1 of 223

Towards neurophysiological biomarkers to assess the repetitive transcranial magnetic stimulation treatment for patients with

schizophrenia and auditory verbal hallucinations

PhD in Clinical Neurophysiology, Dissertation

PhD Candidate: Examiner:

Ovidiu C. Banea, MD Giorgio Di Lorenzo, MD, PhD

Department of Engineering, Reykjavik University Laboratory of Psychophysiology and Cognitive Neuroscience

Clinical Neurophysiology Unit, Landspítali Associate Professor and Chair of Psychiatry

Department of Systems Medicine

University of Rome Tor Vergata, Italy

Place: Reykjavík University, Room M35

Date: 31st of January 2022

2 of 223

Summary

PART ONE

  • Schizophrenia & AVH
    • History, Biomarker & Endophenotype
  • TMS

PART TWO

  • The present work
  • Studies
  • Conclusions
  • Ecology of the brain

3 of 223

Schizophrenia

Chronic and severe mental disorder affecting approximately 24 million people or 1 in 300 people worldwide (WHO, 2022). Median incidence: 15.2 cases per 100,000 persons (McGrath et al, 2004). Lifetime prevalence 0.5-1% (Keshavan et al, 2019)

In Iceland, during the period 2012-2018, a total number of 567 patients were diagnosed with SCZ (Jónasson et al, 2019) 348,450 inhabitants (2018).

Multifactorial origin: It is thought that an interaction between genes, a range of environmental and probably psychosocial factors may cause schizophrenia.

4 of 223

Schizophrenia

Onset between 16-30 year old, found more in migrant / local, urban / rural. Much higher incidence at higher latitudes (Saha et al, 2005).

Symptoms: positive, negative, and cognitive (Patel et al., 2014)

  1. Positive: People may “lose touch” with some aspects of reality. Hallucinations, Delusions, Thought disorders, Movement disorders (agitated body movements)
  2. Negative: disruptions to normal emotions and behaviors “Flat affect”, Reduced feelings of pleasure in everyday life, Difficulty beginning and sustaining activities, reduced speaking

5 of 223

Schizophrenia

  1. Cognitive:
    • Poor “executive functioning” (the ability to understand information and use it to make decisions).
    • Problems with “working memory” (the ability to use information immediately after learning it)
    • Trouble focusing or paying attention (sensory gating dysfunction).

6 of 223

History (The History of Schizophrenia | Pasadena Villa, n.d.)

Word schizophrenia comes from the Greek words schizo = split, and phren = mind, to describe fragmented thinking

  • 1st century Celsus (25 BC-50 AD), Roman encyclopedist known for the medical work De Medicina believed in the idea that madness was punishment from the Gods Insania.
  • XIV-XVII centuries: Burn of heretics (doctrine considered false by ecclesiastical authority) and witches (Copeman et al., 1966)
  • 1563 In the book “De Praestigiis DaemonumOn the Tricks of Demons Dutch physician Johann Weyer, argued that the madness resulted from natural causes, not divine punishment or demonic possession.

7 of 223

History (The History of Schizophrenia | Pasadena Villa, n.d.)

  • XVI England, monasteries and hospitals were gradually converted into asylums, which later became tourist attractions (patients mentally ill).
  • 1893 Dementia praecox was a term given to young patients with cognitive deterioration by German psychiatrist Emil Kraepelin (disruption in cognitive or mental functioning in attention, memory, and goal-directed behavior). In 1904 he admitted that some patients recover from dementia praecox.
  • 1908 Swiss psychiatrist Eugen Bleuler introduced the term "schizophrenia" in a Berlin lecture (Fusar-Poli & Politi, 2008). He also made the difference between basic symptoms and the accessory symptoms, primary symptoms and secondary symptoms.
  • 1950s The first antipsychotic drug, chlorpromazine became available.

8 of 223

Between physical disease and phenomenology

  • Kraepelin (1890s): product of neural processes and physical origin progressive disorder and described deteriorated brain cells that underlie the disorder”

  • Bleuler (1911): “physical disease process but considered it “not absolutely necessary.” He agreed with Freud, Jung, and Janet, that the understanding of schizophrenia requires the study of unconscious psychological processes

  • Karl Jaspers (1913): “General Psychopathology” analyzing a symptom by form rather than by their content, biographical understanding and phenomenological approach, distinction between neurosis and psychosis

  • McGhie and Chapman (1961): proposed that symptoms of schizophrenia indicate a primary deficit in mechanisms of attention related to selection and inhibition (McGHIE & CHAPMAN, 1961)

9 of 223

Schizophrenia & Sensory Gating

  • Sensory gating describes neurological processes of filtering out redundant or unnecessary stimuli in the brain from all possible environmental stimuli.

  • Psychopathology may be ‘‘the result of abnormalities in filtering stimuli, focusing attention, or sensory gating.’’ (Andreasen et al., 1994)

  • Contemporary sensory gating definitions are generally tied to the perceptual and attentional phenomenology described by McGhie and Chapman (1961) (Hetrick et al., 2012)

10 of 223

How to quantify Sensory Gating?

  • Psychometrically sound self-report rating scale such as Sensory Gating Inventory (SGI) (Hetrick et al., 2012)
  • Using auditory evoked potentials (P50 suppression)
  • MRI or PET scan

Contrary to expectations, schizophrenic patients who reported perceptual anomalies exhibited P50 ratios that did not differ from normal controls.

11 of 223

Schizophrenia: Structural findings: THALAMUS (MRI)

MRI from 39 male patients and 47 male controls were transformed with a "bounding box" to produce an "average schizophrenic brain" and an "average normal brain." Image subtraction –

Specific regional abnormalities were observed in the (R) thalamus and adjacent white matter. Interpreted as result from a defect in filtering or gating sensory input, which is one of the primary functions of the thalamus in the human brain.

Source: (Andreasen et al., 1994) Science

12 of 223

Schizophrenia: Structural findings: THALAMUS (PET)

Patients with SCZ (20 never medicated) showed a diminished metabolic rate in the right thalamus, with a loss of the normal pattern of right greater than left asymmetry (compared with 15 HC).

The area of the thalamus was smaller in the patients than in the volunteers, and this difference was greatest in the left anterior region.

Deficit in sensory filtering in schizophrenia. (Buchsbaum et al., 1996)

13 of 223

14 of 223

Biomarker and endophenotype

The research efforts of Psychiatry must be equally focused on biological processes, psychopathological experiences, psychological - biographical connections, and social interactions,

Instead of being reduced to analyzing processes inside the brain (Thomas Fuchs, 2021)

15 of 223

Biomarker - definition (FDA-NIH Biomarker Group 2016)

  • A defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or responses to an exposure or intervention, including therapeutic interventions.

  • TYPES
    • Histologic, molecular (glucose), radiographic (tumor size), and physiologic (blood pressure) characteristics.

16 of 223

Biomarkers

  • Biomarkers have three general functions: to improve diagnostic procedure as diagnostic biomarkers, to predict development of the disease as prognostic biomarkers (predisposition screening), and to be used in therapy monitoring as theranostic biomarkers (Perkovic et al., 2017).

  • Unfortunately for clinicians, there is no (and there will likely never be) quantitative measure of any neurochemical measure in the blood to be able to support a diagnosis of psychosis because of elevation or deficits in any neurotransmitter (DeLisi, 2020).

17 of 223

Biomarkers

  • In a 2012 consensus report, the APA Work Group on Neuroimaging Markers of Psychiatric Disorders suggested several criteria that should be met in order to establish the validity of a neuroimaging biomarker. (Kraguljac et al., 2021)

  • A diagnostic biomarker should have
    • Sensitivity .80% in detecting a particular psychiatric disorder,
    • Specificity .80% in distinguishing this disorder from other psychiatric disorders, and a PPV that approaches 90%.
    • The data used to establish a biomarker should require confirmation by at least two independent sets of qualified investigators, with results published in peer-reviewed journals

18 of 223

Endophenotype - definition

  • Biomarkers that are genetically correlated with disease liability, can be measured in all individuals (both affected and unaffected), and that provide greater power to identify disease-related genes than does disease “yes/no” status alone

  • The term was coined in 1966 and applied in psychiatry by Gottesman and Shields in 1972 (Kropotov, 2009)

  • The term “biological marker” should be used to signify differences that do not have genetic underpinnings and “endophenotype” when certain heritability indicators are fulfilled (Gottesman & Gould, 2003)

19 of 223

Biomarkers and endophenotype candidates

Neuroimage biomarkers (fMRI, MRS): Dopamine hyperactivity, NMDA receptor hypofunction, hippocampal hyperactivity, immune dysregulation, dysconnectivity, cortical gray matter volume loss (Kraguljac et al., 2021)

Neurophysiological Biomarker: TMS as potential method by which sensory processing can be assessed, since TMS paradigms (CSP) can be used to measure GABAB-mediated cortical inhibition that is linked with sensory gating (Kim et al., 2020).

Endophenotypes: Assessment of P50 suppression and PPI of the startle response, eye tracking dysfunction (smooth pursuit), working memory and cognition by dysfunction and abnormality found at DLPFC level (Gottesman & Gould, 2003) , QEEG and spectral analysis (Kropotov, 2009).

20 of 223

Spectral EEG endophenotype (Kropotov, 2009)

7 days test-retest reliability (left) and heritability of spectral EEG in monozigotic versus dizigotic twins (right) (Kropotov, 2009).

21 of 223

Neuroanatomy (MRI)

Despite that both Kraepelin and Bleuler were convinced that schizophrenia would ultimately be linked to an organic brain disorder the neuropathology of SCZ remains difficult to understand.

  • At the onset of illness, acquiring an MRI scan could be part of the routine evaluation to determine the disease progression.

  • The practice is not recognized by the American Psychiatric Association in any of its guidelines on the treatment of schizophrenia (DeLisi, 2020).

22 of 223

Neuroanatomy (MRI)

Most studies have shown a grey matter volume reduction of STG but not localized to a particular type of cortex, i.e.,

Both the allocortex including the hippocampus and the isocortex including the superior temporal gyrus (STG) are affected (Psychology and Schizophrenia - Janet E. Pletson - Google Books, n.d.)

23 of 223

Left temporoparietal site T3P3 (TPJ)

The dorsal edge of the principal sulcus,

- mainly the supramarginal gyrus (BA40)

-The angular gyrus (BA39)

-Superior temporal cortex (BA 22)

(Wernicke area) language reception and processing

STG includes primary (Heschl's gyrus) and secondary auditory cortices, processing auditory signals, and the planum temporale (PT) in the posterior STG is a language-related area.

Source: Brain Anatomy (Columbia University, NY)

24 of 223

Network disease

  • The fMRI studies suggest that for any given task that is performed poorly by individuals with schizophrenia, there is a network of affected brain regions related to the abnormal function, rather than a single abnormal brain region, raising the issue of the state of the interconnections between regions (Ross et al., 2006).

25 of 223

Diagnostic of SCZ, ICD-11 (2018) January 2022

  • New ICD-11 code proposed for schizophrenia is 6A20. Exclusions are Schizotypal disorder (6A22), Schizophrenic reaction (6A22), and Acute and transient psychotic disorder (6A23)

  • ICD-11 made 3 changes in the characterization of schizophrenia from ICD-10:
    • Removed the subtypes of schizophrenia from ICD-10
    • Introduced a symptom specifier which records information on the presence or absence of symptoms, their longitudinal course, response to treatment and prognosis in the disorder
    • Modified the ICD-10 schizophrenia course specifier with “first episode”, multiple episodes, continuous course and unspecified (Valle, 2020).

26 of 223

Diagnostic of SCZ, ICD-11 (2018) January 2022

  • Disturbances in multiple mental modalities, including thinking, perception (e.g., hallucinations), self experience (e.g., the experience that one's impulses, thoughts, or behavior are under the control of an external force), cognition (e.g., impaired attention, verbal memory), volition (e.g., loss of motivation), affect, and behavior.

  • Psychomotor disturbances, including catatonia, may be present.

  • Persistent delusions, persistent hallucinations, thought disorder, and experiences of influence, passivity, or control are considered core symptoms. Symptoms must have persisted for at least one month for a diagnosis of schizophrenia to be assigned. DSM-5: Some signs of the disorder must last for a continuous period of at least 6 months.

  • The symptoms are not a manifestation of another health condition (e.g., a brain tumor) and are not because of a substance or medication on the central nervous system (e.g., corticosteroids), including withdrawal (e.g., alcohol withdrawal).

27 of 223

Auditory Verbal Hallucinations

Auditory verbal hallucinations is the most common positive symptom in schizophrenia.

The first treatment option for hallucinations in schizophrenia is antipsychotic medication, which can induce a rapid decrease in severity.

Antipsychotic medication does not ameliorate AVH in 20-30% of patients.

28 of 223

"Paradoxical" brain activation in relation to AVHs (Kompus et al, 2011).

MRI studies examining patients with schizophrenia during the processing of auditory stimuli showed overlapping:

  • increased activation in the absence of an external stimulus !

  • decreased activation in the presence of an external auditory stimulus in the left primary auditory cortex and in the right rostral prefrontal cortex

Auditory Verbal Hallucinations

29 of 223

Transcranial magnetic stimulation

TMS is a technique in which a strong pulse of electrical current is sent through a coil. When the coil is placed over a person’s skull, this induces a magnetic field pulse in a small brain area, depolarizing local neurons up to a depth of 2 cm.

When TMS is applied repetitive, it is thought to induce longer-lasting effects as a result of long-term potentiation or depression at the neuronal level.���

30 of 223

Left T3P3 STG

In 1999, Hoffman and colleagues started to explore RTMS for the treatment of AVH. When the coil was directed at the left temporoparietal cortex, they were able to ameliorate medication-resistant AVH (Hoffman 1999)

Low frequency 1Hz INHIBITORY EFFECT

31 of 223

RTMS and AVH

32 of 223

Auditory Verbal Hallucinations

Levels of Evidence:�IA Evidence from meta-analysis of randomized controlled trials�� Grades of Recommendations:A Directly based on Level I evidence���

33 of 223

Auditory Verbal Hallucinations

Levels of Evidence:

DROPPED left TP LF-RTMS in SCZ with AVH from LEVEL A (2011) to LEVEL C (2020)

NEW FRENCH GUIDELINES (Lefaucheur et al., 2020)

�“Highly controversial results were reported concerning the effect of LF-rTMS applied to the left STG/TPJ on auditory hallucinations, with as many ‘‘positive” studies showing rTMS efficacy as many ‘‘negative” studies showing RTMS inefficacy.”

“However, considering effect size calculated in various meta-analyses, literature data appeared to be in favor of a possible efficacy of LF RTMS of the left TPC on auditory hallucinations (Level C).” (Lefaucheur et al., 2020)

34 of 223

Slotema et al 2014

35 of 223

Problem

RTMS applied at the left temporoparietal area with a frequency of 1 Hz, how effective in time?

All studies on effectiveness were assessed using psychometric scales before and after treatment.

There is still no consensus for the stimulus parameter or for the effectiveness in time.

No objective measurements as neurophysiological biomarkers were assessed for this kind of treatment

2018 RTMS study to assess efficacy on first episodes (P. Homan)

Hallucination Change Scale

36 of 223

PART TWO ��Towards neurophysiological biomarkers to assess the repetitive transcranial magnetic stimulation treatment for patients with schizophrenia and auditory verbal hallucinations�

37 of 223

The aim

To determine the degree to which repetitive transcranial magnetic stimulation (RTMS) is effective for the treatment of patients with schizophrenia and persistent auditory verbal hallucinations (AVH).

Exploratory objectives

      • To determine if the symptoms of patients with schizophrenia change after intervention with 10 days of low-frequency 1Hz RTMS

      • To explore if neurophysiological tests like quantifying auditory event-related responses (P50 suppression, N100-P300 complex), EEG (electroencephalography) relative power, functional connectivity, and the cortical silent period (CSP) show changes after the RTMS treatment.

38 of 223

Subjects

Ten patients (mean age 32.4, SD = 6.85) and six HC aged between (mean age 30.3, SD = 7.5) participated in the studies.

Inclusion criteria

18-55 years of age and had treatment resistant AVH for at least 1 year (lack of clinically meaningful response to two trials of pharmacotherapy at the recommended dosage, lasting at least 6 weeks)

Exclusion criteria

History of epilepsy, cannabis or other illegal drugs use within one month prior to the study or during the study, drinking more than 3 units of alcohol daily, or RTMS contraindication

The study was approved by the Ethics Committee of the National University Hospital of Iceland (Approval No 21. 2018).

39 of 223

Methods

Clinical outcome

Quality of Life (QoL) is a self-report scale consisting of five conceptual domains of quality of life: material and physical well-being, relationships with other people, social community and civic activities, personal development, and fulfillment, and recreation. The scale maximum score is 112 (Flanagan, 1978).

Depression Anxiety Stress Scales (DASS) is a measure of mental health focusing on three traits of depression, anxiety, and stress. Maximum score is 126 (Brown et al., 1997).

Psychotic Symptom Rating Scales (PSYRATS) The score of the PSYRATS auditory hallucinations subscale (AHS) is a structured interview that measures auditory hallucinations (11 items) The maximum score of AHS is 44 (Drake et al., 2007; Haddock et al., 1999).

40 of 223

Methods

Neurophysiological assessment

  • EEG recordings in three conditions: Resting-state (RS), auditory-motor task with the left hand (AMT-l), auditory-motor task with the right hand (AMT-r)

  • Auditory paired-click paradigm

  • Oddball auditory paradigm

  • Cortical silent period

  • Cutaneous silent period

41 of 223

Protocol design

42 of 223

Methods

The S1 response was identified as the most prominent peak in the 40 to 80 ms post-stimulus windows.

The preceding negative trough was used to calculate the S1 amplitude. For the S2 response, the positive peak with latency closest to that of the S1 peak was selected.

P50 suppression was calculated as the ratio of the mean value of the S2 amplitude to the mean value of the S1 amplitude (S2:S1).

The difference of S1 minus S2 amplitude was also used as a comparison.

43 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

44 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

45 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

46 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

47 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

48 of 223

Results��P50 gating in �one patient in T2

49 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

50 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

51 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

52 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

53 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

54 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

55 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

56 of 223

57 of 223

58 of 223

59 of 223

Right Hand

Post TMS

60 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

61 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

62 of 223

Methods

The S1 response was identified as the most prominent peak in the 40 to 80 ms post-stimulus windows.

The preceding negative trough was used to calculate the S1 amplitude. For the S2 response, the positive peak with latency closest to that of the S1 peak was selected.

P50 suppression was calculated as the ratio of the mean value of the S2 amplitude to the mean value of the S1 amplitude (S2:S1).

The difference of S1 minus S2 amplitude was also used as a comparison.

63 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

64 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

65 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

66 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

67 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

68 of 223

Results��P50 gating in �one patient in T2

69 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

70 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

71 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

72 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

73 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

74 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

75 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

76 of 223

77 of 223

78 of 223

79 of 223

Right Hand

Post TMS

80 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

81 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

82 of 223

Methods

Cortical silent period (CuSP) duration was measured in sub-maximally contracted abductor pollicis brevis muscle by stimulating the motor cortex using TMS with an intensity of 140% resting motor threshold (RMT). Calculations of CSP were determined from the motor evoked potential (MEP) onset to the recovery of any voluntary EMG activity.

Three to five trials were repeated to obtain the average CSP duration for each subject.

To obtain cutaneous silent period (CuSP) electrical stimuli were delivered to the ipsilateral dermatome of the 2nd digit from the dominant hand.

Stimuli were delivered with ring electrodes at 10 times x sensory threshold. Surface EMG electrodes recorded CuSP from abductor pollicis brevis muscle while the patient was requested to continuously sustain a submaximal contraction. When the stimulus produced a clear and visible silent period, three trials were collected for each subject.

83 of 223

RESULTS CSP

84 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

85 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

86 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

87 of 223

Exploratory Study 1. Using high-density EEG to assess TMS treatment in patients with schizophrenia (Marcu et al., 2020)

Sensory gating is impaired in patients with schizophrenia (Adler et al., 1985; Olincy et al., 2010) and P300 showed decreased amplitude in patients when compared to healthy controls (Turetsky et al., 2015).

RESEARCH QUESTION

We looked mostly to reproduce paired-click, and oddball auditory paradigms in healthy controls and patients with schizophrenia and to develop a quantitative method with dense-array 256 channel EEG.

88 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

89 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

90 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

91 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

92 of 223

Results��P50 gating in �one patient in T2

93 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

94 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

95 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

96 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

97 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

98 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

99 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

100 of 223

101 of 223

102 of 223

103 of 223

Right Hand

Post TMS

104 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

105 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

106 of 223

Methods

Cortical silent period (CuSP) duration was measured in sub-maximally contracted abductor pollicis brevis muscle by stimulating the motor cortex using TMS with an intensity of 140% resting motor threshold (RMT). Calculations of CSP were determined from the motor evoked potential (MEP) onset to the recovery of any voluntary EMG activity.

Three to five trials were repeated to obtain the average CSP duration for each subject.

To obtain cutaneous silent period (CuSP) electrical stimuli were delivered to the ipsilateral dermatome of the 2nd digit from the dominant hand.

Stimuli were delivered with ring electrodes at 10 times x sensory threshold. Surface EMG electrodes recorded CuSP from abductor pollicis brevis muscle while the patient was requested to continuously sustain a submaximal contraction. When the stimulus produced a clear and visible silent period, three trials were collected for each subject.

107 of 223

RESULTS CSP

108 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

109 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

110 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

111 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

112 of 223

Results��P50 gating in �one patient in T2

113 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

114 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

115 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

116 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

117 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

118 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

119 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

120 of 223

121 of 223

122 of 223

123 of 223

Right Hand

Post TMS

124 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

125 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

126 of 223

Methods

Cortical silent period (CuSP) duration was measured in sub-maximally contracted abductor pollicis brevis muscle by stimulating the motor cortex using TMS with an intensity of 140% resting motor threshold (RMT). Calculations of CSP were determined from the motor evoked potential (MEP) onset to the recovery of any voluntary EMG activity.

Three to five trials were repeated to obtain the average CSP duration for each subject.

To obtain cutaneous silent period (CuSP) electrical stimuli were delivered to the ipsilateral dermatome of the 2nd digit from the dominant hand.

Stimuli were delivered with ring electrodes at 10 times x sensory threshold. Surface EMG electrodes recorded CuSP from abductor pollicis brevis muscle while the patient was requested to continuously sustain a submaximal contraction. When the stimulus produced a clear and visible silent period, three trials were collected for each subject.

127 of 223

RESULTS CSP

128 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

129 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

130 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

131 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

132 of 223

Results��P50 gating in �one patient in T2

133 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

134 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

135 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

136 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

137 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

138 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

139 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

140 of 223

141 of 223

142 of 223

143 of 223

Right Hand

Post TMS

144 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

145 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

146 of 223

Methods

Cortical silent period (CuSP) duration was measured in sub-maximally contracted abductor pollicis brevis muscle by stimulating the motor cortex using TMS with an intensity of 140% resting motor threshold (RMT). Calculations of CSP were determined from the motor evoked potential (MEP) onset to the recovery of any voluntary EMG activity.

Three to five trials were repeated to obtain the average CSP duration for each subject.

To obtain cutaneous silent period (CuSP) electrical stimuli were delivered to the ipsilateral dermatome of the 2nd digit from the dominant hand.

Stimuli were delivered with ring electrodes at 10 times x sensory threshold. Surface EMG electrodes recorded CuSP from abductor pollicis brevis muscle while the patient was requested to continuously sustain a submaximal contraction. When the stimulus produced a clear and visible silent period, three trials were collected for each subject.

147 of 223

RESULTS CSP

148 of 223

Methods

The regions of interest (ROI) were defined using a MATLAB script as follow: Left Anterior (LA), Left Posterior (LP), Medial Anterior (MA), Medial Central (MC), Medial Posterior (MP), Right Anterior (RA) and Right Posterior (RP)

Fifteen electrodes were selected from 3 parallel lines for each region (105 electrodes).

149 of 223

Methods

N100-P300 complex values for each ROI were calculated as the difference between the most negative voltage value and the most positive voltage value within the time range of 80-500 ms.

150 of 223

EEG pre-processing

EEG was recorded using a 256-channel system (ANT Neuro, Netherlands ) with EOG electrode placed below the right eye, and a ground electrode placed on the left side of the neck.

The data were sampled at 1024 Hz and re-referenced to the average of L and R mastoid electrodes (R19R, L19L).

A bandpass filter was set between 0.1-80 Hz and a notch filter from 49-51 Hz Bad channels were removed when EEG voltage was greater than ±80 μV; if more than 10% of the channels showed too much noise or bad signal, the whole trial was rejected.

Epochs:

P50, 500 ms starting 100 ms prior to the presentation of each auditory stimulus (-100 ms to +400 ms)

P300 epoch: 1000 ms starting 100 ms prior to the presentation of auditory stimulus (-100 to +900 ms).

Baseline correction was performed using a pre-stimulus 100 ms window.

“Bad “channels were removed and interpolated. Individual trials were visually inspected and rejected when indicative of excessive muscle activity, eye movements, or other artifacts.

151 of 223

Results

  • We observed major P50 suppression (reduced ratio) in HC compared with P50 suppression of two patients.

  • The patients showed major difference with higher ratios (less suppression) on the left anterior and left posterior regions

  • Healthy participants showed better responses over the right posterior or temporoparietal region (lower value of S2:S1 ratio).

  • S1-S2 P50 amplitude difference (method) showed more gating in patients over left anterior and left posterior regions

152 of 223

Results��P50 gating in �one patient in T2

153 of 223

Results N100-P300 baseline (HC-N4, SCZ-N5)

At baseline, healthy subjects showed better topographical representation from left and right anterior regions which are located over the frontal lobes and midline, MA, and MC regions, while the patients with schizophrenia showed better representation over LA, RA and posterior regions and less amplitude from the midline.

154 of 223

Results N100-P300 after RTMS (HC-N4, SCZ-N5)

All patients (TG + CG) showed reduced N100-P300 complex voltage at the mid posterior region, with TG patients (n2) showing the major global difference with reduced voltages in all regions

Individual data showed reduced P300 amplitude in all patients after RTMS treatment, with the S15 subject showing no values after RTMS

Passive task

155 of 223

Exploratory Study 2. A Novel Technique to Trigger High Beta and Low Gamma Activity in Patients with�Schizophrenia (E. Ívarsson et al., 2020)

RESEARCH QUESTION

It has been suggested that schizophrenic symptoms can be explained by over-arousal causing depression in neural activity or inverted-U relationship between performance and arousal (Grossberg, 2000; Yerkes & Dodson, 1908).

Our study followed this suggestion and we expected that in schizophrenia the high-beta and gamma activity will break up in clusters and decrease during a task requiring attention, whereas in normal subjects these bands activity will be increased.

A cognitively driven auditory-motor task was suggested as “triggering beta and gamma neural synchrony” in the cortical regions involved in the working memory and auditory-motor cortices.

156 of 223

Method

The auditory-motor task (AMT) was employed in a sound-attenuated room, and each patient was awake, un-sedated, and comfortably seated on a chair during the tasks.

Subjects held a button in one hand and placed the hand on the thigh. We instructed each subject to press the button using the thumb when the pre-recorded verbal command “press” was given and not to press the button when the verbal command “no press” was given fig.

Subsequently, each subject completed two auditory-motor tasks (one for each hand). The AMT task contained 40 trials: 20 auditory-verbal “press” commands and 20 “no press” commands. The interstimulus interval was 3 seconds (Nagasawa et al., 2010).

The commands were presented in a pseudorandom sequence during each task. The AMT was performed with the dominant hand (AMT-r), and non-dominant hand (AMT-l) one time in HC and two times in the patients (SCZ), before the RTMS treatment (SCZ-T1) and within one week after completing ten sessions of RTMS treatment (SCZ-T2).

157 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

Here are presented the results of motor cortical activation (MCA) for one HS and one patient with auditory verbal hallucinations (AVH), using power spectral density (PSD) and topographical frequency maps including both hands sensory-motor regions.

158 of 223

Method

First, we analyzed 40 ± 10 sec resting-state EEG with eyes closed for three patients with schizophrenia and we compared it with the EEG of three healthy control participants.

For the auditory-motor task, epochs related to the cognitively driven “motor cortical activation” (MCA) were segmented between -500 ms to +500 ms (1000 ms) relative to the onset of the button code (hand reaction) seen in the EEG trace. The reference period was set between +1500 ms to +2500 ms post-verbal command or auditory stimulus onset (1000 ms). Finally, the remaining artifact free trials were analyzed.

159 of 223

Method

Later, ACA, MCA, and NCA epochs were modified (Study 6) to avoid overlapping of the EEG data.

We looked at specific frequencies related to the motor reaction: low beta (13–20 Hz), high beta (20–30 Hz), low gamma (30–45 Hz), and high gamma (45-80 Hz).

We compared data with the reference period and with the resting state EEG.

160 of 223

161 of 223

162 of 223

163 of 223

Right Hand

Post TMS

164 of 223

Results

  • The results appeared to be congruent with our hypotheses, showing more clustered and fragmented cortical distribution in patients with schizophrenia and reduced activity during the auditory-motor task, even during the reference period between the auditory commands.

  • The activity seems to be stable during the task, with little lateralization effect of hand response.

165 of 223

Exploratory Study 3. The cortical and cutaneous silent period in patients with schizophrenia and AVH

RESEARCH QUESTION

Results pertaining to CSP in schizophrenia patients are controversial. Prolonged CSP was observed among both first-episode patients and clozapine medicated chronic patients compared with healthy controls, suggesting alterations within the GABAB-mediated neurotransmitter system (Daskalakis et al., 2002).

One study found no significant differences in CSP between SCZ patients and healthy controls, and others have reported a shortened CSP in either the chronic or unmedicated patients (Fitzgerald et al., 2002; S.-K. Liu et al., 2009).

Based on these findings, the working hypothesis was that CSP will be prolonged after RTMS as a signal of improvement in the inhibitory system related to sensory gating. Additionally, we looked to cutaneous silent period, which was expected to be prolonged, too. Both data are presented as case studies

166 of 223

Methods

Cortical silent period (CuSP) duration was measured in sub-maximally contracted abductor pollicis brevis muscle by stimulating the motor cortex using TMS with an intensity of 140% resting motor threshold (RMT). Calculations of CSP were determined from the motor evoked potential (MEP) onset to the recovery of any voluntary EMG activity.

Three to five trials were repeated to obtain the average CSP duration for each subject.

To obtain cutaneous silent period (CuSP) electrical stimuli were delivered to the ipsilateral dermatome of the 2nd digit from the dominant hand.

Stimuli were delivered with ring electrodes at 10 times x sensory threshold. Surface EMG electrodes recorded CuSP from abductor pollicis brevis muscle while the patient was requested to continuously sustain a submaximal contraction. When the stimulus produced a clear and visible silent period, three trials were collected for each subject.

167 of 223

RESULTS CSP

168 of 223

RESULTS CuSP

169 of 223

170 of 223

171 of 223

Study 4. Effects of transcranial magnetic stimulation on AVH and mid latency auditory evoked potentials in patients with schizophrenia

RESEARCH QUESTION

We expected that RTMS would reduce AVH severity (hypothesis H1), that stress and anxiety would be reduced, and that QoL would be increased after the RTMS treatment in patients with SCZ and AVH (hypothesis H2).

Based on the assumption that there is impaired triggering of attention in patients with schizophrenia, as made evident by reduced N100 amplitude we expected that N100 amplitude would be higher after the RTMS (hypothesis H3) and that N100 and P200 sensory gating which appeared to be impaired in patients with schizophrenia and AVH will improve (hypothesis H4).

172 of 223

Methods

Clinical symptoms were used as primary outcomes.

Additionally, we measured N1 component (N100) and P2 component (P200) neurophysiological markers as a secondary outcome. A paired-click paradigm was performed to elicit the P50, N100, and P200 components. In this study, we focused on N100 and P200 responses.

Variables

The largest negative deflection between 80 and 150 ms was identified as the N100 or N1, and the largest positive deflection between 150 and 250 ms was identified as the P200 or P2.

173 of 223

Methods

We divided the cortical EEG 256 channels in seven regions of interest (ROI), each of them with an averaged signal derived from 15 electrodes selected from 3 parallel lines (105 electrodes out of a total of 256). At the end, we measured 9 metrics for each ROI and for the average of all electrodes (AA and AVG) providing 72 (63 + 9) variables.

The patient groups were compared to the healthy control group at baseline and the two patient groups were also compared against each other at baseline.

174 of 223

Results

TG and CG neurophysiological data measured separately in pre-post RTMS conditions didn´t show significant changes.

Clinical symptoms showed a similar change direction for all psychometric scales: PSYRATS AHS decreased in both groups, QoL improved in both groups and DASS decreased in both groups.

The null hypotheses H1 and H2 were not true (type 1 error), and the alternative hypotheses could not be tested because of type 2 error (e.g., for PSYRATS the pre-study calculated N to achieve statistical power was 16).

We calculated pre-post RTMS changes for all 72 metrics in all patients (N=10). Two variables showed the most marked changes after the treatment with small-medium effect size: N1S1BmLP or N100 to S1 measured “baseline to peak” in left posterior region, which changed from -0.57 μV (SD 0.97) to -2.39 μV (SD 1.59), (p = 0.006, η2 = 0.346) and N1S1BmMP or N100 to S1 measured “baseline to peak” in medial posterior region (p = 0.038, η2 = 0.218). The p-values after Bonferroni correction should be < 0.00714 (7 variables).

175 of 223

Results�HS

176 of 223

Results�T3P3�S25 in T1

177 of 223

Results�T3P3�S25 in T2

178 of 223

Study 5. P300 Analysis Using HD EEG to Decipher Neural Response to RTMS in Patients with SCZ and AVH (Aubonnet et al., 2020)

RESEARCH QUESTION

We looked at the N100-P300 complex voltage before and after the treatment expecting that after the treatment the amplitude of P300 will be higher in patients receiving RTMS at T3-P3 EEG location.

The question was if P300 related oscillations and local connectivity participation index derived from a dense-array 256 channel EEG system can be considered as candidates for biomarkers of the patients with schizophrenia

179 of 223

Methods

PSD has been performed from the event-related oscillations and we looked to the network organization and the difference between T2 (postTMS) and T1 (pre-TMS) conditions. Network organization was analyzed with participation coefficient, a metric of functional segregation.

180 of 223

Methods

The connectivity has been computed at the cortical level using the "EEG source connectivity" method. It consists of estimating the brain sources (over 68 regions of interest - ROI) and then computing the statistical coupling between these reconstructed sources. The weighted minimum norm estimate (wMNE) and the Phase Locking Value (PLV) was used to solve the inverse problem and compute the functional connectivity, respectively.

The analysis has been performed on the beta and gamma bands. To quantify the network integration, we used the participation coefficient (PC), to calculate the interactions between brain modules (distant sub-networks), on the threshold connectivity matrices (here 20%). We used the brain connectivity toolbox (BCT) to compute the PC (Rubinov & Sporns, 2010).

181 of 223

Results

The individual analysis revealed general consistent results.

The analysis of the psychometric tests revealed that four out of five subjects in TG and three out of five subjects in CG felt improved condition after the treatment, whereas the other subjects remained neutral or reported worse psychometric scores.

In the time domain analysis, the N1-P3 amplitude was globally higher post-treatment than pre-treatment, for six subjects, two in TG and four in CG.

The PSD increased post-treatment mainly for the alpha band and beta band globally, for six subjects as well, two in TG and four in CG. No trends were detectable for the gamma and theta bands. In several subjects, the right temporal area showed an opposite behavior compared to the other regions.

The connectivity results showed an increased network integration (increase in participation coefficient) during post-treatment for frequent, for the beta band especially, for seven subjects, four in CG, three in TG.

182 of 223

Improvement in clinical outcomeS17 TG S23 CG

The yellow areas in frequency analysis are related to a higher PSD in T2, whereas the blue ones are related to a higher PSD in T1. Amount of increase (green) or decrease (orange) participation coefficient (PC) values. The positive bars in time analysis are related to a higher N1-P3 amplitude in T2 (Red Rare)

183 of 223

Stagnation or worse in psychometric scalesS26 CG S15 TG

The yellow areas in frequency analysis are related to a higher PSD in T2, whereas the blue ones are related to a higher PSD in T1. Amount of increase (green) or decrease (orange) participation coefficient (PC) values. The positive bars in time analysis are related to a higher N1-P3 amplitude in T2 (Red Rare)

184 of 223

Study 6. Network signatures of RTMS treatment in patients with SCZ and AVH during an auditory-motor task using HD-EEG (Ovidiu C. Banea et al., 2021)

RESEARCH QUESTION

Following our observations that the cortical distribution of the relative power in different EEG bands showed topographical EEG fragmentation the question

was if there will be increasing values of the relative power after the RTMS, especially for beta and gamma bands.

Auditory-motor task results were compared with the resting state. We expected that network organization measured with the graph theory and small worldness will show an improved small world effect.

185 of 223

Methods

Intervention: 10 session rTMS at 1Hz (900 pulses) at 100%RMT

Tests:

HD EEG (256 channels) gamma & beta cortical activation in resting condition and during auditory motor task before and after rTMS

Pshychometrics: PSYRATS AHS (11 items)

186 of 223

Patients

Gender

Diagnostic

Medication

TG1

M

Paranoid Schizophrenia

Clozapine, Fluoxetine, Wellbutrin, Metformin

TG2

M

Paranoid Schizophrenia

Clozapine, Olanzapine, Perphenazine, Alprazolam, Levomepromazine, Oxazepam and Melatonin

TG3

M

Paranoid Schizophrenia

Clozapine and Flupenthixol

CG1

F

Paranoid Schizophrenia

Aripiprazole, Olanzapine, Chlorprothixene and Pregabalin

TG4

M

Paranoid Schizophrenia

Non-medicated

CG2

M

Hebephrenic schizophrenia

Clozapine, Pregabalin, Amisulpride

Six patients (mean age = 30.2, SD 2.9, 5 males and 1 female) diagnosed with schizophrenia following the ICD-10 schizophrenia classification (F20) (SCZ group) and six healthy controls (HC) (mean age = 28.7, SD 4.3, 4 males and 2 females) were included in this study.

187 of 223

Hypotheses - AM task

Previous studies showed that pre-stimulus baseline (NCA) gamma activity is elevated (Gandal M) and that task-driven “evoked” gamma-band responses are reduced in schizophrenia (Gandal MJ, Edgar JC, Klook K, Siegel SJ.2012) Leicht et al., 2011). On the other hand, during working memory tasks, where there is a need for increased effort, it was observed that an efficient network organization is expressed by a higher SM index (Micheloyannis et al., 2006a).

Based on Gandal review and Micheloyannis findings (Gandal. Et al 2012, Micheloyannis et al., 2006a) we hypothesized

(1) that 10 days of rTMS will increase beta and gamma PSD over sensorimotor and temporoparietal cortices after auditory triggering of the hand-motor task,

(2) that rTMS will decrease gamma PSD during pre-stimulus baseline period and

(3) that small worldness index will improve after the rTMS.

188 of 223

189 of 223

ROI

190 of 223

RESTING STATE EEG HS, eyes closed (Power Spectral Density & Frequency map) / 60 seconds

191 of 223

RESTING STATE EEG HS, eyes opened (Power Spectral Density & Frequency map) / 60 seconds

192 of 223

First assessment: Eye-bird view maps for RP

Maps for Healthy Subjects (No rTMS) and Patients in T1 (pre TMS) and T2 (post TMS) conditions:

EEG bands: alpha (8-13Hz), beta (13-30Hz), low gamma (30-49Hz) and high gamma (51-100Hz)

193 of 223

Auditory-motor task protocol and windows of Interest

194 of 223

Healthy Subjects (No RTMS)

195 of 223

NCA

196 of 223

197 of 223

198 of 223

199 of 223

200 of 223

201 of 223

n e t w o r k o r g a n i z a t i o n

Clustered nodes

N4 - Connector hub

Number of Nodes

Number of Edges

Degree of centrality

Clustering Coefficient

Hub coefficient

Average Path Length

Small-worldness

Global synchrony

202 of 223

SMALL WROLDNESS And GRAPHS Theory

The global impairments in brain connectivity during working memory tasks can also be studied according to graph theory analysis, especially through measures such as small-world (SM) networks that exhibit high clustering and decreased characteristic path length.

The SM network phenomenon is observed when the connectivity among nodes of a given network is in the middle between a completely regular and random network (Watts and Strogatz, 1998).

SCZ show disrupted functional integration expressed by decreases in SM index (Micheloyannis et al., 2006b) with more contribution of frontotemporal networks and occipital regions (van den Heuvel et al., 2008; Zalesky et al., 2011).

203 of 223

1 - 2

1 - 3

1 - 4

256 - 255

256 - 256

Matrix with 256

rows and 256 columns

Significant values

204 of 223

Left Frontal

Right Frontal

Left sensory-motor

Right sensory-motor

Left temporo-parietal

Right temporo-parietal

T1 Subject I5 - Left Hand - Non-cortical activity

T2 Subject I5 - Left Hand - Non-cortical activity

205 of 223

Network organization AM Task

  • For Healthy Subjects we would expect higher clustering (existence of motifs, modules and hubs) with shorter path length for gamma band activity during working memory than during resting state

  • In Patients Group we expect higher global SWN organization after 10 days rTMS

  • Reference period (NCA): we would expect increased connectivity

206 of 223

RESULTS

Significant improvement on psychotic symptoms as measured by PSYRATS between pre-rTMS (M = 28,6, Std = 4,17) and post-rTMS (M = 23,6, std = 5,27) conditions, F(1, 5) = 23,44,

p < 0,05, ηp² = 0,967.

207 of 223

RTMS modulates gamma PSD in SCZ patients during

Resting state

LTP

LOW Gamma

208 of 223

RTMS modulates gamma PSD

over the

left temporoparietal in SCZ patients during AMT performed with non-dominant hand

209 of 223

Small worldness

210 of 223

Results

10 days of low frequency rTMS did not show significant improvement of neurophysiological parameters in patients with schizophrenia and pharmaco-resistant AVH.

Small worldness index showed changes in patients with schizophrenia and AVH after rTMS similarly to those found in healthy controls suggesting that this might represent an interesting biomarker for rTMS treatment effectiveness.

The sample size and a working memory psychological test as first outcome (we used positive symptoms PSYRATS) are important limitations of this study.

211 of 223

Summary

  • Study 1 and Study 4 showed that a dense-array 256 channel EEG system is not an easy option to record ERP in patients with schizophrenia when the data is to be analyzed in the time domain.

  • To quantify the RTMS aftereffects in EEG gamma oscillatory activity and network organization, in Study 2 we triggered beta and gamma activity with a cognitively driven auditory-motor task.

  • Study 3 contains data of the cortical silent period as a measure of central GABAB mediated cortical inhibition that is linked with sensory gating. The cutaneous silent period which is a peripherally spinal cutaneous-muscular reflex showed similar changes after RTMS in both groups of patients. It remains to complete the work to suggest that this neurophysiological marker is modulated at central CST level or intracortical inhibitory neurons.

212 of 223

Summary

  • Study 5 focused on the relation between psychometric scores and EEG data analysis in time, frequency domains, and functional connectivity. After conducting RTMS, most patients showed an evolution in psychometric data as well as on the neurophysiological quantitative data, independent of the stimulation site. When the psychometric improved post-TMS, we could observe an increased network organization mainly, through the participation coefficient in the beta bands, a higher alpha and beta band power, and different N1-P2 or N1-P3 behavioral responses.

  • In Study 6, we measured the change in phase connectivity and EEG coherence before and after the RTMS by analyzing EEG data obtained from the auditory-motor task with graph theory and small worldness

213 of 223

Conclusions

This thesis dissertation is a hypothesis-generating research analyzing the influence of RTMS on brain mechanisms in patients with schizophrenia and pharmaco-resistant AVH.

1) Based on the patient´s clinical evaluations and all the neurophysiological measurements it might be suspected that both, left temporoparietal and vertex RTMS induce positive changes in patients with schizophrenia and auditory verbal hallucinations.

2) By dividing the scalp regions and analyzing ERP data in seven regions of interest we were able to observe local changes of N100 amplitude after the RTMS. This finding suggests that this auditory evoked potential of triggering attention might be an interesting marker of RTMS induced neural plasticity to be analyzed in patients with schizophrenia and AVH.

214 of 223

Conclusions

3) P300 evoked EEG oscillations spectral analysis did not show group differences before and after RTMS treatment. It might be that in our study P300 was not elicited and we faced a habituation effect (The passive task elicits P200). However, the results suggest that brain connectivity, through the participation coefficient, and PSD, were highly related to the psychometric score and that N1-P3 complex, despite the variability, should be further investigated.

4) This report revealed for the first time that patients with schizophrenia exhibit higher gamma power spectral density during the auditory-motor task compared to healthy controls, which was modified by RTMS without being significant.

PSD of alpha, beta, and gamma oscillatory EEG activity and network organization showed the most marked changes in the prestimulus period or "delay epoch” compared with “auditory” or “motor” windows.

215 of 223

Conclusions

5) The change of low and high gamma PSD was visible, locally, over the left temporoparietal region, when the auditory-motor task was done with the nondominant hand, showing that during this condition, gamma synchronization can be a marker of “neural effort” and workload during the working memory-related time.

6) Small worldness index of low gamma activity analyzed in patients with schizophrenia and AVH during the working memory period of an auditory-motor task, when it is performed with the nondominant hand, showed changes after RTMS similarly to those found in healthy controls. This might represent an interesting biomarker for RTMS treatment effectiveness.

216 of 223

Limitations

  • The weaknesses of the works presented here are that they are based on a small sample size. The patients were also undergoing their usual treatment, including antipsychotic and sedative medications. This did not change between pre-and-post RTMS conditions but might have influenced the background neural activity and the generation of the ERPs (Javitt et al., 2008).
  • The experimental procedure was long (1 hour) and tiring and some patients had difficulty cooperating and maintaining tasks engagement, which may have affected data quality.
  • Muscle and movement artifacts added noise to the EEG signal, requiring a thorough pre-processing and the exclusion of many trials.

217 of 223

218 of 223

Brain Ecology �to understand mental illness

Thomas Fuchs 2005, Der Nervenarzt

  • Ecological or systemic view of the brain regards cognition and consciousness as components in a circular causality of organism and environment.

  • The Brain is allowing the organism to develop integral options of perception (A) and action (E) in its environment.

219 of 223

Brain, Organism and Environment = Dynamic Unity

  • The research efforts of Psychiatry must be equally focused on biological processes, psychopathological experiences, psychological - biographical connections, and social interactions, instead of being reduced to analyzing processes inside the brain (Thomas Fuchs, 2021)

  • Neuronal processes = components of a comprehensive process that can be viewed on different levels:
    • the macro- level of psychosocial processes or the interactions of persons,
    • the medium, individual level of interactions between brain, organism and environment, and
    • the micro- level of neuronal and molecular processes within the brain.

220 of 223

Ecological conception�has consequences for:

  • understanding mental illness

  • therapeutic approaches . A psychotherapeutic intervention on the macro level modifies the brain structures involved— top- down. The altered neuronal structure, however, in turn enables the patient's interactions with the environment to change— bottom- up

  • Role of subjectivity in psychiatry (Fuchs, 2005)

Fuchs, 2021 Psychiatry Between Psyche and Brain Chapter in Book

221 of 223

Schizophrenia - ECOLOGICAL VIEW

Schizophrenia

A key cognitive symptom in patients with schizophrenia is impaired sensory gating, defined as reduced ability to suppress processing of irrelevant and uninformative sensory input (Fröhlich, 2016)

222 of 223

„Just one more thing“

223 of 223

Takk fyrir!�

Mulțumesc!��Thank you!��Grazie!