number of paired surrogate conditions made of sets of randomly selected trials,
without replacement from the pool of trials of the two conditions. The surrogate conditions
serves as a baseline or null hypothesis for the case of no inhibition, i.e. no difference between
the powers of conditions. The power differences extracted from the original conditions as
well as the surrogate ones consist of a set of values over voxels, in the time-frequency-
electrode space for each subject. From this “landscape” of values over the time-frequency-
electrode space, a threshold (p-value) was chosen to label each voxel as inhibited=1 or not-
inhibited=0. A voxel received a value of “1” if its value is greater than 99% of the values in
131
the same voxel in the null distribution, otherwise is “0”. The number of randomization in the
null distribution was set to 500. Inhibition was then evaluated by counting the 1’s that forms
a cluster in this time-frequency-electrode domain. A voxel belongs to a cluster if it has a
value of 1, and has at least a neighbour in time, frequency and electrode that is also in the
cluster [Garcia Dominguez et al., 2014]. Thus, for each subject an index of inhibition is
defined as the total sum of significant values in the largest cluster, if more than one cluster is
found. The cluster is only considered in a time domain from the moment of the test pulse to
500 ms after, and frequencies from 2-50Hz. If the analysis is restricted to the gamma band
we sum only over the range: 30-50Hz. The size of the larger cluster of significant values (or
index of inhibition) is a way to capture the degree of inhibition at the subject level, since a
larger inhibition correlates with the extent of significant voxels in the time-frequency-spatial
domain.
7.4.2 Between-Group Analyses After calculating inhibition by subject, we compared groups by a t-test analysis with pooled
variance. In all comparisons, the two-tailed analyses are reported. However, based on our
previous finding [Radhu et al., 2015], one analysis was single-tailed when we compared
healthy subjects to SCZ patients in overall (2-50Hz) and gamma (30-50Hz) inhibition. The
null hypothesis being that the control group is not more inhibited than the SCZ group, in both
the overall frequency band and the gamma frequency band.
Figure 1 depicts differences in inhibition evaluated non- parametrically for patients with
SCZ, their first-degree relatives and healthy subjects, this is how it was calculated:
For each group 19 subjects that were chosen with replacement and a value of inhibition was
obtained from this subset as the largest size of the cluster with significant values resulting
from voxel-by-voxel paired t-test between the single and paired-pulse stimulation. Since
inhibition corresponds to the fact that the power of single pulse is larger than that of the
paired, the analysis was single-tailed [Garcia Dominguez et al., 2014; Radhu et al., 2015].
132
This analysis was repeated 2000 times, selecting a different pool of 19 subjects from the
same group.
Nineteen subjects were chosen to avoid differences in sample size between the groups that
could potentially affect the voxel-by-voxel t-test favoring inhibition in the larger group
because of increased statistical power. Nineteen represents the lowest number of subjects
within the SCZ group. We replicated the same analysis for figure 2 for healthy controls,
OCD patients and their unaffected first-degree relatives. Figures 3 and 4 depict group
inhibition over the time-frequency domain evaluated non- parametrically. In this case,
subsets of subjects were also chosen with replacement for each group with the same
cardinality to maintain comparable statistical power. Figure 3 illustrates healthy subjects,
first-degree relatives of SCZ patients, and SCZ patients. Figure 4 shows healthy subjects,
first-degree relatives of OCD patients and OCD patients.
7.4.3 Assessing Clinical Severity and Effects of Medication Analyses
in Schizophrenia Patients A Spearman's rho correlation analysis was performed between the BPRS (total score) and the
index of frontal inhibition (overall and gamma frequency bands) for each SCZ subject. This
correlation was also conducted with chlorpromazine equivalents [American Psychiatric
Association, 1997; Bezchlibnyk-Butler et al., 2014; Chue et al., 2005; Woods, 2003].
7.4.4 Evaluating Clinical Severity in First-Degree Relatives of
Schizophrenia A Spearman's rho correlation analysis was performed between the SPQ (total score) and the
index of frontal inhibition (overall and gamma frequency bands) for the first-degree relatives
of SCZ group.
7.4.5 Stratification of Age in First-Degree Relatives of Schizophrenia
Patients
133
DLPFC inhibition (overall and gamma frequency bands) was compared between young (<50
years) first-degree relatives of SCZ patients and older (>50 years) first-degree relatives of
SCZ patients to determine whether there were effects of age. An independent samples t-test
was used to compare the two groups.
7.4.6 The Heritability of Inhibition in Schizophrenia Variance components methods in SOLAR v.4.3.1 were used to estimate the narrow sense
heritability for overall inhibition of the DLPFC, defined as the phenotypic variance explained
by additive genetic factors [Almasy and Blangero, 1998]. Normalized trait values were used
for all analyses, and age and sex were screened as covariates and found to be not significant
(p > 0.05). In first-degree relatives of SCZ patients, there were no differences in ethnicity
between Caucasians and non-Caucasians in frontal overall inhibition (p = 0.39) and frontal
gamma inhibition (p = 0.85). In SCZ patients, there were no difference in ethnicity between
Caucasians and non-Caucasians in frontal overall inhibition (p = 0.26) and frontal gamma
inhibition (p = 0.61), thus, ethnicity was not used a covariate. Corrections were made for
ascertainment bias, since the families were recruited through the identification of a proband
with SCZ and are thus not representative of the general population [Beaty and Liang, 1987].
The significance of the heritability estimate was determined by comparing the full polygenic
model with significant covariates included to a sporadic model in which the genetic
component had been removed.
7.5 Results 7.5.1 Frontal Overall (2-50 Hz) Inhibition Frontal inhibition was significantly greater in healthy subjects compared to subjects with
SCZ (t = 1.76, df = 66, p = 0.041). First-degree relatives of SCZ patients showed
significantly more inhibition than their SCZ probands (t = 2.14, df = 47, p = 0.038). No
differences were demonstrated between first-degree relatives of SCZ and healthy subjects
(t = -0.44, df = 77, p = 0.66). Figure 1 shows that the pattern of frontal inhibition. This
analysis showed that the pattern of inhibition was: healthy subjects > first-degree relatives of
SCZ > SCZ probands, over a wide range of p-value thresholds and was independent of the
134
specific threshold chosen. A trending heritability estimate of 0.65 was found with a standard
error of 0.58 and p = 0.107 in overall inhibition. No differences were found between healthy
subjects compared to OCD patients (t = -1.08, df = 60, p = 0.29) and their first-degree
relatives (t = -0.33, df = 65, p = 0.75). Lastly, no inhibition differences were found between
OCD patients and their first-degree relatives (t = -0.64, df = 29, p = 0.52) (figure 2).
7.5.2 Frontal Gamma (30-50 Hz) Inhibition Gamma inhibition was significantly lower in SCZ compared to healthy subjects (t = 2.22, df
= 66, p = 0.015). No significant differences were found between healthy subjects and first-
degree relatives of SCZ (t = 0.16, df = 77, p = 0.87). When comparing first-degree relatives
of SCZ to their probands, no differences in gamma inhibition (t = 1.77, df = 47, p = 0.083)
were found. No significant differences were found between healthy controls compared to
OCD (t = -1.31, df = 60, p = 0.19) and when compared to first-degree relatives of OCD (t = -
0.40, df = 65, p = 0.69). Lastly, no significant differences were found between first-degree
relatives of OCD compared to their OCD probands (t = -0.72, df = 29, p = 0.48) (figures 3
and 4).
135
Figure 19. The index of frontal inhibition from a cluster analysis at different thresholds of p-
values in healthy controls, first-degree relatives of schizophrenia patients and their related
probands. A cluster analysis was performed for each group by sampling a subset 19 of
subjects with replacement. The procedure was repeated 2000 times for each threshold. Error
bars indicate one standard error of the mean.
136
Figure 20.The index of frontal inhibition from a cluster analysis at different thresholds of p-
values in healthy controls, first-degree relatives of obsessive-compulsive disorder patients
and their related probands. A cluster analysis was performed for each group by sampling a
subset 13 of subjects with replacement. The procedure was repeated 2000 times for each
threshold. Error bars indicate one standard error of the mean.
7.5.3 Motor Cortex Overall (2-50 Hz) Inhibition No significant inhibitory differences were found between patients with healthy subjects and
SCZ patients (t = 0.22, df = 66, p = 0.41) and when compared to their first-degree relatives
(t = -0.44, df = 77, p = 0.66). No significant differences were found between SCZ patients
and their first-degree relatives (t = 0.57, df = 47, p = 0.57). No differences were found
between healthy subjects compared to OCD patients (t = -0.29, df = 60, p = 0.78) and when
compared to their first-degree relatives (t = 0.24, df = 65, p = 0.81). No significant
differences were found between OCD patients and their first-degree relatives (t = -0.58, df =
29, p = 0.57).
137
7.5.4 Motor Cortex Gamma (30-50 Hz) Inhibition No significant inhibitory differences were found between patients with healthy subjects and
SCZ patients (t = 0.04, df = 66, p = 0.49) and when compared to their first-degree relatives
(t = 0.22, df = 77, p =0.83). No significant differences were found between SCZ patients and
their first-degree relatives (t = -0.16, df = 47, p = 0.87). No differences were found between
healthy subjects compared to OCD patients (t = 0.15, df = 60, p = 0.88) and when compared
to their first-degree relatives (t = 0.63, df = 65, p = 0.53). No significant differences were
found between OCD patients and their first-degree relatives (t = -0.53, df = 29, p = 0.60).
7.5.5 Stratification of Age in First-Degree Relatives of Schizophrenia
Patients No significant inhibition differences were found when comparing younger (n = 9) (< 50
years) to older (n = 21) (> 50 years) first-degree relatives of SCZ patients in frontal overall
inhibition (p = 0.48) and frontal gamma inhibition (p = 0.66).
7.5.6 Clinical Severity Analysis in First-Degree Relatives of
Schizophrenia Patients In first-degree relatives of SCZ patients, no significant relationship was found between the
SPQ (total score) and frontal overall inhibition (Spearman’s rho = 0.27, p = 0.92). In first-
degree relatives of SCZ, no significant relationship was found between the SPQ (total score)
and frontal gamma inhibition (Spearman’s rho = 0.14, p = 0.76).
7.5.7 Effect of Antipsychotic Medications and Anti-depressant
Medications No significant correlation between overall inhibition and chlorpromazine equivalents was
shown (Spearman’s rho = 0.012, p = 0.52) and no relationship was found between frontal
gamma inhibition and chlorpromazine equivalents (Spearman’s rho = -0.25, p = 0.15).
138
In the DLPFC, no significant differences were found between antidepressant-treated OCD
patients (n = 7) and unmedicated OCD patients (n = 6) in overall inhibition (p = 0.15) and
gamma inhibition (p = 0.32). In the DLPFC, no significant differences were found between
antidepressant-treated SCZ patients (n = 4) and SCZ patients who were not treated with
antidepressants (n = 15) in overall frontal inhibition (p = 0.85) and frontal gamma inhibition
(p = 0.98). Lastly, in the DLPFC, no significant differences were found between OCD
patients who were treated with SSRIs (n = 4) and SCZ patients who were treated with SSRIs
(n = 2) in overall frontal inhibition (p = 0.40) and gamma inhibition (p = 0.31).
7.5.8 Clinical Severity in Schizophrenia Patients No significant relationship was found between the BPRS and overall frontal inhibition,
(Spearman’s rho = -0.28, p = 0.12). A trending negative correlation was found between the
BPRS and frontal gamma inhibition (Spearman’s rho = -0.36, p = 0.068). The greater the
BPRS score is indicative of increased severity of SCZ symptoms. This negative correlation
shows that the higher BPRS score is related to a lower degree of frontal inhibition in 19 SCZ
patients (at trending significance).
139
Figure 21. The frequency of significant values for each group, summarized from subject data,
on each voxel for all the nine central electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2).
The threshold for significance was chosen to be p < 0.01. Each graph corresponds to healthy
subjects, first-degree relatives of SCZ patients, and their related probands. The stimulation
area was the dorsolateral prefrontal cortex. Values are masked over the left bottom area (dark
navy blue) indicating that those windows of the wavelet analysis, which contains points from
the pre-stimulus interval.
140
Figure 22. The frequency of significant values for each group, summarized from subject data,
on each voxel for all the nine central electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2).
The threshold for significance was chosen to be p < 0.01. Each graph corresponds to healthy
subjects, first-degree relatives of obsessive-compulsive disorder patients, and their related
probands. The stimulation area was the dorsolateral prefrontal cortex. Values are masked
over the left bottom area (dark navy blue) indicating that those windows of the wavelet
analysis, which contains points from the pre-stimulus interval.
141
7.6 Discussion We found that first-degree relatives of SCZ patients showed an intermediate pattern of
frontal inhibition compared to their related probands and healthy controls. Significant
differences were found in frontal overall inhibition between SCZ patients and their
unaffected first-degree relatives. We showed a heritability estimate of 0.65 for frontal
inhibition that was trending towards significance. No differences were demonstrated in
frontal inhibition between unaffected first-degree relatives of SCZ and healthy subjects.
Significant deficits were shown in frontal overall and gamma inhibition in patients with SCZ
compared to healthy subjects. No significant frontal inhibition differences were found in
OCD patients when compared to their unaffected first-degree relatives and healthy subjects.
7.6.1 Frontal Inhibition in First-Degree Relatives of Schizophrenia
Probands We demonstrated that first-degree relatives of SCZ had significant differences in overall
frontal inhibition when compared to their related probands. First-degree relatives of SCZ
were not significantly different from healthy subjects. The pattern of frontal inhibition in
first-degree relatives of SCZ was intermediate between their related probands and healthy
controls (figure 1). We showed preliminary evidence suggesting a heritability estimate of
0.65 that was trending and may not have reached signifance due to the relatively small
sample size in our study. Epidemiological studies indicate a heritability of up to 80% for
SCZ, reflecting a strong genetic influence [Cardno et al., 1999b; Sullivan, 2005]. Data from
biological relatives of probands are important for assessing disease-related effects in a
complex disorder like SCZ. The Consortium of Genetics in Schizophrenia (COGS) has
investigated several neurophysiological measures as potential endophenotypes as a means for
understanding the genetic determinants of SCZ [Calkins et al., 2007]. Measures include P50
suppression, antisaccade task for eye movements and prepulse inhibition. Greenwood et al.
demonstrated that P50 suppression shows a low heritability of 0.10 that was not significant in
a sample of 183 nuclear families [Greenwood et al., 2007]. Furthermore, for the antisaccade
task for eye movements, moderate to strong heritability of 0.42 was found [Greenwood et al.,
2007] and a modest heritability of 0.32 for prepulse inhibition was shown [Greenwood et al.,
142
2007]. Hasenkamp et al. demonstrated moderate heritability of 0.45 for prepulse inhibition at
the 60 ms interstimulus interval and a trending heritability of 0.33 at the 120 ms interstimulus
interval [Hasenkamp et al., 2010]. These above mentioned studies suggest low heritability for
P50 suppression, moderate to strong for the antisaccade task for eye movements and modest
for pre-pulse inhibition. Identification of better endophenotypes are needed to facilitate the
future of diagnosis and may also facilitate the identification of genes contributing to SCZ
susceptibility.
7.6.2 Endophenotypes (Intermediate Phenotypes) in Schizophrenia Currently, no objective measures exist to inform psychiatric diagnoses. The clinical interview
dominates the methodological approach with respect to diagnosis, with little evolvement over
the last 100 years. Compared to current subjective clinical diagnoses, endophenotypes (i.e.
intermediate phenotypes) are discrete, genetically determined disease-related intermediate
phenotypes that are detected in the laboratory with demonstrated replicability, diagnostic
specificity, trait stability, test-retest reliability and heritability (first-degree relatives of
probands show intermediate values) [Braff et al., 2007; Braff, 2015; Braff et al., 2008;
Gottesman and Gould, 2003]. Psychiatric disorders are genetically complex, no specific
constellation of genes or environmental conditions characterize a large subset of ill
individuals, thus, requiring the study of intermediate phenotypes [Meyer-Lindenberg and
Weinberger, 2006]. Intermediate phenotypes refer to pathophysiological phenomena,
whereby, susceptability-related phenotypes are between genetics and the neurobiological
systems underlying behavioral disturbance [Tan et al., 2008]; they allow for a target to find
disease-associated genetic variants and elucidation of disease-related mechanisms [Flint et
al., 2014]. For example, patients with SCZ have changes in frontal brain function, cognition
and in brain structure that are found more frequently in their unaffected siblings, including
unaffected monozygotic co-twins, significantly more than in healthy control subjects,
suggesting that these intermediate deviations represent biological expressions of increased
genetic risk [Callicott et al., 2003; Cannon et al., 2000; Ettinger et al., 2007; Toulopoulou et
al., 2007]. In a family study of the association of a SCZ susceptibility gene, DISC1 and the
P300 waveform, almost every subject with a structural abnormality in the DISC1 gene had a
143
significant reduction in the amplitude of the P300 event-related potential, even if they had no
psychiatric diagnosis [Blackwood et al., 2001]. Individuals in the same family lacking the
DISC1 abnormality also lacked a P300 abnormality. This suggests that a gene linked with
behavioral abnormalities is more strongly associated with a measure of brain function related
to SCZ and genetic risk for SCZ even in the absence of clinical presentation. However, most
intermediate phenotypes related to brain physiology are polygenic and heterogeneous [Tan et
al., 2008]. Our findings demonstrate potential for an intermediate phenotype as we found that
first-degree relatives of SCZ were intermediate of probands and healthy controls in frontal
inhibition. These biological changes are found in at-risk individuals who do not manifest a
psychiatric diagnosis suggesting evidence for susceptibility-related phenotypes, intermediate
between genetics and psychopathology [Preston and Weinberger, 2005; Tan et al., 2008].
7.6.3 Frontal Inhibitory Deficits in Schizophrenia The SCZ patients in this study included the results of 13 of the 19 patients that were
published previously [Radhu et al., 2015]. For these current analyses, the probands were
included if their first-degree relatives had been neurophysiologically assessed. Several lines
of evidence support our findings of frontal LICI deficits in patients with SCZ. Previous
studies have showed a reduced density of GABA interneurons in superficial layers of the
prefrontal cortex has been demonstrated which may account for the inhibitory deficits in SCZ
patients. For example, GABAergic deficits have been found in SCZ based on post-mortem
and animal studies showing reduced expression of pre- and postsynaptic markers of
GABAergic neurotransmission in subpopulations of GABAergic interneurons in the
prefrontal cortex [Benes and Berretta, 2001; Coyle, 2004; Lewis et al., 2004]. Recent work
[Marsman et al., 2014] using magnetic resonance spectroscopy (MRS) has found that SCZ
patients had significantly lower GABA/creatine ratios specific to the medial prefrontal
cortex. These findings are consistent with postmortem SCZ studies demonstrating diminished
GABA production based on decreased levels of mRNA encoding for glutamate
decarboxylase67 (GAD67), an enzyme that facilitates GABA synthesis from glutamate
[Kondziella et al., 2007; Lisman et al., 2008; Olney et al., 1999; Stone et al., 2009]. As a
result of the above mentioned findings, excessive excitability in the cortex may result in
144
aberrant neuronal activation that may lead to the disorganized behavior and impulsivity that
is commonly found in SCZ [Uhlhaas et al., 2006; Uhlhaas and Singer, 2010]. Two studies
have demonstrated impaired frontal gamma inhibition specific to SCZ using combined TMS
and EEG [Farzan et al., 2010a; Radhu et al., 2015]. Gamma oscillations appear to be
dependent on inhibitory neurotransmission from parvalbumin-containing GABA
interneurons. The lack of GABAergic neurotransmission in SCZ may translate into excessive
gamma oscillations leading to a pathological plasticity (long-term potentiation) and
ultimately translate into abnormal learning and inflexible thinking that over time may lead to
delusions, a manifestation of erroneous information that is learned and reinforced.
We found no significant frontal inhibitory deficits in patients with OCD compared to their
first-degree relatives and compared to healthy subjects. The results of seven of the OCD
patients had been reported [Radhu et al., 2015] and 6 additional OCD patients were included
in these current analyses. Similar to previous results [Radhu et al., 2015], we found no
differences between OCD patients and healthy controls. Furthermore, the findings that first-
degree relatives of SCZ were intermediate of their probands and healthy controls were
specific to SCZ and their first-degree relatives as this was not demonstrated in OCD and their
first-degree relatives. Future studies are needed to determine the specificity and reliability of
our findings, since there are substantial areas of pathophysiological overlap between SCZ
and OCD [Poyurovsky and Koran, 2005].
7.7 Limitations The results of the present study should be interpreted in light of several limitations. First,
clear neurophysiological and molecular findings in first-degree relatives of SCZ are lacking,
further research needs to be done to disentangle the role of GABA in this population. Second,
when interpreting these results, inferences have been made for the role of LICI vis à vis
cognition, we did not directly measure cognition in this study. Going forward, studies should
establish relationships between LICI, cognition and behavior. Another shortcoming is that
TMS-EEG data is heavily contaminated by artifact. Artifact selection can be very subjective,
making it difficult to decide which ICA component to accurately remove without heavily
145
impacting neuronal activity. Further refinement of removing noise from the EEG data is
necessary. Lastly, this study included a small sample of first-degree relatives. Larger samples
of first-degree relatives are needed including more complete nuclear families such as:
unaffected biological parents, a full sibling and the proband to compute heritability analyses.
With family-based approaches, challenges include recruitment of large samples; further
investigations with larger multi-center research trials are needed to develop TMS-EEG as a
neurophysiological method for use in diagnosis.
7.8 Summarizing the Main Findings
The present results offer preliminary evidence to suggest that first-degree relatives of SCZ,
showed inhibition intermediate of probands and healthy controls. First-degree relatives of
SCZ were significantly different than probands. We showed impairment in frontal inhibition
specific to SCZ patients compared to healthy controls. The ability to evaluate the response
profiles of different oscillatory frequency bands via EEG in response to TMS may be a
potential measure for diagnosis. An advantage of using endophenotypes is that they relate to
specific objective biological functions and substrates associated with the disease. The search
for liability genes for complex disorders such as SCZ may be aided by identifying
endophenotypes and relating these genes to cortical inhibition. Such efforts may ultimately
help to enhance our understanding of the complex neurobiological mechanisms underlying
SCZ to help with diagnosis in the future.
146
Chapter 8
General Discussion, Future Directions and Conclusions
147
8.1 Summary of the Literature Review The introduction of this dissertation is comprised of three literature review chapters. These
chapters summarized TMS and EEG studies assessing inhibition, excitability, plasticity and
connectivity in motor and frontal regions of the brain. Applications of such methods were
applied to psychiatry, neurology, sleep and loss of consciousness research. The first chapter
was a review which summarized motor cortex inhibitory and excitatory findings in patients
with OCD, MDD and SCZ. The literature review focused on neurophysiological studies
which linked dysfunctional GABAergic inhibitory neurotransmission and psychiatric
disorders. The second chapter demonstrated how TMS-EEG is used to directly assess the
DLPFC in psychiatric disorders without being substantially affected by TMS-stimulus
artifact, due to advancements in technology. The third chapter of the introduction assessed
the neurobiology of SCZ and potential candidate endophenotypes for this disorder.
8.1.1 Summary of the Original Research The three main original research studies presented within this dissertation have evaluated
inhibitory measures in severe psychiatric disorders. The first study (chapter 5) provided the
motivation of this current work, hypothesizing an overall inhibitory deficit in severe
psychopathology. This publication quantified all motor cortex inhibitory and excitatory
paradigms with OCD, MDD and SCZ. Our analysis showed that inhibitory deficits were a
ubiquitous finding across OCD, MDD and SCZ and enhancement of excitability (ICF) was
only found in OCD [Radhu et al., 2013]. Specifically, we found significant effect sizes
(Hedge’s G) for decreased SICI, enhanced ICF and reduced CSP within the OCD population.
For MDD, significant effect sizes (Hedge’s G) were found for decreased CSP and SICI.
Lastly, significant deficits in SICI were shown in SCZ. These findings are in line with
previous literature that suggests motor inhibitory deficits among psychiatric disorders;
however, this study suggests that each disease may have a distinct illness profile and
response to treatment.
The second study (chapter 6) evaluated GABAB inhibition in both the DLPFC and motor
cortex using TMS-EEG in SCZ, OCD and healthy controls [Radhu et al., 2015]. The main
148
objective was to evaluate the specificity of frontal inhibition deficits in large sample of SCZ
patients with SCZ. We found deficits in frontal inhibition (overall inhibition and gamma
inhibition) in SCZ patients. No differences were found in the motor cortex between the three
groups. We did not find inhibition abnormalities in OCD patients suggesting frontal
inhibitory deficits may be specific to SCZ, allowing us to differentiate SCZ from OCD.
The last study (chapter 7) assessed both frontal and motor LICI (via TMS-EEG) in first
degree-relatives of SCZ and OCD, their probands and healthy subjects. We found that first-
degree relatives of SCZ were significantly different than their probands; however, they were
similar to healthy controls (no significant differences) in both frontal gamma and overall
inhibition. We showed that first-degree relatives had a pattern of frontal LICI at the
intermediate point between their related SCZ probands and healthy subjects, SCZ were the
most impaired in frontal LICI. Furthermore, no differences were found between OCD, their
relatives and healthy subjects. The heritability of 0.65 (trending) in overall inhibition in SCZ
provides pilot data to suggest that frontal inhibition may represent a candidate
endophenotype for this illness, warranting further investigation. Future multi-site trials are
needed to assess larger samples of first-degree relatives.
8.2 How Can Neuroscience Revolutionize Psychiatry? Neuroscience studies rely on disease definitions that are solely based on the DSM, a
subjective tool used to diagnose psychiatric illnesses. To provide an alternative framework
for research into psychiatric disorders, the United States National Institute of Mental Health
introduced its Research Domain Criteria project to provide a complementary way of
classifying mental illness [Casey et al., 2013]. The main focus of the strategy is to develop
new scientific ways of classifying psychiatric diseases based on behavioural dimensions and
neurobiological measures that is not intended to replace the DSM [Casey et al., 2013]. This
discussion will focus on the need for the biomarker approach, as more research needs to be
conducted with psychiatric populations emphasizing the use of laboratory measures in order
to accelerate both diagnosis and treatment monitoring.
149
8.2.1 Combined TMS and EEG TMS is a non-invasive brain stimulation technique that allows for in vivo examination of
cortical processes and to test both excitatory and inhibitory circuits [Barker et al., 1985]. In
1985, Barker et al. introduced TMS as a tool for investigating the functional state of the
motor pathways in patients with neurological disorders as well as in healthy human subjects
[Barker et al., 1985]. TMS has been used as an investigational tool to measure
neurophysiological processes [McClintock et al., 2011]. TMS has been combined with EEG
to evaluate the effects of electromagnetic induction on cortical oscillations, a methodological
combination that has generated important neurophysiological leads in both healthy and
disease states [Garcia Dominguez et al., 2014; Radhu et al., 2015].
8.2.2 Why Implement TMS-EEG? Many of the studies highlighted in this dissertation demonstrate the tremendous potential for
the recording of TMS-evoked potentials in both motor and non-motor regions of the brain.
As illustrated in the introduction, TMS-EMG studies have shown to be invaluable in
assessing the pathophysiology of neuropsychiatric disorders [Fitzgerald et al., 2002c;
Levinson et al., 2010; Levinson et al., 2007] and the effects of various medications on
different neurotransmitter pathways in the cortex [Ziemann et al., 1998; Ziemann et al.,
1996a; b; Ziemann et al., 1997a; Ziemann et al., 1996c; Ziemann et al., 1997b]. However,
combined TMS and EEG has the potential to extend such findings to frontal brain regions
[Daskalakis et al., 2008c] and to provide evidence about important physiological mechanisms
that are unique to individual brain regions [Paus et al., 2001].
8.3 Advancements in Analyses Recent advancements in post-processing analyses and methods have rendered EEG as a
powerful, cost-effective and easy-to-use technique in clinical and experimental settings. In
agreement with previous studies, we have identified differences between the conditioned and
unconditioned response in the LICI paradigm. However, this time we have analyzed a larger
temporal window. Our findings have shown large differences over roughly 300 ms between
the activities generated by the two conditions. Actual inhibition mediated by the release of
150
GABAB and subsequent generation of inhibitory post-synaptic potentials, can still occur over
a much shorter time scales (i.e., the first 100–150 ms after the test pulse). However, the effect
of such transient inhibition can also resonate over a much larger time scale. The particular
spatial signature of LICI and its evolution in time has been shown to be related to GABAB-
mediated inhibitory neurotransmission.
8.3.1 Detecting and Removing Artifacts in EEG EEG data is dominated by non-physiologic noise (e.g., line noise, electrode, movement, TMS
artifacts) and also contributed by physiological processes (e.g., muscle activity, eye blinks
and saccades) [Rogasch et al., 2014]. Dealing with non-brain artifacts is challenging in
clinical samples. Two rounds of ICA were implemented in the studies presented in this
dissertation to address the noise in the data; the first round was mainly to remove the TMS-
related decay artifact and the second round was to remove eye-blink and muscle-related
artefacts. The second round of ICA was done on the mixed signal after the TMS-related
decay was removed, components were mixed again after the first round of ICA and a new
round of ICA was applied. The ICA does not presume independence of signals. It is a model
that takes dependent signals and produces independent components that explain the same
data. It may happen that the number of components that explain the data is less than the
number of channels. This is the case when the second ICA is applied after the removal of a
few components from the first one. Ideally, one ICA would be enough. The reason we
applied two rounds of ICA was mainly that we wanted to filter the data before ICA was
applied to obtain cleaner components, however this was impossible in some cases because
the large amplitude artifact resulting from the TMS-related decay distorts the output of the
filter. Thus, the final solution was two rounds of ICA [Garcia Dominguez et al., 2014].
8.3.2 Cluster-Based Analyses The statistical analysis described in this dissertation was based on the specific applications of
the cluster-based permutation test, which have produced particularly sensitive results without
assumptions over the null distribution. This non-parametric test is based on random
permutations of conditions across subjects and groups. It corrects for multiple comparisons
151
by reducing the whole time-frequency domain to a single number that accounts for the size of
the main cluster of inhibition. In the context of TMS-EEG, limited studies have examined the
entire time-frequency and spatial domain since presenting a multidimensional analysis
increases the likelihood of committing a Type I error due to the problem of multiple
comparisons [Maris and Oostenveld, 2007]. Taken together, the presented methodology is
parameter-free, while at the same time, avoids the multiple comparison issue without the
need to discard vital information as done in previous TMS-EEG analyses [Garcia Dominguez
et al., 2014].
8.3.3 LICI Analyses Our analysis presents a wider picture of LICI, in which the extent of the estimated inhibition
in the magnitude and time span is larger than previously documented. We noticed that the
size of the period of inhibition is strongly associated to the frequency, which was particularly
broad at lower frequencies. The location and strength of the peak of inhibition not only
depends on the frequency but also the stimulation site. Our results show that inhibition can be
indexed, with high sensitivity, from areas distant to the stimulation site, mainly over central
and contralateral channels. While GABAB-mediated inhibition can be local in nature, its
effect can potentially amplify as we move away from the stimulation site in both time and
space. Thus, indexing inhibition from distant sites can offer a more sensitive characterization
of the phenomenon. The electrodes closer to the site of stimulation are more affected by coil-
induced artifacts, and thus the least reliable. This could conceivably help to explain why
inhibition was not particularly strong over the area of stimulation. Additionally, distant
inhibition has been reported as interhemispheric inhibition and is mechanistically similar to
LICI [Daskalakis et al., 2002b]. In summary, the advancements in pre- and post-processing
EEG techniques applied in these neurophysiological experiments may serve as a road map
for investigators to model their data processing pipelines.
152
8.4 Pathophysiology of Schizophrenia SCZ is one of the most severe neuropsychiatric disorders, the diagnosis remains symptom-
based, not etiologically based. Studies have previously reported that patients with SCZ show
gamma inhibition deficits in the DLPFC [Farzan et al., 2010a; Radhu et al., 2015]. This
imbalanced circuitry is evident from previous studies in the context of SCZ pathophysiology.
We have shown the specificity of a novel combined TMS-EEG index of GABAB inhibitory
neurotransmission [Radhu et al., 2015]. SCZ patients exhibited LICI deficits in the DLPFC,
and no differences were shown between OCD and healthy subjects. Interestingly, no
differences were observed among the 3 groups in motor cortex, indicating that LICI deficits
in DLPFC were specific to SCZ. This work may represent critical biomarker validation that
can be used to inform diagnostic decisions.
8.4.1 GABAergic Deficits in Schizophrenia Impairments of several neurotransmission systems such as dopamine, glutamate, serotonin,
and GABAergic systems are separately linked to the pathophysiology of SCZ.
Pharmacological studies have examined the neurochemical basis of several TMS paradigms
by examining the effect of dopamine, glutamate and GABA agents on the extent of inhibition
or excitation following the application of a specific TMS paradigm to the motor cortex of
healthy subjects. The results of such studies, as reviewed by Paulus et al. [Paulus et al.,
2008a] indicate that inhibition observed through LICI is likely associated with activation of
GABAB receptors. Previous evidence has demonstrated an association between GABAergic
inhibitory neurotransmission and gamma oscillations [Bartos et al., 2007; Bragin et al., 1995;
Brown et al., 2007; Jefferys et al., 1996; Marrosu et al., 2006; Scanziani, 2000; Traub et al.,
1997; Traub et al., 1996; Wang and Buzsaki, 1996; Whittington et al., 1995]. Furthermore,
the modulation of gamma oscillations represent an important neurophysiological process that
may, in part, be responsible for optimal cognitive functioning in the DLPFC. It has been
postulated that the functional role of gamma modulation may be to provide a temporal frame
for information processing and filtering out information. For example, it has been shown that
the successful encoding of information may depend on its arrival time relative to the gamma
cycle. Information arriving at the fading phase of inhibition would be potentiated, whereas
153
information arriving at the beginning of the inhibitory period will be blocked from further
propagation, likely through LTP-like mechanisms (i.e., NMDA receptor-mediated
neurotransmission). Gamma band abnormalities may correspond to a down-regulation of
gamma due to the aberration in the ability of the neural circuits to support this critical
frequency range. Disturbances in chandelier cell functioning could impair the ability of
cortical circuits to engage in high frequency synchronous oscillations [Lewis et al., 2005], as
a result, disrupted LICI may result from disordered synaptic wiring in key cognitive
networks. CI aids suppression of neural noise by filtering irrelevant sensory information
imperative for attention and cognitive performance Thus, gamma oscillations represent an
important neurophysiological process that may, in part, be responsible for optimal cognitive
function and may explain why their functioning (i.e., generation and modulation) is largely
localized to the DLPFC [Farzan et al., 2009], shown to be dysfunctional in SCZ.
Disrupted gamma oscillations in the DLPFC as indexed by EEG have been consistently
demonstrated in SCZ patients [Basar-Eroglu et al., 2007; Cho et al., 2006; Uhlhaas and
Singer, 2010]. For example, it has been shown that patients with SCZ have deficits in gamma
oscillatory activity in response to 40 Hz auditory stimulation [Light et al., 2006] or during
perception of gestalt objects [Spencer et al., 2003] as compared to healthy individuals, while
other studies have shown an impairment in modulation of gamma oscillations in the DLPFC
during working memory tasks [Cho et al., 2006]. Also, both increase and reduction of gamma
oscillations have been reported in patients with SCZ [Ferrarelli et al., 2008; Lee et al., 2003].
We have specifically demonstrated that patients with SCZ exhibit excessive power of gamma
oscillations during working memory performance [Barr et al., 2010]. Barr et al. examined
frontal gamma oscillations using EEG in SCZ and healthy subjects during working memory
performance at three different working memory loads. It was shown that SCZ patients
performed worse than healthy subjects, and generated excessive power of gamma oscillations
from the frontal local electrodes only most notably during the task with the highest working
memory load, the 3-back task condition [Barr et al., 2010]. It has been suggested that the 3-
back may involve attentional components and/or short term memory aspects of working
memory. In this view, working memory is regarded as controlled attention and been shown to
modulate with gamma oscillatory activity. Thus, it is possible that the excessive power of
154
gamma oscillations was associated with the impairment in the modulation of gamma
oscillations. To ascertain these findings further, future studies should examine the correlation
between frontal inhibitory deficits, attentional processing and working memory performance
in SCZ patients.
8.5 Pathophysiology of OCD OCD is a debilitating psychiatric illness that has been reported to affect 1-3% of the world’s
population [Horwath and Weissman, 2000; Torres et al., 2006]. OCD consists mainly of
obsessions (i.e., persistent, intrusive thoughts or impulses that potentiate anxiety) and
compulsions (i.e., repetitive ritualistic physical or mental actions performed to reduce
obsession-provoked anxiety) [Abramowitz et al., 2009; Heyman et al., 2006; Stein, 2002].
Studies show that the rate of co-morbidity between SCZ and OCD is approximately 7% to
26% [Eisen et al., 1997; Fabisch et al., 1997; Porto et al., 1997; Poyurovsky et al., 1999;
Tibbo et al., 2000].
Several genetic studies have reported associations between OCD and dysfunctional
GABAergic and glutamatergic genes [Arnold et al., 2006; Dickel et al., 2006; Samuels et al.,
2011; Stewart et al., 2007; Voyiaziakis et al., 2011; Zai et al., 2005]. Arnold and colleagues
[Arnold et al., 2004] found a positive association between variants in the 3′ untranslated
region of the GRIN2B gene— the gene encoding the NR2 subunit of the N-methyl-D-
aspartate (NMDA) glutamate receptor and OCD in 178 affected individuals from 130
families. Similarly, Whiteside et al., demonstrated increased levels of a combined measure of
glutamate and glutamine relative to creatine were found in orbitofrontal white matter in
patients with OCD [Whiteside et al., 2006]. Furthermore, Chakrabarty et al., showed
significantly higher levels of glutamate in OCD [Chakrabarty et al., 2005]. Animal models
confirm the role of corticolimbic glutamatergic hyperactivation in patients with OCD
[Nordstrom and Burton, 2002]. Zai et al., found a positive association between OCD and the
GABAB receptor gene (GABR1) [Zai et al., 2005], implicating a relationship between
dysfunctional GABAB and the pathophysiology of OCD. TMS studies with OCD patients
have demonstrated decreased inhibition [Greenberg et al., 2000; Greenberg et al., 1998;
155
Richter et al., 2012] and enhanced cortical excitability [Richter et al., 2012]. Richter et al.
reported [Richter et al., 2012] that patients with OCD have abnormalities in both GABAB and
NMDA receptor-mediated neurotransmission. Deficits were found in inhibition and
excessive intracortical facilitation of the motor cortex, a paradigm reflecting excessive
NMDA-receptor-mediated excitatory neurotransmission, independent of medication status.
Collectively these findings are consistent with genetic findings reporting GABA and NMDA-
related genes involved in the pathophysiology of OCD [Arnold et al., 2006; Dickel et al.,
2006; Samuels et al., 2011; Stewart et al., 2007; Voyiaziakis et al., 2011; Zai et al., 2005].
However, motor cortex TMS studies are of limited interest as the pathophysiology of many
psychiatric disorders are more closely associated with frontal brain abnormalities. Therefore,
it is essential to evaluate the neurophysiology in brain regions that are more proximal to the
underlying phenotype such as the DLPFC.
As demonstrated in this dissertation, we did not find frontal inhibitory deficits in OCD
patients and their unaffected first-degree relatives. Our findings could be accounted for by
the large medicated OCD sample, as the recent Richter et al., study included a majority of
unmedicated patients with OCD (68%) [Richter et al., 2012]. SSRIs inhibitors are the
established pharmacologic first-line treatment for OCD [Decloedt and Stein, 2010; Kellner,
2010]. Serotonin is able to modulate excitatory and inhibitory effects, respectively mediated
by glutamate and GABA [Ciranna, 2006]. The serotonin receptor (5-HT) induces a decrease
of glutamate transmission and a parallel increase in GABA transmission evident in the
hippocoampus, frontal cortex and the cerebellum [Ciranna, 2006]. Previous studies have
shown that SSRI’s increase GABA measured by magnetic resonance spectroscopy
[Bhagwagar et al., 2004] and TMS [Robol et al., 2004], thus concealing any potential LICI
deficits in the present study. The modulatory action of the serotonin receptor (5-HT) may
serve as a "brake" on neuronal excitability. Given this inconsistency, further studies are
warranted to disentangle the effects of medication. Future directions of this work may be to
evaluate LICI (TMS-EEG) before and after SSRI treatment for OCD to establish a
relationship between inhibition and therapeutic response with larger samples of patients.
156
8.6 Neurophysiological Biomarkers in Clinical Practice The identification of biological markers for heterogeneous disorders such as SCZ has several
potential benefits. First, there is a great need for developing a biological marker in psychiatry
as no laboratory tests have entered the clinic to inform diagnoses as well as to guide and
monitor treatments. Clinicians still rely on behavioural observation and careful interview
techniques to make inferences about patients’ inner experiences and make assumptions about
the impacted neural systems. Although we have refined indirect clinical assessments for
diagnosis and treatment, these methods have only slightly evolved over the past 100 years
[Light and Makeig, 2015]. Biological identification may also be used for diagnosing
individuals who are at high risk for developing SCZ, allowing for an early intervention. Since
SCZ is a very heterogeneous disorder across a variety of domains including: symptoms,
functional outcome, genetic architecture, and pathophysiology, mixed research data arise as a
result of this heterogeneity, making biomarker work extremely difficult [Light and Makeig,
2015].
The translation from neuroscience research to clinical practice has been challenging,
endophenotypes provide a way for that to occur. Endophenotypes are heritable biomarkers of
psychiatric disorders. To be considered as an endophenotype, a biomarker must fulfill criteria
such as: 1) heritability (i.e., measure of the strength of genetic effects on a trait) 2) trait
stability (e.g., a biomarker that is unrelated to illness duration or pharmacological
treatments); 3) test–retest reliability; 4) diagnostic specificity (i.e., a biomarker is present in
the disease of interest but not present is other disorders); and 5) large effect size differences
between healthy controls and patients [Gottesman and Gould, 2003]. Neurophysiological
tools have potential for providing information on diagnoses, treatment predictors and
treatment monitoring. More research is required to ensure their effective application in
clinical settings. The ability to evaluate physiological response profiles of different
oscillatory frequencies in response to TMS-induced cortical evoked potentials may ultimately
serve to identify endophenotypes or biomarkers for the identification of a variety of
neurological and psychiatric disorders. Much more work is needed to effectively use TMS-
EEG biomarkers as a diagnostic or treatment tool.
157
8.6.1 Assessing Biological First-Degree Relatives of Patients
As discussed and emphasized in chapter 7, endophenotypes are a subtype of biological
markers that are independent of illness state and occur with a higher probability in biological
relatives of patients with SCZ. As such, endophenotypes (also known as intermediate
phenotypes) may be used to distinguish biological relatives who are at risk of developing
SCZ prior to illness onset. It has been proposed that an efficient preventive strategy may
involve using laboratory tests to assess children of parents affected with SCZ, and then
provide preventive strategies for children whereby impairments are observed and are at
greater risk for developing the illness later in life compared to their siblings
Estimation of genetic influence on neurophysiological responses necessitates measures of
heritability, which is ideally achieved with twin studies or large samples of family cohorts.
The true test for a biomarker is that the genetic association should be seen within the
intermediate phenotype in individuals who do not have the clinical diagnosis [Tan et al.,
2008]. Our sample utilized nuclear families and families had to consist of at least one relative
per proband (at minimum) and two parents and one sibling (at maximum). Data from
biological relatives of probands are important tools for assessing the etiology of psychiatric
illnesses. First-degree relatives allow for assessment of intermediate phenotypes which refer
to pathophysiological phenomena, whereby, susceptability-related phenotypes are between
genetics and the neurobiological systems underlying behavioral disturbance [Tan et al.,
2008]. Specifically, they allow for a target to find disease-associated genetic variants and
elucidation of disease-related mechanisms [Flint et al., 2014]. Biological markers and in
particular endophenotypes could be used for early identification, etiological diagnosis, and
selection and discovery of optimal treatment strategies for each patient. Intermediate
phenotypes integrate basic neurobiology with specific human phenotypes that are potentially
tractable genetically [Tan et al., 2008].
In this dissertation, we have showed a reliable approach for indexing the inhibition of cortical
oscillations in vivo in humans. Our technique in combination with neuroimaging and
158
neurocognitive measures may provide a valuable means for in vivo identification of
biological markers for heterogeneous and complex psychiatric disorders such as SCZ.
8.7 Limitations The experiments outlined in this PhD dissertation are a new beginning for a series of future
experiments. However, there are several limitations to this work, discussed in greater detail
within this section.
First, the functional and regional specificity of modulation of cortical oscillations remains
unsolved. The modulatory pattern of cortical oscillations in other cortices, such as in
occipital, temporal, and parietal cortex should be examined in future experiments. Therefore,
without further evidence, it is not possible to infer whether or not gamma inhibition or
overall inhibition is specific to DLPFC. Similarly, to examine the functional significance of
gamma and overall inhibition, the relationship between LICI and performance on various
domains of cognitive function such as working memory should be investigated. In order to
better link neurophysiological findings to the symptoms of SCZ and OCD, future studies
should establish better relationships between LICI, cognition and behaviour.
Second, this study included a pilot sample of unaffected first-degree relatives and their
probands. Replication studies using a multi-site trial approach assessing larger samples of
probands and their relatives are needed. For example, the COGS is a multi-site National
Institute of Mental Health sponsored collaboration investigating the genetic basis of
candidate endophenotypes in SCZ patients and their relatives. Findings from the COGS
family-based study confirmed the presence of robust deficits in SCZ probands and their first-
degree relatives in several different neurophysiological measures, including P50 suppression
[Olincy et al., 2010], N100 evoked amplitude [Turetsky et al., 2008] and antisaccade
performance [Radant et al., 2010]. The COGS strategy has been used to acquire
endophenotype measures across multiple geographically distributed sites to maximize sample
ascertainment [Calkins et al., 2007]. Larger samples of first-degree relatives are needed in
future studies to further assess the heritability of frontal inhibitory deficits in SCZ. Larger
159
sample sizes which include both biological parents and one full sibling are required to
compute heritability analyses, however, with family-based approaches; challenges include
recruitment of large samples. A sample size calculation was completed, to achieve a
heritability estimate of 0.60 with 80% power, 44 families would be required (1 affected
proband and two biological parents) or 17 families would be required (1 affected proband, 1
full sibling and two biological parents) [Schork and Schork, 1993]. To achieve a heritability
estimate of 0.60 with 90% power, 58 families would be required (1 affected proband and two
biological parents) or 23 families would be required (1 affected proband, 1 full sibling and
two biological parents) [Schork and Schork, 1993]. Further investigation with larger multi-
center research trials are needed to develop TMS as a neurophysiological marker for use in
diagnosis and for monitoring treatment outcomes.
Third, frontal inhibition was assessed in SCZ and OCD using a cross-sectional design. Future
studies can consider evaluating patients over time (i.e. before and after starting treatment) to
assess the effects of psychotropic medications, to establish a relationship between inhibition
and therapeutic response.
Lastly, a main shortcoming of TMS-EEG experiments is the contamination of brain signals
by various sources of artifact. As indiciated in the introduction, combined TMS and EEG
measures are susceptible to unwanted physiological and non-physiological artifacts,
particularly over regions such as the DLPFC. Somatosensory-evoked potentials resulting
from either scalp or peripheral TMS-evoked muscle activity or from stimulation of the
trigeminal nerve may also contribute to the noise in the EEG signal. Muscle activity, eye
movement, eye blink and TMS-related decay artifacts can be removed with minimal impact
on neural activity using ICA, allowing the study of TMS-evoked cortical network properties.
However, artifact selection can be very subjective and time consuming, making it difficult to
decide which component to accurately remove without removing TMS-evoked neuronal
activity [Rogasch et al., 2014].
160
8.8 Future Directions There is significant potential in the future to evaluate a variety of other neurophysiological
processes in the cortex. Future studies may also permit the recording of plasticity in non-
motor brain regions. For example, 30 min of repeated stimulation of the median nerve
applied simultaneously with TMS to the motor cortex results in LTP in the motor cortex
through a paradigm known as PAS [Stefan et al., 2000]. Such plasticity is LTP-like as
repeated and cotemporaneous excitation of sensory afferents and motor interneurons
translates into increased motor excitability. These and other plasticity measures have been
previously shown to be impaired in SCZ [Daskalakis et al., 2008a; Frantseva et al., 2008].
Thus, combining TMS and EEG with PAS can be used to index LTP in the DLPFC, and it
can provide critical advantages when attempting to understand key brain mechanisms
underlying learning and working memory. Future studies may also be used to examine
potential regional pharmacological effects that may be of particular importance to illnesses
whose pathophysiology may be more regionally specific (e.g., major depressive disorder).
Future studies may include twin studies, assessing both monozygotic and dizygotic twins to
further disentangle the heritability of inhibition. Furthermore, assessing genetic mutations
and correlating liability genes with LICI would be very important in future trials. In
summary, further studies are warranted to validate our findings with larger multi-site trials.
This dissertation expands and develops further the initial reports of LICI by our group using
TMS and EEG. TMS can be combined with EEG to assess neurophysiological profiles of
response; current studies have demonstrated several neurophysiological processes including
excitability, inhibition, and interhemispheric signal propagation. We consider the pre-
processing tools and our statistical approach to quantify LICI to be accepted as a future
guideline in this field. Within this line of research, the characterization of LICI that we have
presented here can be used to enhance the pool of parameters that characterize inhibition.
The novel analyses presented in this dissertation have great potential to detect network
differences between healthy and disease states and can be applied in future biomarker work.
161
8.9 Conclusions The ability to evaluate cortical processes such as inhibition, excitation and plasticity in
healthy subjects has further led to the discovery of pathophysiological processes in various
neuropsychiatric disorders. This doctoral work consisted of a series of experiments that
applied TMS to examine the modulation of cortical oscillations whose impairments may be
involved in pathophysiology SCZ. Intermediate phenotypes are vital as they link genetic
variation to complex disease mechanisms. Due to the heterogeneity of SCZ and the above
mentioned limitations, it may be too early to conclude that frontal inhibition may be a central
endophenotype for this disease.
Overall, this dissertation has demonstrated inhibitory deficits in the motor cortex in SCZ in
the meta-analysis. We then showed specific frontal inhibitory deficits in SCZ, differentiating
SCZ from OCD. Lastly, we demonstrated first-degree relatives were intermediate of their
related SCZ probands and healthy controls in frontal inhibition with a trending heritability of
0.65. The use of combined TMS and EEG is a tool that needs to be further explored to apply
to clinical practice and may help to explain the variance in disease mechanisms. Our findings
provide rationale for future biomarker research to further ascertain the role of frontal
inhibitory impairments in SCZ.
162
References or Bibliography
Abarbanel JM, Lemberg T, Yaroslavski U, Grisaru N, Belmaker RH (1996)
Electrophysiological responses to transcranial magnetic stimulation in depression and
schizophrenia. Biol Psychiatry 40:148-150.
Abramowitz JS, Taylor S, McKay D (2009) Obsessive-compulsive disorder. Lancet 374:491-
499.
Acosta GB, Rubio MC (1994) GABAA receptors mediate the changes produced by stress on
GABA function and locomotor activity. Neurosci Lett 176:29-31.
Akbarian S, Kim JJ, Potkin SG, Hagman JO, Tafazzoli A, Bunney WE, Jr., Jones EG (1995)
Gene expression for glutamic acid decarboxylase is reduced without loss of neurons
in prefrontal cortex of schizophrenics. Arch Gen Psychiatry 52:258-266.
Akerman CJ, Cline HT (2007) Refining the roles of GABAergic signaling during neural
circuit formation. Trends Neurosci 30:382-389.
Almasy L, Blangero J (1998) Multipoint quantitative-trait linkage analysis in general
pedigrees. Am J Hum Genet 62:1198-1211.
Amador N, Schlag-Rey M, Schlag J (1998) Primate antisaccades. I. Behavioral
characteristics. J Neurophysiol 80:1775-1786.
American Psychiatric Association (1997) Practice guideline for the treatment of patients with
schizophrenia. Am J Psychiatry 154:1-63.
American Psychiatric Association (2000) Diagnostic and statistical manual of mental
disorders (4th ed., Text Revision). Washington, D.C.: American Psychiatric
Association.
Anokhin AP, Vedeniapin AB, Heath AC, Korzyukov O, Boutros NN (2007) Genetic and
environmental influences on sensory gating of mid-latency auditory evoked
responses: a twin study. Schizophr Res 89:312-319.
Arnold PD, Rosenberg DR, Mundo E, Tharmalingam S, Kennedy JL, Richter MA (2004)
Association of a glutamate (NMDA) subunit receptor gene (GRIN2B) with obsessive-
compulsive disorder: a preliminary study. Psychopharmacology (Berl) 174:530-538.
163
Arnold PD, Sicard T, Burroughs E, Richter MA, Kennedy JL (2006) Glutamate transporter
gene SLC1A1 associated with obsessive-compulsive disorder. Arch Gen Psychiatry
63:769-776.
Arnold SJ, Ivleva EI, Gopal TA, Reddy AP, Jeon-Slaughter H, Sacco CB, Francis AN,
Tandon N, Bidesi AS, Witte B, Poudyal G, Pearlson GD, Sweeney JA, Clementz BA,
Keshavan MS, Tamminga CA (2015) Hippocampal volume is reduced in
schizophrenia and schizoaffective disorder but not in psychotic bipolar I disorder
demonstrated by both manual tracing and automated parcellation (FreeSurfer).
Schizophr Bull 41:233-249.
Babiloni C, Binetti G, Cassetta E, Cerboneschi D, Dal Forno G, Del Percio C, Ferreri F, Ferri
R, Lanuzza B, Miniussi C, Moretti DV, Nobili F, Pascual-Marqui RD, Rodriguez G,
Romani GL, Salinari S, Tecchio F, Vitali P, Zanetti O, Zappasodi F, Rossini PM
(2004) Mapping distributed sources of cortical rhythms in mild Alzheimer's disease.
A multicentric EEG study. Neuroimage 22:57-67.
Babiloni C, Cassetta E, Dal Forno G, Del Percio C, Ferreri F, Ferri R, Lanuzza B, Miniussi
C, Moretti DV, Nobili F, Pascual-Marqui RD, Rodriguez G, Luca Romani G, Salinari
S, Zanetti O, Rossini PM (2006) Donepezil effects on sources of cortical rhythms in
mild Alzheimer's disease: Responders vs. Non-Responders. Neuroimage 31:1650-
1665.
Babiloni C, Vecchio F, Lizio R, Ferri R, Rodriguez G, Marzano N, Frisoni GB, Rossini PM
(2011) Resting state cortical rhythms in mild cognitive impairment and Alzheimer's
disease: electroencephalographic evidence. J Alzheimers Dis 26 Suppl 3:201-214.
Baddeley A (1986) Working memory. Oxford: Claredon Press.
Baddeley A (1992) Working memory. Science 255:556-559.
Bajbouj M, Gallinat J, Niehaus L, Lang UE, Roricht S, Meyer BU (2004) Abnormalities of
inhibitory neuronal mechanisms in the motor cortex of patients with schizophrenia.
Pharmacopsychiatry 37:74-80.
Bajbouj M, Lang UE, Niehaus L, Hellen FE, Heuser I, Neu P (2006a) Effects of right
unilateral electroconvulsive therapy on motor cortical excitability in depressive
patients. J Psychiatr Res 40:322-327.
164
Bajbouj M, Lisanby SH, Lang UE, Danker-Hopfe H, Heuser I, Neu P (2006b) Evidence for
impaired cortical inhibition in patients with unipolar major depression. Biol
Psychiatry 59:395-400.
Barbey AK, Koenigs M, Grafman J (2012) Dorsolateral prefrontal contributions to human
working memory. Cortex 49:1195-1205.
Barker AT, Jalinous R, Freeston IL (1985) Non-invasive magnetic stimulation of human
motor cortex. Lancet 1:1106-1107.
Barr MS, Farzan F, Arenovich T, Chen R, Fitzgerald PB, Daskalakis ZJ (2011) The effect of
repetitive transcranial magnetic stimulation on gamma oscillatory activity in
schizophrenia. PLoS One 6:e22627.
Barr MS, Farzan F, Rusjan PM, Chen R, Fitzgerald PB, Daskalakis ZJ (2009) Potentiation of
gamma oscillatory activity through repetitive transcranial magnetic stimulation of the
dorsolateral prefrontal cortex. Neuropsychopharmacology 34:2359-2367.
Barr MS, Farzan F, Tran LC, Chen R, Fitzgerald PB, Daskalakis ZJ (2010) Evidence for
excessive frontal evoked gamma oscillatory activity in schizophrenia during working
memory. Schizophr Res 121:146-152.
Barrantes-Vidal N, Aguilera M, Campanera S, Fatjo-Vilas M, Guitart M, Miret S, Valero S,
Fananas L (2007) Working memory in siblings of schizophrenia patients. Schizophr
Res 95:70-75.
Bartos M, Vida I, Jonas P (2007) Synaptic mechanisms of synchronized gamma oscillations
in inhibitory interneuron networks. Nat Rev Neurosci 8:45-56.
Basar-Eroglu C, Brand A, Hildebrandt H, Karolina Kedzior K, Mathes B, Schmiedt C (2007)
Working memory related gamma oscillations in schizophrenia patients. Int J
Psychophysiol 64:39-45.
Basar-Eroglu C, Struber D, Schurmann M, Stadler M, Basar E (1996) Gamma-band
responses in the brain: a short review of psychophysiological correlates and
functional significance. Int J Psychophysiol 24:101-112.
Beaty TH, Liang KY (1987) Robust inference for variance components models in families
ascertained through probands: I. Conditioning on proband's phenotype. Genet
Epidemiol 4:203-210.
165
Benazzi F (2007) Bipolar disorder--focus on bipolar II disorder and mixed depression.
Lancet 369:935-945.
Benes FM, Berretta S (2001) GABAergic interneurons: implications for understanding
schizophrenia and bipolar disorder. Neuropsychopharmacology 25:1-27.
Benes FM, Kwok EW, Vincent SL, Todtenkopf MS (1998) A reduction of nonpyramidal
cells in sector CA2 of schizophrenics and manic depressives. Biol Psychiatry 44:88-
97.
Benes FM, Lim B, Matzilevich D, Walsh JP, Subburaju S, Minns M (2007) Regulation of the
GABA cell phenotype in hippocampus of schizophrenics and bipolars. Proc Natl
Acad Sci U S A 104:10164-10169.
Benes FM, McSparren J, Bird ED, SanGiovanni JP, Vincent SL (1991) Deficits in small
interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective
patients. Arch Gen Psychiatry 48:996-1001.
Benes FM, Todtenkopf MS, Taylor JB (1997) Differential distribution of tyrosine
hydroxylase fibers on small and large neurons in layer II of anterior cingulate cortex
of schizophrenic brain. Synapse 25:80-92.
Berlucci G (1990) Commisurotomy studies in animals. In: Handbook of Neuropsychology
(Boller F, Grafman J, eds), vol 4, pp 9-47. Amsterdam: Elsevier.
Bezchlibnyk-Butler KZ, Jeffries JJ, Procyshyn RM, Virani AS (2014) Clinical Handbook of
Psychotropic Drugs, ed 20th Edition: Hogrefe Publishing.
Bhagwagar Z, Wylezinska M, Taylor M, Jezzard P, Matthews PM, Cowen PJ (2004)
Increased Brain GABA Concentrations Following Acute Administration of a
Selective Serotonin Reuptake Inhibitor. Am J Psychiatry 161:368-370.
Blackwood DH, Fordyce A, Walker MT, St Clair DM, Porteous DJ, Muir WJ (2001)
Schizophrenia and affective disorders--cosegregation with a translocation at
chromosome 1q42 that directly disrupts brain-expressed genes: clinical and P300
findings in a family. Am J Hum Genet 69:428-433.
Bloom JS, Hynd GW (2005) The role of the corpus callosum in interhemispheric transfer of
information: excitation or inhibition? Neuropsychol Rev 15:59-71.
166
Blumenfeld LD, Clementz BA (2001) Response to the first stimulus determines reduced
auditory evoked response suppression in schizophrenia: single trials analysis using
MEG. Clin Neurophysiol 112:1650-1659.
Bollen KA, Jackman RW (1990) Regression diagnostics: An expository treatment of outliers
and influential cases. In: Modern Methods of Data Analysis (Fox J, Long JS, eds), pp
257-291. Newbury Park, California: Sage Publications.
Bonato C, Miniussi C, Rossini PM (2006) Transcranial magnetic stimulation and cortical
evoked potentials: a TMS/EEG co-registration study. Clin Neurophysiol 117:1699-
1707.
Bonnard M, Spieser L, Meziane HB, de Graaf JB, Pailhous J (2009) Prior intention can
locally tune inhibitory processes in the primary motor cortex: direct evidence from
combined TMS-EEG. Eur J Neurosci 30:913-923.
Boroojerdi B, Topper R, Foltys H, Meincke U (1999) Transcallosal inhibition and motor
conduction studies in patients with schizophrenia using transcranial magnetic
stimulation. Br J Psychiatry 175:375-379.
Boutros NN, Overall J, Zouridakis G (1991) Test-retest reliability of the P50 mid-latency
auditory evoked response. Psychiatry Res 39:181-192.
Braff D, Schork NJ, Gottesman, II (2007) Endophenotyping schizophrenia. Am J Psychiatry
164:705-707.
Braff DL (2015) The importance of endophenotypes in schizophrenia research. Schizophr
Res 163:1-8.
Braff DL, Geyer MA, Swerdlow NR (2001) Human studies of prepulse inhibition of startle:
normal subjects, patient groups, and pharmacological studies. Psychopharmacology
(Berl) 156:234-258.
Braff DL, Greenwood TA, Swerdlow NR, Light GA, Schork NJ (2008) Advances in
endophenotyping schizophrenia. World Psychiatry 7:11-18.
Braff DL, Grillon C, Geyer MA (1992) Gating and habituation of the startle reflex in
schizophrenic patients. Arch Gen Psychiatry 49:206-215.
Braff DL, Light GA (2004) Preattentional and attentional cognitive deficits as targets for
treating schizophrenia. Psychopharmacology (Berl) 174:75-85.
167
Bragin A, Jando G, Nadasdy Z, Hetke J, Wise K, Buzsaki G (1995) Gamma (40-100 Hz)
oscillation in the hippocampus of the behaving rat. J Neurosci 15:47-60.
Brenner CA, Kieffaber PD, Clementz BA, Johannesen JK, Shekhar A, O'Donnell BF,
Hetrick WP (2009) Event-related potential abnormalities in schizophrenia: a failure to
"gate in" salient information? Schizophr Res 113:332-338.
Brockhaus-Dumke A, Schultze-Lutter F, Mueller R, Tendolkar I, Bechdolf A, Pukrop R,
Klosterkoetter J, Ruhrmann S (2008) Sensory gating in schizophrenia: P50 and N100
gating in antipsychotic-free subjects at risk, first-episode, and chronic patients. Biol
Psychiatry 64:376-384.
Brown JT, Davies CH, Randall AD (2007) Synaptic activation of GABA(B) receptors
regulates neuronal network activity and entrainment. Eur J Neurosci 25:2982-2990.
Butefisch CM, Davis BC, Wise SP, Sawaki L, Kopylev L, Classen J, Cohen LG (2000)
Mechanisms of use-dependent plasticity in the human motor cortex. Proc Natl Acad
Sci U S A 97:3661-3665.
Buzsaki G (2006) Rhythms of the Brain. New York, NY: Oxford University Press.
Calkins ME, Dobie DJ, Cadenhead KS, Olincy A, Freedman R, Green MF, Greenwood TA,
Gur RE, Gur RC, Light GA, Mintz J, Nuechterlein KH, Radant AD, Schork NJ,
Seidman LJ, Siever LJ, Silverman JM, Stone WS, Swerdlow NR, Tsuang DW,
Tsuang MT, Turetsky BI, Braff DL (2007) The Consortium on the Genetics of
Endophenotypes in Schizophrenia: model recruitment, assessment, and
endophenotyping methods for a multisite collaboration. Schizophr Bull 33:33-48.
Callicott JH, Egan MF, Mattay VS, Bertolino A, Bone AD, Verchinksi B, Weinberger DR
(2003) Abnormal fMRI response of the dorsolateral prefrontal cortex in cognitively
intact siblings of patients with schizophrenia. Am J Psychiatry 160:709-719.
Campanella S, Guerit JM (2009) How clinical neurophysiology may contribute to the
understanding of a psychiatric disease such as schizophrenia. Neurophysiol Clin
39:31-39.
Cannon TD, Huttunen MO, Lonnqvist J, Tuulio-Henriksson A, Pirkola T, Glahn D,
Finkelstein J, Hietanen M, Kaprio J, Koskenvuo M (2000) The inheritance of
neuropsychological dysfunction in twins discordant for schizophrenia. Am J Hum
Genet 67:369-382.
168
Cantello R, Gianelli M, Civardi C, Mutani R (1992) Magnetic brain stimulation: the silent
period after the motor evoked potential. Neurology 42:1951-1959.
Cardno AG, Jones LA, Murphy KC, Sanders RD, Asherson P, Owen MJ, McGuffin P
(1999a) Dimensions of psychosis in affected sibling pairs. Schizophr Bull 25:841-
850.
Cardno AG, Marshall EJ, Coid B, Macdonald AM, Ribchester TR, Davies NJ, Venturi P,
Jones LA, Lewis SW, Sham PC, Gottesman, II, Farmer AE, McGuffin P, Reveley
AM, Murray RM (1999b) Heritability estimates for psychotic disorders: the
Maudsley twin psychosis series. Arch Gen Psychiatry 56:162-168.
Casarotto S, Canali P, Rosanova M, Pigorini A, Fecchio M, Mariotti M, Lucca A, Colombo
C, Benedetti F, Massimini M (2013) Assessing the effects of electroconvulsive
therapy on cortical excitability by means of transcranial magnetic stimulation and
electroencephalography. Brain Topogr 26:326-337.
Casarotto S, Maatta S, Herukka SK, Pigorini A, Napolitani M, Gosseries O, Niskanen E,
Kononen M, Mervaala E, Rosanova M, Soininen H, Massimini M (2011)
Transcranial magnetic stimulation-evoked EEG/cortical potentials in physiological
and pathological aging. Neuroreport 22:592-597.
Casey BJ, Craddock N, Cuthbert BN, Hyman SE, Lee FS, Ressler KJ (2013) DSM-5 and
RDoC: progress in psychiatry research? Nat Rev Neurosci 14:810-814.
Cavedini P, Riboldi G, D'Annucci A, Belotti P, Cisima M, Bellodi L (2002) Decision-
making heterogeneity in obsessive-compulsive disorder: ventromedial prefrontal
cortex function predicts different treatment outcomes. Neuropsychologia 40:205-211.
Chakrabarty K, Bhattacharyya S, Christopher R, Khanna S (2005) Glutamatergic dysfunction
in OCD. Neuropsychopharmacology 30:1735-1740.
Chen CM, Stanford AD, Mao X, Abi-Dargham A, Shungu DC, Lisanby SH, Schroeder CE,
Kegeles LS (2014) GABA level, gamma oscillation, and working memory
performance in schizophrenia. Neuroimage Clin 4:531-539.
Chen R, Tam A, Butefisch C, Corwell B, Ziemann U, Rothwell JC, Cohen LG (1998)
Intracortical inhibition and facilitation in different representations of the human motor
cortex. J Neurophysiol 80:2870-2881.
169
Chipchase L, Schabrun S, Cohen L, Hodges P, Ridding M, Rothwell J, Taylor J, Ziemann U
(2012) A checklist for assessing the methodological quality of studies using
transcranial magnetic stimulation to study the motor system: an international
consensus study. Clin Neurophysiol 123:1698-1704.
Cho RY, Konecky RO, Carter CS (2006) Impairments in frontal cortical gamma synchrony
and cognitive control in schizophrenia. Proc Natl Acad Sci U S A 103:19878-19883.
Chroni E, Lekka NP, Tsoussis I, Nikolakopoulou A, Paschalis C, Beratis S (2002) Effect of
exercise on motor evoked potentials elicited by transcranial magnetic stimulation in
psychiatric patients. J Clin Neurophysiol 19:240-244.
Chue P, Eerdekens M, Augustyns I, Lachaux B, Molcan P, Eriksson L, Pretorius H, David
AS (2005) Comparative efficacy and safety of long-acting risperidone and risperidone
oral tablets. Eur Neuropsychopharmacol 15:111-117.
Ciranna L (2006) Serotonin as a modulator of glutamate- and GABA-mediated
neurotransmission: implications in physiological functions and in pathology. Curr
Neuropharmacol 4:101-114.
Classen J, Liepert J, Wise SP, Hallett M, Cohen LG (1998) Rapid plasticity of human
cortical movement representation induced by practice. J Neurophysiol 79:1117-1123.
Claus D, Weis M, Jahnke U, Plewe A, Brunholzl C (1992) Corticospinal conduction studied
with magnetic double stimulation in the intact human. J Neurol Sci 111:180-188.
Clementz BA, Blumenfeld LD (2001) Multichannel electroencephalographic assessment of
auditory evoked response suppression in schizophrenia. Exp Brain Res 139:377-390.
Clementz BA, Blumenfeld LD, Cobb S (1997) The gamma band response may account for
poor P50 suppression in schizophrenia. Neuroreport 8:3889-3893.
Constantinidis C, Williams GV, Goldman-Rakic PS (2002) A role for inhibition in shaping
the temporal flow of information in prefrontal cortex. Nat Neurosci 5:175-180.
Costa E, Davis J, Grayson DR, Guidotti A, Pappas GD, Pesold C (2001) Dendritic spine
hypoplasticity and downregulation of reelin and GABAergic tone in schizophrenia
vulnerability. Neurobiol Dis 8:723-742.
Coyle JT (2004) The GABA-glutamate connection in schizophrenia: which is the proximate
cause? Biochem Pharmacol 68:1507-1514.
170
Cui RQ, Huter D, Lang W, Deecke L (1999) Neuroimage of voluntary movement:
topography of the Bereitschaftspotential, a 64-channel DC current source density
study. Neuroimage 9:124-134.
Curran T, D'Arcangelo G (1998) Role of reelin in the control of brain development. Brain
Res Brain Res Rev 26:285-294.
Daskalakis ZJ (2012) On a quest for the elusive schizophrenia biomarker. Biol Psychiatry
72:714-715.
Daskalakis ZJ, Christensen BK, Chen R, Fitzgerald PB, Zipursky RB, Kapur S (2002a)
Evidence for impaired cortical inhibition in schizophrenia using transcranial magnetic
stimulation. Arch Gen Psychiatry 59:347-354.
Daskalakis ZJ, Christensen BK, Fitzgerald PB, Chen R (2008a) Dysfunctional neural
plasticity in patients with schizophrenia. Arch Gen Psychiatry 65:378-385.
Daskalakis ZJ, Christensen BK, Fitzgerald PB, Moller B, Fountain SI, Chen R (2008b)
Increased cortical inhibition in persons with schizophrenia treated with clozapine. J
Psychopharmacol 22:203-209.
Daskalakis ZJ, Christensen BK, Fitzgerald PB, Roshan L, Chen R (2002b) The mechanisms
of interhemispheric inhibition in the human motor cortex. J Physiol 543:317-326.
Daskalakis ZJ, Farzan F, Barr MS, Maller JJ, Chen R, Fitzgerald PB (2008c) Long-interval
cortical inhibition from the dorsolateral prefrontal cortex: a TMS-EEG study.
Neuropsychopharmacology 33:2860-2869.
Daskalakis ZJ, Farzan F, Barr MS, Rusjan PM, Favalli G, Levinson AJ, Fitzgerald PB
(2008d) Evaluating the relationship between long interval cortical inhibition, working
memory and gamma band activity in the dorsolateral prefrontal cortex. Clin EEG
Neurosci 39:150-155.
Daskalakis ZJ, Farzan F, Radhu N, Fitzgerald PB (2012) Combined transcranial magnetic
stimulation and electroencephalography: its past, present and future. Brain Res
1463:93-107.
Daskalakis ZJ, Fitzgerald PB, Christensen BK (2007) The role of cortical inhibition in the
pathophysiology and treatment of schizophrenia. Brain Res Rev 56:427-442.
171
Daskalakis ZJ, Moller B, Christensen BK, Fitzgerald PB, Gunraj C, Chen R (2006) The
effects of repetitive transcranial magnetic stimulation on cortical inhibition in healthy
human subjects. Exp Brain Res 174:403-412.
Daskalakis ZJ, Paradiso GO, Christensen BK, Fitzgerald PB, Gunraj C, Chen R (2004)
Exploring the connectivity between the cerebellum and motor cortex in humans. J
Physiol 557:689-700.
de Leeuw M, Kahn RS, Zandbelt BB, Widschwendter CG, Vink M Working memory and
default mode network abnormalities in unaffected siblings of schizophrenia patients.
Schizophr Res 150:555-562.
Decloedt EH, Stein DJ (2010) Current trends in drug treatment of obsessive-compulsive
disorder. Neuropsychiatr Dis Treat 6:233-242.
Deco G, Rolls ET (2003) Attention and working memory: a dynamical model of neuronal
activity in the prefrontal cortex. Eur J Neurosci 18:2374-2390.
DeFelipe J, Conley M, Jones EG (1986) Long-range focal collateralization of axons arising
from corticocortical cells in monkey sensory-motor cortex. J Neurosci 6:3749-3766.
Deisz RA (1999a) GABA(B) receptor-mediated effects in human and rat neocortical
neurones in vitro. Neuropharmacology 38:1755-1766.
Deisz RA (1999b) GABA(B) receptor-mediated effects in human and rat neocortical
neurones in vitro. Neuropharmacology 38:1755-1766.
Deisz RA (1999c) The GABA(B) receptor antagonist CGP 55845A reduces presynaptic
GABA(B) actions in neocortical neurons of the rat in vitro. Neuroscience 93:1241-
1249.
Di Lazzaro V, Oliviero A, Profice P, Pennisi MA, Pilato F, Zito G, Dileone M, Nicoletti R,
Pasqualetti P, Tonali PA (2003) Ketamine increases human motor cortex excitability
to transcranial magnetic stimulation. J Physiol 547:485-496.
Dickel DE, Veenstra-VanderWeele J, Cox NJ, Wu X, Fischer DJ, Van Etten-Lee M, Himle
JA, Leventhal BL, Cook EH, Jr., Hanna GL (2006) Association testing of the
positional and functional candidate gene SLC1A1/EAAC1 in early-onset obsessive-
compulsive disorder. Arch Gen Psychiatry 63:778-785.
Douglas R, Martin K (1998) Neocortex. In: The synaptic organization of the brain (Shepherd
GM, ed), pp 459-509. New York: Oxford University Press.
172
Eichhammer P, Wiegand R, Kharraz A, Langguth B, Binder H, Hajak G (2004) Cortical
excitability in neuroleptic-naive first-episode schizophrenic patients. Schizophr Res
67:253-259.
Eisen JL, Beer DA, Pato MT, Venditto TA, Rasmussen SA (1997) Obsessive-compulsive
disorder in patients with schizophrenia or schizoaffective disorder. Am J Psychiatry
154:271-273.
Eisen JL, Mancebo MA, Pinto A, Coles ME, Pagano ME, Stout R, Rasmussen SA (2006)
Impact of obsessive-compulsive disorder on quality of life. Compr Psychiatry 47:270-
275.
Enticott PG, Hoy KE, Herring SE, Johnston PJ, Daskalakis ZJ, Fitzgerald PB (2008)
Reduced motor facilitation during action observation in schizophrenia: a mirror
neuron deficit? Schizophr Res 102:116-121.
Esser SK, Huber R, Massimini M, Peterson MJ, Ferrarelli F, Tononi G (2006) A direct
demonstration of cortical LTP in humans: A combined TMS/EEG study. Brain Res
Bull 69:86-94.
Ethridge LE, Hamm JP, Pearlson GD, Tamminga CA, Sweeney JA, Keshavan MS, Clementz
BA (2015) Event-related potential and time-frequency endophenotypes for
schizophrenia and psychotic bipolar disorder. Biol Psychiatry 77:127-136.
Ettinger U, Picchioni M, Landau S, Matsumoto K, van Haren NE, Marshall N, Hall MH,
Schulze K, Toulopoulou T, Davies N, Ribchester T, McGuire PK, Murray RM (2007)
Magnetic resonance imaging of the thalamus and adhesio interthalamica in twins with
schizophrenia. Arch Gen Psychiatry 64:401-409.
Fabisch K, Fabisch H, Langs G, Wieselmann G, Zapotoczski HG (1997) Obsessive-
compulsive symptoms in schizophrenia. Schizophr Res 24:15.
Farzan F, Barr MS, Hoppenbrouwers SS, Fitzgerald PB, Chen R, Pascual-Leone A,
Daskalakis ZJ (2013) The EEG correlates of the TMS-induced EMG silent period in
humans. Neuroimage 83:120-134.
Farzan F, Barr MS, Levinson AJ, Chen R, Wong W, Fitzgerald PB, Daskalakis ZJ (2010a)
Evidence for gamma inhibition deficits in the dorsolateral prefrontal cortex of
patients with schizophrenia. Brain 133:1505-1514.
173
Farzan F, Barr MS, Levinson AJ, Chen R, Wong W, Fitzgerald PB, Daskalakis ZJ (2010b)
Reliability of long-interval cortical inhibition in healthy human subjects: a TMS-EEG
study. J Neurophysiol 104:1339-1346.
Farzan F, Barr MS, Wong W, Chen R, Fitzgerald PB, Daskalakis ZJ (2009) Suppression of
gamma-oscillations in the dorsolateral prefrontal cortex following long interval
cortical inhibition: a TMS-EEG study. Neuropsychopharmacology 34:1543-1551.
Ferbert A, Priori A, Rothwell JC, Day BL, Colebatch JG, Marsden CD (1992)
Interhemispheric inhibition of the human motor cortex. J Physiol 453:525-546.
Ferrarelli F, Massimini M, Peterson MJ, Riedner BA, Lazar M, Murphy MJ, Huber R,
Rosanova M, Alexander AL, Kalin N, Tononi G (2008) Reduced evoked gamma
oscillations in the frontal cortex in schizophrenia patients: a TMS/EEG study. Am J
Psychiatry 165:996-1005.
Ferrarelli F, Massimini M, Sarasso S, Casali A, Riedner BA, Angelini G, Tononi G, Pearce
RA (2010) Breakdown in cortical effective connectivity during midazolam-induced
loss of consciousness. Proc Natl Acad Sci U S A 107:2681-2686.
Ferreri F, Pasqualetti P, Maatta S, Ponzo D, Ferrarelli F, Tononi G, Mervaala E, Miniussi C,
Rossini PM (2011) Human brain connectivity during single and paired pulse
transcranial magnetic stimulation. Neuroimage 54:90-102.
Fitzgerald PB, Brown TL, Daskalakis ZJ, deCastella A, Kulkarni J (2002a) A study of
transcallosal inhibition in schizophrenia using transcranial magnetic stimulation.
Schizophr Res 56:199-209.
Fitzgerald PB, Brown TL, Daskalakis ZJ, Kulkarni J (2002b) A transcranial magnetic
stimulation study of inhibitory deficits in the motor cortex in patients with
schizophrenia. Psychiatry Res 114:11-22.
Fitzgerald PB, Brown TL, Daskalakis ZJ, Kulkarni J (2002c) A transcranial magnetic
stimulation study of the effects of olanzapine and risperidone on motor cortical
excitability in patients with schizophrenia. Psychopharmacology (Berl) 162:74-81.
Fitzgerald PB, Brown TL, Marston NA, Daskalakis ZJ, de Castella A, Bradshaw JL,
Kulkarni J (2004a) Motor cortical excitability and clinical response to rTMS in
depression. J Affect Disord 82:71-76.
174
Fitzgerald PB, Brown TL, Marston NA, Oxley T, De Castella A, Daskalakis ZJ, Kulkarni J
(2004b) Reduced plastic brain responses in schizophrenia: a transcranial magnetic
stimulation study. Schizophr Res 71:17-26.
Fitzgerald PB, Brown TL, Marston NA, Oxley TJ, de Castella A, Daskalakis ZJ, Kulkarni J
(2003) A transcranial magnetic stimulation study of abnormal cortical inhibition in
schizophrenia. Psychiatry Res 118:197-207.
Fitzgerald PB, Daskalakis ZJ, Hoy K, Farzan F, Upton DJ, Cooper NR, Maller JJ (2008)
Cortical inhibition in motor and non-motor regions: a combined TMS-EEG study.
Clin EEG Neurosci 39:112-117.
Fitzgerald PB, Maller JJ, Hoy K, Farzan F, Daskalakis ZJ (2009a) GABA and cortical
inhibition in motor and non-motor regions using combined TMS-EEG: a time
analysis. Clin Neurophysiol 120:1706-1710.
Fitzgerald PB, Williams S, Daskalakis ZJ (2009b) A transcranial magnetic stimulation study
of the effects of cannabis use on motor cortical inhibition and excitability.
Neuropsychopharmacology 34:2368-2375.
Flint J, Timpson N, Munafo M (2014) Assessing the utility of intermediate phenotypes for
genetic mapping of psychiatric disease. Trends Neurosci 37:733-741.
Ford JM, Roth WT, Menon V, Pfefferbaum A (1999) Failures of automatic and strategic
processing in schizophrenia: comparisons of event-related brain potential and startle
blink modification. Schizophr Res 37:149-163.
Frangou S, Sharma T, Alarcon G, Sigmudsson T, Takei N, Binnie C, Murray RM (1997) The
Maudsley Family Study, II: Endogenous event-related potentials in familial
schizophrenia. Schizophr Res 23:45-53.
Frantseva M, Cui J, Farzan F, Chinta LV, Perez Velazquez JL, Daskalakis ZJ (2012)
Disrupted Cortical Conductivity in Schizophrenia: TMS-EEG Study. Cereb Cortex.
Frantseva MV, Fitzgerald PB, Chen R, Moller B, Daigle M, Daskalakis ZJ (2008) Evidence
for impaired long-term potentiation in schizophrenia and its relationship to motor
skill learning. Cereb Cortex 18:990-996.
Freedman R, Adams CE, Adler LE, Bickford PC, Gault J, Harris JG, Nagamoto HT, Olincy
A, Ross RG, Stevens KE, Waldo M, Leonard S (2000) Inhibitory neurophysiological
175
deficit as a phenotype for genetic investigation of schizophrenia. Am J Med Genet
97:58-64.
Fries P, Nikolic D, Singer W (2007) The gamma cycle. Trends Neurosci 30:309-316.
Friston KJ (1998) The disconnection hypothesis. Schizophr Res 30:115-125.
Fuggetta G, Fiaschi A, Manganotti P (2005) Modulation of cortical oscillatory activities
induced by varying single-pulse transcranial magnetic stimulation intensity over the
left primary motor area: a combined EEG and TMS study. Neuroimage 27:896-908.
Gaiarsa JL, Caillard O, Ben-Ari Y (2002) Long-term plasticity at GABAergic and
glycinergic synapses: mechanisms and functional significance. Trends Neurosci
25:564-570.
Garcia Dominguez L, Radhu N, Farzan F, Daskalakis ZJ (2014) Characterizing Long Interval
Cortical Inhibition over the Time-Frequency Domain. PLoS One 9:e92354.
Gazzaniga MS (2000) Cerebral specialization and interhemispheric communication: does the
corpus callosum enable the human condition? Brain 123 ( Pt 7):1293-1326.
Gerloff C, Cohen LG, Floeter MK, Chen R, Corwell B, Hallett M (1998) Inhibitory influence
of the ipsilateral motor cortex on responses to stimulation of the human cortex and
pyramidal tract. J Physiol 510 ( Pt 1):249-259.
Gilbert CD, Wiesel TN (1992) Receptive field dynamics in adult primary visual cortex.
Nature 356:150-152.
Gonzalez-Burgos G, Lewis DA NMDA receptor hypofunction, parvalbumin-positive
neurons, and cortical gamma oscillations in schizophrenia. Schizophr Bull 38:950-
957.
Goodwin RD, Marusic A, Hoven CW (2003) Suicide attempts in the United States: the role
of physical illness. Soc Sci Med 56:1783-1788.
Goto N, Yoshimura R, Moriya J, Kakeda S, Ueda N, Ikenouchi-Sugita A, Umene-Nakano W,
Hayashi K, Oonari N, Korogi Y, Nakamura J (2009) Reduction of brain gamma-
aminobutyric acid (GABA) concentrations in early-stage schizophrenia patients: 3T
Proton MRS study. Schizophr Res 112:192-193.
Gottesman, II, Gould TD (2003) The endophenotype concept in psychiatry: etymology and
strategic intentions. Am J Psychiatry 160:636-645.
176
Graham FK (1975) Presidential Address, 1974. The more or less startling effects of weak
prestimulation. Psychophysiology 12:238-248.
Greenberg BD, Ziemann U, Cora-Locatelli G, Harmon A, Murphy DL, Keel JC,
Wassermann EM (2000) Altered cortical excitability in obsessive-compulsive
disorder. Neurology 54:142-147.
Greenberg BD, Ziemann U, Harmon A, Murphy DL, Wassermann EM (1998) Decreased
neuronal inhibition in cerebral cortex in obsessive-compulsive disorder on
transcranial magnetic stimulation. Lancet 352:881-882.
Greenwood TA, Braff DL, Light GA, Cadenhead KS, Calkins ME, Dobie DJ, Freedman R,
Green MF, Gur RE, Gur RC, Mintz J, Nuechterlein KH, Olincy A, Radant AD,
Seidman LJ, Siever LJ, Silverman JM, Stone WS, Swerdlow NR, Tsuang DW,
Tsuang MT, Turetsky BI, Schork NJ (2007) Initial heritability analyses of
endophenotypic measures for schizophrenia: the consortium on the genetics of
schizophrenia. Arch Gen Psychiatry 64:1242-1250.
Gross-Isseroff R, Hermesh H, Zohar J, Weizman A (2003) Neuroimaging communality
between schizophrenia and obsessive compulsive disorder: a putative basis for
schizo-obsessive disorder? World J Biol Psychiatry 4:129-134.
Grunhaus L, Polak D, Amiaz R, Dannon PN (2003) Motor-evoked potential amplitudes
elicited by transcranial magnetic stimulation do not differentiate between patients and
normal controls. Int J Neuropsychopharmacol 6:371-378.
Guglietti CL, Daskalakis ZJ, Radhu N, Fitzgerald PB, Ritvo P (2013) Meditation-Related
Increases in GABAB Modulated Cortical Inhibition. Brain Stimul 6:397-402.
Hajek T, Cullis J, Novak T, Kopecek M, Blagdon R, Propper L, Stopkova P, Duffy A,
Hoschl C, Uher R, Paus T, Young LT, Alda M (2013) Brain structural signature of
familial predisposition for bipolar disorder: replicable evidence for involvement of
the right inferior frontal gyrus. Biol Psychiatry 73:144-152.
Hall MH, Taylor G, Sham P, Schulze K, Rijsdijk F, Picchioni M, Toulopoulou T, Ettinger U,
Bramon E, Murray RM, Salisbury DF (2011) The early auditory gamma-band
response is heritable and a putative endophenotype of schizophrenia. Schizophr Bull
37:778-787.
177
Hallett M (2000) Transcranial magnetic stimulation and the human brain. Nature 406:147-
150.
Hamidi M, Slagter HA, Tononi G, Postle BR (2010) Brain responses evoked by high-
frequency repetitive transcranial magnetic stimulation: an event-related potential
study. Brain Stimul 3:2-14.
Hamm JP, Ethridge LE, Shapiro JR, Pearlson GD, Tamminga CA, Sweeney JA, Keshavan
MS, Thaker GK, Clementz BA (2013) Family history of psychosis moderates early
auditory cortical response abnormalities in non-psychotic bipolar disorder. Bipolar
Disord 15:774 - 786.
Hamm JP, Ethridge LE, Shapiro JR, Stevens MC, Boutros NN, Summerfelt AT, Keshavan
MS, Sweeney JA, Pearlson G, Tamminga CA, Thaker G, Clementz BA (2012)
Spatiotemporal and frequency domain analysis of auditory paired stimuli processing
in schizophrenia and bipolar disorder with psychosis. Psychophysiology 49:522-530.
Hari R, Kaila K, Katila T, Tuomisto T, Varpula T (1982) Interstimulus interval dependence
of the auditory vertex response and its magnetic counterpart: implications for their
neural generation. Electroencephalogr Clin Neurophysiol 54:561-569.
Hari R, Pelizzone M, Makela JP, Hallstrom J, Leinonen L, Lounasmaa OV (1987)
Neuromagnetic responses of the human auditory cortex to on- and offsets of noise
bursts. Audiology 26:31-43.
Hasan A, Wobrock T, Grefkes C, Labusga M, Levold K, Schneider-Axmann T, Falkai P,
Muller H, Klosterkotter J, Bechdolf A (2012) Deficient Inhibitory Cortical Networks
in Antipsychotic-Naive Subjects at Risk of Developing First-Episode Psychosis and
First-Episode Schizophrenia Patients: A Cross-Sectional Study. Biol Psychiatry.
Hasenkamp W, Epstein MP, Green A, Wilcox L, Boshoven W, Lewison B, Duncan E (2010)
Heritability of acoustic startle magnitude, prepulse inhibition, and startle latency in
schizophrenia and control families. Psychiatry Res 178:236-243.
Hashimoto T, Bazmi HH, Mirnics K, Wu Q, Sampson AR, Lewis DA (2008) Conserved
regional patterns of GABA-related transcript expression in the neocortex of subjects
with schizophrenia. Am J Psychiatry 165:479-489.
178
Hashimoto T, Volk DW, Eggan SM, Mirnics K, Pierri JN, Sun Z, Sampson AR, Lewis DA
(2003) Gene expression deficits in a subclass of GABA neurons in the prefrontal
cortex of subjects with schizophrenia. J Neurosci 23:6315-6326.
Herbsman T, Forster L, Molnar C, Dougherty R, Christie D, Koola J, Ramsey D, Morgan PS,
Bohning DE, George MS, Nahas Z (2009) Motor threshold in transcranial magnetic
stimulation: the impact of white matter fiber orientation and skull-to-cortex distance.
Hum Brain Mapp 30:2044-2055.
Heyman I, Mataix-Cols D, Fineberg NA (2006) Obsessive-compulsive disorder. BMJ
333:424-429.
Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med
21:1539-1558.
Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring inconsistency in meta-
analyses. BMJ 327:557-560.
Hodgkin AL, Huxley AF (1952) A quantitative description of membrane current and its
application to conduction and excitation in nerve. J Physiol 117:500-544.
Hong LE, Summerfelt A, McMahon RP, Thaker GK, Buchanan RW (2004) Gamma/beta
oscillation and sensory gating deficit in schizophrenia. Neuroreport 15:155-159.
Hoppenbrouwers SS, De Jesus DR, Stirpe T, Fitzgerald PB, Voineskos AN, Schutter DJ,
Daskalakis ZJ (2013) Inhibitory deficits in the dorsolateral prefrontal cortex in
psychopathic offenders. Cortex 49:1377-1385.
Horwath E, Weissman MM (2000) The epidemiology and cross-national presentation of
obsessive-compulsive disorder. Psychiatr Clin North Am 23:493-507.
Howard MW, Rizzuto DS, Caplan JB, Madsen JR, Lisman J, Aschenbrenner-Scheibe R,
Schulze-Bonhage A, Kahana MJ (2003) Gamma oscillations correlate with working
memory load in humans. Cereb Cortex 13:1369-1374.
Hoy KE, Georgiou-Karistianis N, Laycock R, Fitzgerald PB (2007) Using transcranial
magnetic stimulation to investigate the cortical origins of motor overflow: a study in
schizophrenia and healthy controls. Psychol Med 37:583-594.
Huang C, Wahlund L, Dierks T, Julin P, Winblad B, Jelic V (2000) Discrimination of
Alzheimer's disease and mild cognitive impairment by equivalent EEG sources: a
cross-sectional and longitudinal study. Clin Neurophysiol 111:1961-1967.
179
Hwa GG, Avoli M (1992) Excitatory postsynaptic potentials recorded from regular-spiking
cells in layers II/III of rat sensorimotor cortex. J Neurophysiol 67:728-737.
Ilmoniemi RJ, Kicic D (2010) Methodology for combined TMS and EEG. Brain Topogr
22:233-248.
Ilmoniemi RJ, Virtanen J, Ruohonen J, Karhu J, Aronen HJ, Naatanen R, Katila T (1997)
Neuronal responses to magnetic stimulation reveal cortical reactivity and
connectivity. Neuroreport 8:3537-3540.
Impagnatiello F, Guidotti AR, Pesold C, Dwivedi Y, Caruncho H, Pisu MG, Uzunov DP,
Smalheiser NR, Davis JM, Pandey GN, Pappas GD, Tueting P, Sharma RP, Costa E
(1998) A decrease of reelin expression as a putative vulnerability factor in
schizophrenia. Proc Natl Acad Sci U S A 95:15718-15723.
Ives JR, Rotenberg A, Poma R, Thut G, Pascual-Leone A (2006) Electroencephalographic
recording during transcranial magnetic stimulation in humans and animals. Clin
Neurophysiol 117:1870-1875.
Ivleva EI, Bidesi AS, Keshavan MS, Pearlson GD, Meda SA, Dodig D, Moates AF, Lu H,
Francis AN, Tandon N, Schretlen DJ, Sweeney JA, Clementz BA, Tamminga CA
(2013) Gray Matter Volume as an Intermediate Phenotype for Psychosis: Bipolar-
Schizophrenia Network on Intermediate Phenotypes (B-SNIP). Am J Psychiatry
170:1285-1296.
Ivleva EI, Moates AF, Hamm JP, Bernstein IH, O'Neill HB, Cole D, Clementz BA, Thaker
GK, Tamminga CA (2014) Smooth pursuit eye movement, prepulse inhibition, and
auditory paired stimuli processing endophenotypes across the schizophrenia-bipolar
disorder psychosis dimension. Schizophr Bull 40:642-652.
Jacobs KM, Donoghue JP (1991) Reshaping the cortical motor map by unmasking latent
intracortical connections. Science 251:944-947.
Jansma JM, Ramsey NF, van der Wee NJ, Kahn RS (2004) Working memory capacity in
schizophrenia: a parametric fMRI study. Schizophr Res 68:159-171.
Javaid N, Kennedy JL, De Luca V (2011) Ethnicity and Age at Onset in Bipolar Spectrum
Disorders. CNS Spectr.
Jefferys JG, Traub RD, Whittington MA (1996) Neuronal networks for induced '40 Hz'
rhythms. Trends Neurosci 19:202-208.
180
Jiao Y, Zhang C, Yanagawa Y, Sun QQ (2006) Major effects of sensory experiences on the
neocortical inhibitory circuits. J Neurosci 26:8691-8701.
Johannesen JK, Kieffaber PD, O'Donnell BF, Shekhar A, Evans JD, Hetrick WP (2005)
Contributions of subtype and spectral frequency analyses to the study of P50 ERP
amplitude and suppression in schizophrenia. Schizophr Res 78:269-284.
Jones EG (1993) GABAergic neurons and their role in cortical plasticity in primates. Cereb
Cortex 3:361-372.
Kahkonen S, Kesaniemi M, Nikouline VV, Karhu J, Ollikainen M, Holi M, Ilmoniemi RJ
(2001) Ethanol modulates cortical activity: direct evidence with combined TMS and
EEG. Neuroimage 14:322-328.
Kahkonen S, Wilenius J, Nikulin VV, Ollikainen M, Ilmoniemi RJ (2003) Alcohol reduces
prefrontal cortical excitability in humans: a combined TMS and EEG study.
Neuropsychopharmacology 28:747-754.
Kaplan HI, Sadock BJ, Grebb JA (1994) Kaplan and Sadock's synopsis of psychiatry :
behavioral sciences, clinical psychiatry., ed 7th Edition. Baltimore: Williams &
Wilkins.
Kapur S (2003) Psychosis as a state of aberrant salience: a framework linking biology,
phenomenology, and pharmacology in schizophrenia. Am J Psychiatry 160:13-23.
Karlsgodt KH, Glahn DC, van Erp TG, Therman S, Huttunen M, Manninen M, Kaprio J,
Cohen MS, Lonnqvist J, Cannon TD (2007) The relationship between performance
and fMRI signal during working memory in patients with schizophrenia, unaffected
co-twins, and control subjects. Schizophr Res 89:191-197.
Karlsgodt KH, Sanz J, van Erp TG, Bearden CE, Nuechterlein KH, Cannon TD (2009) Re-
evaluating dorsolateral prefrontal cortex activation during working memory in
schizophrenia. Schizophr Res 108:143-150.
Karno M, Golding JM, Sorenson SB, Burnam MA (1988) The epidemiology of obsessive-
compulsive disorder in five US communities. Arch Gen Psychiatry 45:1094-1099.
Karoumi B, Laurent A, Rosenfeld F, Rochet T, Brunon AM, Dalery J, d'Amato T, Saoud M
(2000) Alteration of event related potentials in siblings discordant for schizophrenia.
Schizophr Res 41:325-334.
181
Keel JC, Smith MJ, Wassermann EM (2001) A safety screening questionnaire for
transcranial magnetic stimulation. Clin Neurophysiol 112:720.
Kegeles LS, Mao X, Stanford AD, Girgis R, Ojeil N, Xu X, Gil R, Slifstein M, Abi-Dargham
A, Lisanby SH, Shungu DC (2012) Elevated prefrontal cortex gamma-aminobutyric
acid and glutamate-glutamine levels in schizophrenia measured in vivo with proton
magnetic resonance spectroscopy. Arch Gen Psychiatry 69:449-459.
Kellner M (2010) Drug treatment of obsessive-compulsive disorder. Dialogues Clin Neurosci
12:187-197.
Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE (2005) Lifetime
prevalence and age-of-onset distributions of DSM-IV disorders in the National
Comorbidity Survey Replication. Arch Gen Psychiatry 62:593-602.
Kicic D, Lioumis P, Ilmoniemi RJ, Nikulin VV (2008) Bilateral changes in excitability of
sensorimotor cortices during unilateral movement: combined electroencephalographic
and transcranial magnetic stimulation study. Neuroscience 152:1119-1129.
Kirihara K, Rissling AJ, Swerdlow NR, Braff DL, Light GA (2012) Hierarchical
organization of gamma and theta oscillatory dynamics in schizophrenia. Biol
Psychiatry 71:873-880.
Knable MB, Barci BM, Webster MJ, Meador-Woodruff J, Torrey EF (2004) Molecular
abnormalities of the hippocampus in severe psychiatric illness: postmortem findings
from the Stanley Neuropathology Consortium. Mol Psychiatry 9:609-620, 544.
Komssi S, Aronen HJ, Huttunen J, Kesaniemi M, Soinne L, Nikouline VV, Ollikainen M,
Roine RO, Karhu J, Savolainen S, Ilmoniemi RJ (2002) Ipsi- and contralateral EEG
reactions to transcranial magnetic stimulation. Clin Neurophysiol 113:175-184.
Komssi S, Kahkonen S (2006) The novelty value of the combined use of
electroencephalography and transcranial magnetic stimulation for neuroscience
research. Brain Res Brain Res Rev 52:183-192.
Komssi S, Kahkonen S, Ilmoniemi RJ (2004) The effect of stimulus intensity on brain
responses evoked by transcranial magnetic stimulation. Hum Brain Mapp 21:154-
164.
182
Kondziella D, Brenner E, Eyjolfsson EM, Sonnewald U (2007) How do glial-neuronal
interactions fit into current neurotransmitter hypotheses of schizophrenia? Neurochem
Int 50:291-301.
Korhonen RJ, Hernandez-Pavon JC, Metsomaa J, Maki H, Ilmoniemi RJ, Sarvas J (2011)
Removal of large muscle artifacts from transcranial magnetic stimulation-evoked
EEG by independent component analysis. Med Biol Eng Comput 49:397-407.
Krnjevic K (1997) Role of GABA in cerebral cortex. Can J Physiol Pharmacol 75:439-451.
Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremner JD, Heninger GR,
Bowers MB, Jr., Charney DS (1994) Subanesthetic effects of the noncompetitive
NMDA antagonist, ketamine, in humans. Psychotomimetic, perceptual, cognitive, and
neuroendocrine responses. Arch Gen Psychiatry 51:199-214.
Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A, Wroe S, Asselman
P, Marsden CD (1993) Corticocortical inhibition in human motor cortex. J Physiol
471:501-519.
Kumari V, Das M, Zachariah E, Ettinger U, Sharma T (2005) Reduced prepulse inhibition in
unaffected siblings of schizophrenia patients. Psychophysiology 42:588-594.
Langguth B, Eichhammer P, Spranz C, Landgrebe M, Frick U, Sand P, Hajak G (2008)
Modulation of human motor cortex excitability by quetiapine. Psychopharmacology
(Berl) 196:623-629.
Lee KH, Williams LM, Breakspear M, Gordon E (2003) Synchronous gamma activity: a
review and contribution to an integrative neuroscience model of schizophrenia. Brain
Res Brain Res Rev 41:57-78.
Lee MJ, Shin YB, Sunwoo YK, Jung SH, Kim WH, Kang MH, Lee JS, Bae JN, Kim CE
(2009) Comparative Analysis of Cognitive Function in Schizophrenia with and
without Obsessive Compulsive Disorder. Psychiatry Investig 6:286-293.
Lefaucheur JP, Lucas B, Andraud F, Hogrel JY, Bellivier F, Del Cul A, Rousseva A,
Leboyer M, Paillere-Martinot ML (2008) Inter-hemispheric asymmetry of motor
corticospinal excitability in major depression studied by transcranial magnetic
stimulation. J Psychiatr Res 42:389-398.
Leicht G, Karch S, Karamatskos E, Giegling I, Moller HJ, Hegerl U, Pogarell O, Rujescu D,
Mulert C (2010) Alterations of the early auditory evoked gamma-band response in
183
first-degree relatives of patients with schizophrenia: hints to a new intermediate
phenotype. J Psychiatr Res 45:699-705.
Leung LS, Shen B (2007) GABAB receptor blockade enhances theta and gamma rhythms in
the hippocampus of behaving rats. Hippocampus 17:281-291.
Levinson AJ, Fitzgerald PB, Favalli G, Blumberger DM, Daigle M, Daskalakis ZJ (2010)
Evidence of cortical inhibitory deficits in major depressive disorder. Biol Psychiatry
67:458-464.
Levinson AJ, Young LT, Fitzgerald PB, Daskalakis ZJ (2007) Cortical inhibitory dysfunction
in bipolar disorder: a study using transcranial magnetic stimulation. J Clin
Psychopharmacol 27:493-497.
Levit-Binnun N, Litvak V, Pratt H, Moses E, Zaroor M, Peled A (2010) Differences in TMS-
evoked responses between schizophrenia patients and healthy controls can be
observed without a dedicated EEG system. Clin Neurophysiol 121:332-339.
Lewis DA, Hashimoto T, Volk DW (2005) Cortical inhibitory neurons and schizophrenia.
Nat Rev Neurosci 6:312-324.
Lewis DA, Pierri JN, Volk DW, Melchitzky DS, Woo TU (1999) Altered GABA
neurotransmission and prefrontal cortical dysfunction in schizophrenia. Biol
Psychiatry 46:616-626.
Lewis DA, Volk DW, Hashimoto T (2004) Selective alterations in prefrontal cortical GABA
neurotransmission in schizophrenia: a novel target for the treatment of working
memory dysfunction. Psychopharmacology (Berl) 174:143-150.
Light GA, Braff DL (2005a) Mismatch negativity deficits are associated with poor
functioning in schizophrenia patients. Arch Gen Psychiatry 62:127-136.
Light GA, Braff DL (2005b) Stability of mismatch negativity deficits and their relationship
to functional impairments in chronic schizophrenia. Am J Psychiatry 162:1741-1743.
Light GA, Hsu JL, Hsieh MH, Meyer-Gomes K, Sprock J, Swerdlow NR, Braff DL (2006)
Gamma band oscillations reveal neural network cortical coherence dysfunction in
schizophrenia patients. Biol Psychiatry 60:1231-1240.
Light GA, Makeig S (2015) Electroencephalographic biomarkers of psychosis: present and
future. Biol Psychiatry 77:87-89.
184
Light GA, Swerdlow NR, Thomas ML, Calkins ME, Green MF, Greenwood TA, Gur RE,
Gur RC, Lazzeroni LC, Nuechterlein KH, Pela M, Radant AD, Seidman LJ, Sharp
RF, Siever LJ, Silverman JM, Sprock J, Stone WS, Sugar CA, Tsuang DW, Tsuang
MT, Braff DL, Turetsky BI Validation of mismatch negativity and P3a for use in
multi-site studies of schizophrenia: Characterization of demographic, clinical,
cognitive, and functional correlates in COGS-2. Schizophr Res.
Lioumis P, Kicic D, Savolainen P, Makela JP, Kahkonen S (2009) Reproducibility of TMS-
Evoked EEG responses. Hum Brain Mapp 30:1387-1396.
Lisman JE, Coyle JT, Green RW, Javitt DC, Benes FM, Heckers S, Grace AA (2008)
Circuit-based framework for understanding neurotransmitter and risk gene
interactions in schizophrenia. Trends Neurosci 31:234-242.
Litvak V, Komssi S, Scherg M, Hoechstetter K, Classen J, Zaaroor M, Pratt H, Kahkonen S
(2007) Artifact correction and source analysis of early electroencephalographic
responses evoked by transcranial magnetic stimulation over primary motor cortex.
Neuroimage 37:56-70.
Liu SK, Fitzgerald PB, Daigle M, Chen R, Daskalakis ZJ (2009) The relationship between
cortical inhibition, antipsychotic treatment, and the symptoms of schizophrenia. Biol
Psychiatry 65:503-509.
Maeda F, Keenan JP, Pascual-Leone A (2000) Interhemispheric asymmetry of motor cortical
excitability in major depression as measured by transcranial magnetic stimulation. Br
J Psychiatry 177:169-173.
Manchia M, Lampus S, Chillotti C, Sardu C, Ardau R, Severino G, Del Zompo M (2008)
Age at onset in Sardinian bipolar I patients: evidence for three subgroups. Bipolar
Disord 10:443-446.
Manganotti P, Bortolomasi M, Zanette G, Pawelzik T, Giacopuzzi M, Fiaschi A (2001)
Intravenous clomipramine decreases excitability of human motor cortex. A study with
paired magnetic stimulation. J Neurol Sci 184:27-32.
Maris E, Oostenveld R (2007) Nonparametric statistical testing of EEG- and MEG-data. J
Neurosci Methods 164:177-190.
Marrosu F, Santoni F, Fa M, Puligheddu M, Barberini L, Genugu F, Frau R, Manunta M,
Mereu G (2006) Beta and gamma range EEG power-spectrum correlation with
185
spiking discharges in DBA/2J mice absence model: role of GABA receptors.
Epilepsia 47:489-494.
Marsman A, Mandl RC, Klomp DW, Bohlken MM, Boer VO, Andreychenko A, Cahn W,
Kahn RS, Luijten PR, Hulshoff Pol HE (2014) GABA and glutamate in
schizophrenia: A 7 T (1)H-MRS study. Neuroimage Clin 6:398-407.
Massimini M, Ferrarelli F, Esser SK, Riedner BA, Huber R, Murphy M, Peterson MJ,
Tononi G (2007) Triggering sleep slow waves by transcranial magnetic stimulation.
Proc Natl Acad Sci U S A 104:8496-8501.
Massimini M, Ferrarelli F, Huber R, Esser SK, Singh H, Tononi G (2005) Breakdown of
cortical effective connectivity during sleep. Science 309:2228-2232.
Massimini M, Ferrarelli F, Murphy M, Huber R, Riedner B, Casarotto S, Tononi G (2010)
Cortical reactivity and effective connectivity during REM sleep in humans. Cogn
Neurosci 1:176-183.
McClintock SM, Freitas C, Oberman L, Lisanby SH, Pascual-Leone A (2011) Transcranial
magnetic stimulation: a neuroscientific probe of cortical function in schizophrenia.
Biol Psychiatry 70:19-27.
McCormick DA (1989) GABA as an inhibitory neurotransmitter in human cerebral cortex. J
Neurophysiol 62:1018-1027.
McDonnell MN, Orekhov Y, Ziemann U (2006) The role of GABA(B) receptors in
intracortical inhibition in the human motor cortex. Exp Brain Res 173:86-93.
McGrath J, Saha S, Chant D, Welham J (2008) Schizophrenia: a concise overview of
incidence, prevalence, and mortality. Epidemiol Rev 30:67-76.
Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ, Bullmore ET (2008)
Integrating evidence from neuroimaging and neuropsychological studies of obsessive-
compulsive disorder: the orbitofronto-striatal model revisited. Neurosci Biobehav
Rev 32:525-549.
Merikangas KR, Jin R, He JP, Kessler RC, Lee S, Sampson NA, Viana MC, Andrade LH, Hu
C, Karam EG, Ladea M, Medina-Mora ME, Ono Y, Posada-Villa J, Sagar R, Wells
JE, Zarkov Z (2011) Prevalence and correlates of bipolar spectrum disorder in the
world mental health survey initiative. Arch Gen Psychiatry 68:241-251.
186
Meyer-Lindenberg A, Weinberger DR (2006) Intermediate phenotypes and genetic
mechanisms of psychiatric disorders. Nat Rev Neurosci 7:818-827.
Michie PT, Innes-Brown H, Todd J, Jablensky AV (2002) Duration mismatch negativity in
biological relatives of patients with schizophrenia spectrum disorders. Biol Psychiatry
52:749-758.
Minelli A, Bortolomasi M, Scassellati C, Salvoro B, Avesani M, Manganotti P (2010) Effects
of intravenous antidepressant drugs on the excitability of human motor cortex: a study
with paired magnetic stimulation on depressed patients. Brain Stimul 3:15-21.
Miniussi C, Thut G (2010) Combining TMS and EEG offers new prospects in cognitive
neuroscience. Brain Topogr 22:249-256.
Monaghan DT, Bridges RJ, Cotman CW (1989) The excitatory amino acid receptors: their
classes, pharmacology, and distinct properties in the function of the central nervous
system. Annu Rev Pharmacol Toxicol 29:365-402.
Montez T, Poil SS, Jones BF, Manshanden I, Verbunt JP, van Dijk BW, Brussaard AB, van
Ooyen A, Stam CJ, Scheltens P, Linkenkaer-Hansen K (2009) Altered temporal
correlations in parietal alpha and prefrontal theta oscillations in early-stage Alzheimer
disease. Proc Natl Acad Sci U S A 106:1614-1619.
Morbidi F, Garulli A, Prattichizzo D, Rizzo C, Manganotti P, Rossi S (2007) Off-line
removal of TMS-induced artifacts on human electroencephalography by Kalman
filter. J Neurosci Methods 162:293-302.
Moreau AW, Kullmann DM (2013) NMDA receptor-dependent function and plasticity in
inhibitory circuits. Neuropharmacology 74:23-31.
Naatanen R, Paavilainen P, Alho K, Reinikainen K, Sams M (1989) Do event-related
potentials reveal the mechanism of the auditory sensory memory in the human brain?
Neurosci Lett 98:217-221.
Nakamura H, Kitagawa H, Kawaguchi Y, Tsuji H (1997) Intracortical facilitation and
inhibition after transcranial magnetic stimulation in conscious humans. J Physiol 498
( Pt 3):817-823.
Nikulin VV, Kicic D, Kahkonen S, Ilmoniemi RJ (2003) Modulation of
electroencephalographic responses to transcranial magnetic stimulation: evidence for
changes in cortical excitability related to movement. Eur J Neurosci 18:1206-1212.
187
Nordstrom EJ, Burton FH (2002) A transgenic model of comorbid Tourette's syndrome and
obsessive-compulsive disorder circuitry. Mol Psychiatry 7:617-625, 524.
Nunez PL, Srinivasan R (2006) A theoretical basis for standing and traveling brain waves
measured with human EEG with implications for an integrated consciousness. Clin
Neurophysiol 117:2424-2435.
Ogura C, Nageishi Y, Matsubayashi M, Omura F, Kishimoto A, Shimokochi M (1991)
Abnormalities in event-related potentials, N100, P200, P300 and slow wave in
schizophrenia. Jpn J Psychiatry Neurol 45:57-65.
Olincy A, Braff DL, Adler LE, Cadenhead KS, Calkins ME, Dobie DJ, Green MF,
Greenwood TA, Gur RE, Gur RC, Light GA, Mintz J, Nuechterlein KH, Radant AD,
Schork NJ, Seidman LJ, Siever LJ, Silverman JM, Stone WS, Swerdlow NR, Tsuang
DW, Tsuang MT, Turetsky BI, Wagner BD, Freedman R (2010) Inhibition of the P50
cerebral evoked response to repeated auditory stimuli: results from the Consortium on
Genetics of Schizophrenia. Schizophr Res 119:175-182.
Olincy A, Martin L (2005) Diminished suppression of the P50 auditory evoked potential in
bipolar disorder subjects with a history of psychosis. Am J Psychiatry 162:43-49.
Olney JW, Newcomer JW, Farber NB (1999) NMDA receptor hypofunction model of
schizophrenia. J Psychiatr Res 33:523-533.
Ongur D, Prescot AP, McCarthy J, Cohen BM, Renshaw PF (2010) Elevated gamma-
aminobutyric acid levels in chronic schizophrenia. Biol Psychiatry 68:667-670.
Overall JE, Gorham DR (1962) The Brief Psychiatric Rating Scale. Psychological Reports,
10:790-812.
Oxley T, Fitzgerald PB, Brown TL, de Castella A, Daskalakis ZJ, Kulkarni J (2004)
Repetitive transcranial magnetic stimulation reveals abnormal plastic response to
premotor cortex stimulation in schizophrenia. Biol Psychiatry 56:628-633.
Palaniyappan L, Balain V, Liddle PF (2012) The neuroanatomy of psychotic diathesis: a
meta-analytic review. J Psychiatr Res 46:1249-1256.
Palva JM, Palva S, Kaila K (2005) Phase synchrony among neuronal oscillations in the
human cortex. J Neurosci 25:3962-3972.
Pascual-Leone A, Manoach DS, Birnbaum R, Goff DC (2002) Motor cortical excitability in
schizophrenia. Biol Psychiatry 52:24-31.
188
Paulus W, Classen J, Cohen LG, Large CH, Di Lazzaro V, Nitsche M, Pascual-Leone A,
Rosenow F, Rothwell JC, Ziemann U (2008a) State of the art: Pharmacologic effects
on cortical excitability measures tested by transcranial magnetic stimulation. Brain
Stimul 1:151-163.
Paulus W, Classen J, Cohen LG, Large CH, Di Lazzaro V, Nitsche M, Pascual-Leone A,
Rosenow F, Rothwell JC, Ziemann U (2008b) State of the art: Pharmacologic effects
on cortical excitability measures tested by transcranial magnetic stimulation. Brain
Stimulation 1:151-163.
Paus T, Sipila PK, Strafella AP (2001) Synchronization of neuronal activity in the human
primary motor cortex by transcranial magnetic stimulation: an EEG study. J
Neurophysiol 86:1983-1990.
Pope KJ, Fitzgibbon SP, Lewis TW, Whitham EM, Willoughby JO (2009) Relation of
gamma oscillations in scalp recordings to muscular activity. Brain Topogr 22:13-17.
Porto L, Bermanzohn P, Siris S, Pollack S, Morrissey R (1997) A profile of obsessive
compulsive symptoms in schizophrenia. Schizophr Res 24:20.
Potkin SG, Turner JA, Brown GG, McCarthy G, Greve DN, Glover GH, Manoach DS,
Belger A, Diaz M, Wible CG, Ford JM, Mathalon DH, Gollub R, Lauriello J, O'Leary
D, van Erp TG, Toga AW, Preda A, Lim KO (2009) Working memory and DLPFC
inefficiency in schizophrenia: the FBIRN study. Schizophr Bull 35:19-31.
Poyurovsky M, Fuchs C, Weizman A (1999) Obsessive-compulsive disorder in patients with
first-episode schizophrenia. Am J Psychiatry 156:1998-2000.
Poyurovsky M, Koran LM (2005) Obsessive-compulsive disorder (OCD) with schizotypy vs.
schizophrenia with OCD: diagnostic dilemmas and therapeutic implications. J
Psychiatr Res 39:399-408.
Premoli I, Castellanos N, Rivolta D, Belardinelli P, Bajo R, Zipser C, Espenhahn S,
Heidegger T, Muller-Dahlhaus F, Ziemann U (2014a) TMS-EEG Signatures of
GABAergic Neurotransmission in the Human Cortex. J Neurosci 34:5603-5612.
Premoli I, Rivolta D, Espenhahn S, Castellanos N, Belardinelli P, Ziemann U, Muller-
Dahlhaus F (2014b) Characterization of GABAB-receptor mediated
neurotransmission in the human cortex by paired-pulse TMS-EEG. Neuroimage
103:152-162.
189
Preston GA, Weinberger DR (2005) Intermediate phenotypes in schizophrenia: a selective
review. Dialogues Clin Neurosci 7:165-179.
Radant AD, Dobie DJ, Calkins ME, Olincy A, Braff DL, Cadenhead KS, Freedman R, Green
MF, Greenwood TA, Gur RE, Gur RC, Light GA, Meichle SP, Millard SP, Mintz J,
Nuechterlein KH, Schork NJ, Seidman LJ, Siever LJ, Silverman JM, Stone WS,
Swerdlow NR, Tsuang MT, Turetsky BI, Tsuang DW (2010) Antisaccade
performance in schizophrenia patients, their first-degree biological relatives, and
community comparison subjects: data from the COGS study. Psychophysiology
47:846-856.
Radhu N, Daskalakis ZJ, Guglietti CL, Farzan F, Barr MS, Arpin-Cribbie CA, Fitzgerald PB,
Ritvo P (2012) Cognitive behavioral therapy-related increases in cortical inhibition in
problematic perfectionists. Brain Stimul 5:44-54.
Radhu N, de Jesus DR, Ravindran LN, Zanjani A, Fitzgerald PB, Daskalakis ZJ (2013) A
meta-analysis of cortical inhibition and excitability using transcranial magnetic
stimulation in psychiatric disorders. Clin Neurophysiol 124:1309-1320.
Radhu N, Garcia Dominguez L, Farzan F, Richter MA, Semeralul MO, Chen R, Fitzgerald
PB, Daskalakis ZJ (2015) Evidence for inhibitory deficits in the prefrontal cortex in
schizophrenia. Brain 138:483-497.
Raine A (1991) The SPQ: a scale for the assessment of schizotypal personality based on
DSM-III-R criteria. Schizophr Bull 17:555-564.
Rajji TK, Liu SK, Frantseva MV, Mulsant BH, Thoma J, Chen R, Fitzgerald PB, Daskalakis
ZJ (2011) Exploring the effect of inducing long-term potentiation in the human motor
cortex on motor learning. Brain Stimul 4:137-144.
Reid PD, Daniels B, Rybak M, Turnier-Shea Y, Pridmore S (2002) Cortical excitability of
psychiatric disorders: reduced post-exercise facilitation in depression compared to
schizophrenia and controls. Aust N Z J Psychiatry 36:669-673.
Rice DS, Curran T (2001) Role of the reelin signaling pathway in central nervous system
development. Annu Rev Neurosci 24:1005-1039.
Richter MA, de Jesus DR, Hoppenbrouwers S, Daigle M, Deluce J, Ravindran LN,
Fitzgerald PB, Daskalakis ZJ (2012) Evidence for cortical inhibitory and excitatory
190
dysfunction in obsessive compulsive disorder. Neuropsychopharmacology 37:1144-
1151.
Robol E, Fiaschi A, Manganotti P (2004) Effects of citalopram on the excitability of the
human motor cortex: a paired magnetic stimulation study. J Neurol Sci 221:41-46.
Rogasch NC, Daskalakis ZJ, Fitzgerald PB (2013) Mechanisms underlying long-interval
cortical inhibition in the human motor cortex: a TMS-EEG study. J Neurophysiol
109:89-98.
Rogasch NC, Thomson RH, Farzan F, Fitzgibbon BM, Bailey NW, Hernandez-Pavon JC,
Daskalakis ZJ, Fitzgerald PB (2014) Removing artefacts from TMS-EEG recordings
using independent component analysis: importance for assessing prefrontal and motor
cortex network properties. Neuroimage 101:425-439.
Roick H, von Giesen HJ, Benecke R (1993) On the origin of the postexcitatory inhibition
seen after transcranial magnetic brain stimulation in awake human subjects. Exp
Brain Res 94:489-498.
Rosanova M, Casali A, Bellina V, Resta F, Mariotti M, Massimini M (2009) Natural
frequencies of human corticothalamic circuits. J Neurosci 29:7679-7685.
Rosanova M, Gosseries O, Casarotto S, Boly M, Casali AG, Bruno MA, Mariotti M,
Boveroux P, Tononi G, Laureys S, Massimini M (2012) Recovery of cortical
effective connectivity and recovery of consciousness in vegetative patients. Brain
135:1308-1320.
Rosburg T, Boutros NN, Ford JM (2008) Reduced auditory evoked potential component
N100 in schizophrenia--a critical review. Psychiatry Res 161:259-274.
Rossi S, Hallett M, Rossini PM, Pascual-Leone A (2009) Safety, ethical considerations, and
application guidelines for the use of transcranial magnetic stimulation in clinical
practice and research. Clin Neurophysiol 120:2008-2039.
Rossini PM, Barker AT, Berardelli A, Caramia MD, Caruso G, Cracco RQ, Dimitrijevic MR,
Hallett M, Katayama Y, Lucking CH, et al. (1994) Non-invasive electrical and
magnetic stimulation of the brain, spinal cord and roots: basic principles and
procedures for routine clinical application. Report of an IFCN committee.
Electroencephalogr Clin Neurophysiol 91:79-92.
191
Roth BJ, Pascual-Leone A, Cohen LG, Hallett M (1992) The heating of metal electrodes
during rapid-rate magnetic stimulation: a possible safety hazard. Electroencephalogr
Clin Neurophysiol 85:116-123.
Rowland LM, Kontson K, West J, Edden RA, Zhu H, Wijtenburg SA, Holcomb HH, Barker
PB (2013) In vivo measurements of glutamate, GABA, and NAAG in schizophrenia.
Schizophr Bull 39:1096-1104.
Rusjan PM, Barr MS, Farzan F, Arenovich T, Maller JJ, Fitzgerald PB, Daskalakis ZJ (2010)
Optimal transcranial magnetic stimulation coil placement for targeting the
dorsolateral prefrontal cortex using novel magnetic resonance image-guided
neuronavigation. Hum Brain Mapp 31:1643-1652.
Saka MC, Atbasoglu EC, Ozguven HD, Sener HO, Ozay E (2005) Cortical inhibition in first-
degree relatives of schizophrenic patients assessed with transcranial magnetic
stimulation. Int J Neuropsychopharmacol 8:595-599.
Samuels J, Wang Y, Riddle MA, Greenberg BD, Fyer AJ, McCracken JT, Rauch SL,
Murphy DL, Grados MA, Knowles JA, Piacentini J, Cullen B, Bienvenu OJ, 3rd,
Rasmussen SA, Geller D, Pauls DL, Liang KY, Shugart YY, Nestadt G (2011)
Comprehensive family-based association study of the glutamate transporter gene
SLC1A1 in obsessive-compulsive disorder. Am J Med Genet B Neuropsychiatr Genet
156B:472-477.
Sanacora G, Saricicek A (2007) GABAergic contributions to the pathophysiology of
depression and the mechanism of antidepressant action. CNS Neurol Disord Drug
Targets 6:127-140.
Sanchez-Morla EM, Santos JL, Aparicio A, Garcia-Jimenez MA, Villanueva C, Martinez-
Vizcaino V, Arango C (2009) Antipsychotic effects on auditory sensory gating in
schizophrenia patients. Eur Neuropsychopharmacol 19:905-909.
Sanger TD, Garg RR, Chen R (2001) Interactions between two different inhibitory systems
in the human motor cortex. J Physiol 530:307-317.
Scanziani M (2000) GABA spillover activates postsynaptic GABA(B) receptors to control
rhythmic hippocampal activity. Neuron 25:673-681.
Schieber MH, Hibbard LS (1993) How somatotopic is the motor cortex hand area? Science
261:489-492.
192
Schork NJ, Schork MA (1993) The relative efficiency and power of small-pedigree studies of
the heritability of a quantitative trait. Hum Hered 43:1-11.
Schwenkreis P, Witscher K, Janssen F, Addo A, Dertwinkel R, Zenz M, Malin JP,
Tegenthoff M (1999) Influence of the N-methyl-D-aspartate antagonist memantine on
human motor cortex excitability. Neurosci Lett 270:137-140.
Shajahan PM, Glabus MF, Gooding PA, Shah PJ, Ebmeier KP (1999) Reduced cortical
excitability in depression. Impaired post-exercise motor facilitation with transcranial
magnetic stimulation. Br J Psychiatry 174:449-454.
Siebner HR, Dressnandt J, Auer C, Conrad B (1998) Continuous intrathecal baclofen
infusions induced a marked increase of the transcranially evoked silent period in a
patient with generalized dystonia. Muscle Nerve 21:1209-1212.
Silvanto J, Cattaneo Z, Battelli L, Pascual-Leone A (2008a) Baseline cortical excitability
determines whether TMS disrupts or facilitates behavior. J Neurophysiol 99:2725-
2730.
Silvanto J, Muggleton N, Walsh V (2008b) State-dependency in brain stimulation studies of
perception and cognition. Trends Cogn Sci 12:447-454.
Silvanto J, Muggleton NG (2008) Testing the validity of the TMS state-dependency
approach: targeting functionally distinct motion-selective neural populations in visual
areas V1/V2 and V5/MT+. Neuroimage 40:1841-1848.
Silvanto J, Muggleton NG, Cowey A, Walsh V (2007) Neural adaptation reveals state-
dependent effects of transcranial magnetic stimulation. Eur J Neurosci 25:1874-1881.
Silvanto J, Pascual-Leone A (2008) State-dependency of transcranial magnetic stimulation.
Brain Topogr 21:1-10.
Skudlarski P, Schretlen DJ, Thaker GK, Stevens MC, Keshavan MS, Sweeney JA,
Tamminga CA, Clementz BA, O'Neil K, Pearlson GD (2013) Diffusion tensor
imaging white matter endophenotypes in patients with schizophrenia or psychotic
bipolar disorder and their relatives. Am J Psychiatry 170:886-898.
Sohal VS, Zhang F, Yizhar O, Deisseroth K (2009) Parvalbumin neurons and gamma
rhythms enhance cortical circuit performance. Nature 459:698-702.
Somogyi P, Tamas G, Lujan R, Buhl EH (1998) Salient features of synaptic organisation in
the cerebral cortex. Brain Res Brain Res Rev 26:113-135.
193
Soubasi E, Chroni E, Gourzis P, Zisis A, Beratis S, Papathanasopoulos P (2010) Cortical
motor neurophysiology of patients with schizophrenia: a study using transcranial
magnetic stimulation. Psychiatry Res 176:132-136.
Spencer KM, Nestor PG, Niznikiewicz MA, Salisbury DF, Shenton ME, McCarley RW
(2003) Abnormal neural synchrony in schizophrenia. J Neurosci 23:7407-7411.
Sponheim SR, Clementz BA, Iacono WG, Beiser M (2000) Clinical and biological
concomitants of resting state EEG power abnormalities in schizophrenia. Biol
Psychiatry 48:1088-1097.
Stan A, Lewis DA (2012) Altered Cortical GABA Neurotransmission in Schizophrenia:
Insights into Novel Therapeutic Strategies. Curr Pharm Biotechnol.
Steele JD, Glabus MF, Shajahan PM, Ebmeier KP (2000) Increased cortical inhibition in
depression: a prolonged silent period with transcranial magnetic stimulation (TMS).
Psychol Med 30:565-570.
Stefan K, Kunesch E, Cohen LG, Benecke R, Classen J (2000) Induction of plasticity in the
human motor cortex by paired associative stimulation. Brain 123 Pt 3:572-584.
Stein DJ (2002) Obsessive-compulsive disorder. Lancet 360:397-405.
Stephan KE, Baldeweg T, Friston KJ (2006) Synaptic plasticity and dysconnection in
schizophrenia. Biol Psychiatry 59:929-939.
Stewart SE, Platko J, Fagerness J, Birns J, Jenike E, Smoller JW, Perlis R, Leboyer M,
Delorme R, Chabane N, Rauch SL, Jenike MA, Pauls DL (2007) A genetic family-
based association study of OLIG2 in obsessive-compulsive disorder. Arch Gen
Psychiatry 64:209-214.
Stone JM, Day F, Tsagaraki H, Valli I, McLean MA, Lythgoe DJ, O'Gorman RL, Barker GJ,
McGuire PK (2009) Glutamate dysfunction in people with prodromal symptoms of
psychosis: relationship to gray matter volume. Biol Psychiatry 66:533-539.
Sullivan PF (2005) The genetics of schizophrenia. PLoS Med 2:e212.
Sumich A, Harris A, Flynn G, Whitford T, Tunstall N, Kumari V, Brammer M, Gordon E,
Williams LM (2006) Event-related potential correlates of depression, insight and
negative symptoms in males with recent-onset psychosis. Clin Neurophysiol
117:1715-1727.
194
Swartz BE, Goldensohn ES (1998) Timeline of the history of EEG and associated fields.
Electroencephalogr Clin Neurophysiol 106:173-176.
Swerdlow NR, Light GA, Sprock J, Calkins ME, Green MF, Greenwood TA, Gur RE, Gur
RC, Lazzeroni LC, Nuechterlein KH, Radant AD, Ray A, Seidman LJ, Siever LJ,
Silverman JM, Stone WS, Sugar CA, Tsuang DW, Tsuang MT, Turetsky BI, Braff
DL (2014) Deficient prepulse inhibition in schizophrenia detected by the multi-site
COGS. Schizophr Res 152:503-512.
Tallon-Baudry C, Bertrand O (1999) Oscillatory gamma activity in humans and its role in
object representation. Trends Cogn Sci 3:151-162.
Tamminga CA, Ivleva EI, Keshavan MS, Pearlson GD, Clementz BA, Witte B, Morris DW,
Bishop J, Thaker GK, Sweeney JA (2013) Clinical Phenotypes of Psychosis in the
Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP). Am J
Psychiatry 170:1263-1274.
Tan HY, Callicott JH, Weinberger DR (2008) Intermediate phenotypes in schizophrenia
genetics redux: is it a no brainer? Mol Psychiatry 13:233-238.
Taylor PC, Walsh V, Eimer M (2008) Combining TMS and EEG to study cognitive function
and cortico-cortico interactions. Behav Brain Res 191:141-147.
Tayoshi S, Nakataki M, Sumitani S, Taniguchi K, Shibuya-Tayoshi S, Numata S, Iga J, Ueno
S, Harada M, Ohmori T (2010) GABA concentration in schizophrenia patients and
the effects of antipsychotic medication: a proton magnetic resonance spectroscopy
study. Schizophr Res 117:83-91.
Tergau F, Wanschura V, Canelo M, Wischer S, Wassermann EM, Ziemann U, Paulus W
(1999) Complete suppression of voluntary motor drive during the silent period after
transcranial magnetic stimulation. Exp Brain Res 124:447-454.
Thomas C, vom Berg I, Rupp A, Seidl U, Schroder J, Roesch-Ely D, Kreisel SH, Mundt C,
Weisbrod M (2010) P50 gating deficit in Alzheimer dementia correlates to frontal
neuropsychological function. Neurobiol Aging 31:416-424.
Thompson SM, Gahwiler BH (1992) Effects of the GABA uptake inhibitor tiagabine on
inhibitory synaptic potentials in rat hippocampal slice cultures. J Neurophysiol
67:1698-1701.
195
Thut G, Northoff G, Ives JR, Kamitani Y, Pfennig A, Kampmann F, Schomer DL, Pascual-
Leone A (2003) Effects of single-pulse transcranial magnetic stimulation (TMS) on
functional brain activity: a combined event-related TMS and evoked potential study.
Clin Neurophysiol 114:2071-2080.
Tibbo P, Kroetsch M, Chue P, Warneke L (2000) Obsessive-compulsive disorder in
schizophrenia. J Psychiatr Res 34:139-146.
Tibbo P, Warneke L (1999) Obsessive-compulsive disorder in schizophrenia: epidemiologic
and biologic overlap. J Psychiatry Neurosci 24:15-24.
Torres AR, Prince MJ, Bebbington PE, Bhugra D, Brugha TS, Farrell M, Jenkins R, Lewis
G, Meltzer H, Singleton N (2006) Obsessive-compulsive disorder: prevalence,
comorbidity, impact, and help-seeking in the British National Psychiatric Morbidity
Survey of 2000. Am J Psychiatry 163:1978-1985.
Torrey EF, Barci BM, Webster MJ, Bartko JJ, Meador-Woodruff JH, Knable MB (2005)
Neurochemical markers for schizophrenia, bipolar disorder, and major depression in
postmortem brains. Biol Psychiatry 57:252-260.
Tot S, Ozge A, Comelekoglu U, Yazici K, Bal N (2002) Association of QEEG findings with
clinical characteristics of OCD: evidence of left frontotemporal dysfunction. Can J
Psychiatry 47:538-545.
Toulopoulou T, Picchioni M, Rijsdijk F, Hua-Hall M, Ettinger U, Sham P, Murray R (2007)
Substantial genetic overlap between neurocognition and schizophrenia: genetic
modeling in twin samples. Arch Gen Psychiatry 64:1348-1355.
Traub RD, Jefferys JG, Whittington MA (1997) Simulation of gamma rhythms in networks
of interneurons and pyramidal cells. J Comput Neurosci 4:141-150.
Traub RD, Whittington MA, Colling SB, Buzsaki G, Jefferys JG (1996) Analysis of gamma
rhythms in the rat hippocampus in vitro and in vivo. J Physiol 493 ( Pt 2):471-484.
Turetsky BI, Calkins ME, Light GA, Olincy A, Radant AD, Swerdlow NR (2007)
Neurophysiological endophenotypes of schizophrenia: the viability of selected
candidate measures. Schizophr Bull 33:69-94.
Turetsky BI, Greenwood TA, Olincy A, Radant AD, Braff DL, Cadenhead KS, Dobie DJ,
Freedman R, Green MF, Gur RE, Gur RC, Light GA, Mintz J, Nuechterlein KH,
Schork NJ, Seidman LJ, Siever LJ, Silverman JM, Stone WS, Swerdlow NR, Tsuang
196
DW, Tsuang MT, Calkins ME (2008) Abnormal auditory N100 amplitude: a heritable
endophenotype in first-degree relatives of schizophrenia probands. Biol Psychiatry
64:1051-1059.
Uhlhaas PJ, Haenschel C, Nikolic D, Singer W (2008) The role of oscillations and synchrony
in cortical networks and their putative relevance for the pathophysiology of
schizophrenia. Schizophr Bull 34:927-943.
Uhlhaas PJ, Linden DE, Singer W, Haenschel C, Lindner M, Maurer K, Rodriguez E (2006)
Dysfunctional long-range coordination of neural activity during Gestalt perception in
schizophrenia. J Neurosci 26:8168-8175.
Uhlhaas PJ, Singer W (2010) Abnormal neural oscillations and synchrony in schizophrenia.
Nat Rev Neurosci 11:100-113.
Valls-Sole J, Pascual-Leone A, Wassermann EM, Hallett M (1992) Human motor evoked
responses to paired transcranial magnetic stimuli. Electroencephalogr Clin
Neurophysiol 85:355-364.
van den Heuvel OA, Veltman DJ, Groenewegen HJ, Cath DC, van Balkom AJ, van
Hartskamp J, Barkhof F, van Dyck R (2005) Frontal-striatal dysfunction during
planning in obsessive-compulsive disorder. Arch Gen Psychiatry 62:301-309.
van Os J, Kapur S (2009) Schizophrenia. Lancet 374:635-645.
Venables NC, Bernat EM, Sponheim SR (2009) Genetic and disorder-specific aspects of
resting state EEG abnormalities in schizophrenia. Schizophr Bull 35:826-839.
Venkatasubramanian G, Rao NP, Behere RV (2009) Neuroanatomical, neurochemical, and
neurodevelopmental basis of obsessive-compulsive symptoms in schizophrenia.
Indian J Psychol Med 31:3-10.
Virtanen J, Ruohonen J, Naatanen R, Ilmoniemi RJ (1999) Instrumentation for the
measurement of electric brain responses to transcranial magnetic stimulation. Med
Biol Eng Comput 37:322-326.
Voineskos AN, Farzan F, Barr MS, Lobaugh NJ, Mulsant BH, Chen R, Fitzgerald PB,
Daskalakis ZJ (2010) The role of the corpus callosum in transcranial magnetic
stimulation induced interhemispheric signal propagation. Biol Psychiatry 68:825-831.
197
Volk D, Austin M, Pierri J, Sampson A, Lewis D (2001) GABA transporter-1 mRNA in the
prefrontal cortex in schizophrenia: decreased expression in a subset of neurons. Am J
Psychiatry 158:256-265.
Volk DW, Austin MC, Pierri JN, Sampson AR, Lewis DA (2000) Decreased glutamic acid
decarboxylase67 messenger RNA expression in a subset of prefrontal cortical
gamma-aminobutyric acid neurons in subjects with schizophrenia. Arch Gen
Psychiatry 57:237-245.
Voyiaziakis E, Evgrafov O, Li D, Yoon HJ, Tabares P, Samuels J, Wang Y, Riddle MA,
Grados MA, Bienvenu OJ, Shugart YY, Liang KY, Greenberg BD, Rasmussen SA,
Murphy DL, Wendland JR, McCracken JT, Piacentini J, Rauch SL, Pauls DL,
Nestadt G, Fyer AJ, Knowles JA (2011) Association of SLC6A4 variants with
obsessive-compulsive disorder in a large multicenter US family study. Mol
Psychiatry 16:108-120.
Wagner T, Valero-Cabre A, Pascual-Leone A (2007) Noninvasive human brain stimulation.
Annu Rev Biomed Eng 9:527-565.
Waldo MC, Adler LE, Freedman R (1988) Defects in auditory sensory gating and their
apparent compensation in relatives of schizophrenics. Schizophr Res 1:19-24.
Wang XJ, Buzsaki G (1996) Gamma oscillation by synaptic inhibition in a hippocampal
interneuronal network model. J Neurosci 16:6402-6413.
Weinberger DR, Berman KF, Zec RF (1986) Physiologic dysfunction of dorsolateral
prefrontal cortex in schizophrenia. I. Regional cerebral blood flow evidence. Arch
Gen Psychiatry 43:114-124.
Werhahn KJ, Kunesch E, Noachtar S, Benecke R, Classen J (1999a) Differential effects on
motorcortical inhibition induced by blockade of GABA uptake in humans. J Physiol
517 ( Pt 2):591-597.
Werhahn KJ, Kunesch E, Noachtar S, Benecke R, Classen J (1999b) Differential effects on
motorcortical inhibition induced by blockade of GABA uptake in humans. J Physiol
(Lond) 517:591-597.
Whiteside SP, Port JD, Deacon BJ, Abramowitz JS (2006) A magnetic resonance
spectroscopy investigation of obsessive-compulsive disorder and anxiety. Psychiatry
Res 146:137-147.
198
Whitham EM, Lewis T, Pope KJ, Fitzgibbon SP, Clark CR, Loveless S, DeLosAngeles D,
Wallace AK, Broberg M, Willoughby JO (2008) Thinking activates EMG in scalp
electrical recordings. Clin Neurophysiol 119:1166-1175.
Whitham EM, Pope KJ, Fitzgibbon SP, Lewis T, Clark CR, Loveless S, Broberg M, Wallace
A, DeLosAngeles D, Lillie P, Hardy A, Fronsko R, Pulbrook A, Willoughby JO
(2007) Scalp electrical recording during paralysis: quantitative evidence that EEG
frequencies above 20 Hz are contaminated by EMG. Clin Neurophysiol 118:1877-
1888.
Whittington MA, Traub RD, Jefferys JG (1995) Synchronized oscillations in interneuron
networks driven by metabotropic glutamate receptor activation. Nature 373:612-615.
Whittington MA, Traub RD, Kopell N, Ermentrout B, Buhl EH (2000) Inhibition-based
rhythms: experimental and mathematical observations on network dynamics. Int J
Psychophysiol 38:315-336.
Wilson FA, O'Scalaidhe SP, Goldman-Rakic PS (1994) Functional synergism between
putative gamma-aminobutyrate-containing neurons and pyramidal neurons in
prefrontal cortex. Proc Natl Acad Sci U S A 91:4009-4013.
Winterer G, Egan MF, Radler T, Coppola R, Weinberger DR (2001) Event-related potentials
and genetic risk for schizophrenia. Biol Psychiatry 50:407-417.
Wobrock T, Hasan A, Malchow B, Wolff-Menzler C, Guse B, Lang N, Schneider-Axmann
T, Ecker UK, Falkai P (2010) Increased cortical inhibition deficits in first-episode
schizophrenia with comorbid cannabis abuse. Psychopharmacology (Berl) 208:353-
363.
Wobrock T, Schneider-Axmann T, Retz W, Rosler M, Kadovic D, Falkai P, Schneider M
(2009) Motor circuit abnormalities in first-episode schizophrenia assessed with
transcranial magnetic stimulation. Pharmacopsychiatry 42:194-201.
Wobrock T, Schneider M, Kadovic D, Schneider-Axmann T, Ecker UK, Retz W, Rosler M,
Falkai P (2008) Reduced cortical inhibition in first-episode schizophrenia. Schizophr
Res 105:252-261.
Woo TU, Walsh JP, Benes FM (2004) Density of glutamic acid decarboxylase 67 messenger
RNA-containing neurons that express the N-methyl-D-aspartate receptor subunit
199
NR2A in the anterior cingulate cortex in schizophrenia and bipolar disorder. Arch
Gen Psychiatry 61:649-657.
Woods SW (2003) Chlorpromazine equivalent doses for the newer atypical antipsychotics. J
Clin Psychiatry 64:663-667.
Wu Y, Blichowski M, Daskalakis ZJ, Wu Z, Liu CC, Cortez MA, Snead OC, 3rd (2011)
Evidence that clozapine directly interacts on the GABAB receptor. Neuroreport
22:637-641.
Yamanaka K, Kadota H, Nozaki D (2013) Long-latency TMS-evoked potentials during
motor execution and inhibition. Front Hum Neurosci 7:1-9.
Yizhar O, Fenno LE, Prigge M, Schneider F, Davidson TJ, O'Shea DJ, Sohal VS, Goshen I,
Finkelstein J, Paz JT, Stehfest K, Fudim R, Ramakrishnan C, Huguenard JR,
Hegemann P, Deisseroth K (2011) Neocortical excitation/inhibition balance in
information processing and social dysfunction. Nature 477:171-178.
Yoon JH, Maddock RJ, Rokem A, Silver MA, Minzenberg MJ, Ragland JD, Carter CS
(2010) GABA concentration is reduced in visual cortex in schizophrenia and
correlates with orientation-specific surround suppression. J Neurosci 30:3777-3781.
Zaaroor M, Pratt H, Starr A (2003) Time course of motor excitability before and after a task-
related movement. Neurophysiol Clin 33:130-137.
Zai G, Arnold P, Burroughs E, Barr CL, Richter MA, Kennedy JL (2005) Evidence for the
gamma-amino-butyric acid type B receptor 1 (GABBR1) gene as a susceptibility
factor in obsessive-compulsive disorder. Am J Med Genet B Neuropsychiatr Genet
134B:25-29.
Ziemann U, Chen R, Cohen LG, Hallett M (1998) Dextromethorphan decreases the
excitability of the human motor cortex. Neurology 51:1320-1324.
Ziemann U, Lonnecker S, Steinhoff BJ, Paulus W (1996a) The effect of lorazepam on the
motor cortical excitability in man. Exp Brain Res 109:127-135.
Ziemann U, Lonnecker S, Steinhoff BJ, Paulus W (1996b) Effects of antiepileptic drugs on
motor cortex excitability in humans: a transcranial magnetic stimulation study. Ann
Neurol 40:367-378.
200
Ziemann U, Paulus W, Rothenberger A (1997a) Decreased motor inhibition in Tourette's
disorder: evidence from transcranial magnetic stimulation. Am J Psychiatry
154:1277-1284.
Ziemann U, Rothwell JC, Ridding MC (1996c) Interaction between intracortical inhibition
and facilitation in human motor cortex. J Physiol 496 ( Pt 3):873-881.
Ziemann U, Tergau F, Bruns D, Baudewig J, Paulus W (1997b) Changes in human motor
cortex excitability induced by dopaminergic and anti-dopaminergic drugs.
Electroencephalogr Clin Neurophysiol 105:430-437.