Influence of sensorimotor µ rhythm phase and power on motor cortex excitability and plasticity induction, assessed with EEG-triggered TMS

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://hdl.handle.net/10900/125619
http://nbn-resolving.de/urn:nbn:de:bsz:21-dspace-1256198
http://dx.doi.org/10.15496/publikation-66982
Dokumentart: Dissertation
Erscheinungsdatum: 2022-03-28
Sprache: Englisch
Fakultät: 4 Medizinische Fakultät
Fachbereich: Medizin
Gutachter: Ziemann, Ulf (Prof. Dr.)
Tag der mündl. Prüfung: 2022-02-10
DDC-Klassifikation: 500 - Naturwissenschaften
610 - Medizin, Gesundheit
Schlagworte: Transkranielle magnetische Stimulation , Hirnstimulation , Elektroencephalographie , Synaptische neuronale Plastizität , Neurologie , Neurowissenschaften ,
Freie Schlagwörter: paarig assoziierte Stimulation
gepulste Inhibition
EEG-getriggerte transkranielle Magnetstimulation
Motorkortex
nicht-invasive Hirnstimulation
pulsed inhibition
EEG-triggered transcranial magnetic stimulation
paired associative stimulation
transcranial magnetic stimulation
motor cortex
neuroscience
non-invasive brain stimulation
EEG
neuroplasticity
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=en
Gedruckte Kopie bestellen: Print-on-Demand
Zur Langanzeige

Inhaltszusammenfassung:

In dieser Arbeit werden zwei Experimente vorgestellt, bei denen EEG-getriggerte transkranielle Magnetstimulation (TMS) an gesunden Probanden eingesetzt wurde, um die Rolle des sensomotorischen 8-14Hz µ-Rhythmus auf die kortikospinale Erregbarkeit (CSE) und die Induktion positiver Plastizität zu untersuchen. Unser Ziel war es, für Plastizitätsinduktion günstige Zeitpunkte im EEG zu identifizieren, um in Zukunft die Effektivität solcher zurzeit oft noch unzuverlässigen Anwendungen zu steigern. Unser EEG-TMS System interpretierte Oszillationen im EEG in Echtzeit und löste einen Stimulus aus, wenn bestimmte, vorher festgelegte Eigenschaften zutrafen. Die ‘Gehirnwellen’ im EEG entstehen durch synchronisierte Fluktuationen des Membranpotentials kortikaler Neurone, welche aufgrund ihrer intrakortikalen Kommunikationsfunktion wertvolle Informationen über neuronale Erregbarkeit vermitteln. Im Gegensatz zu “open-loop” TMS ermöglicht EEG-TMS nicht nur eine präzisere Erforschung der Funktion von Gehirnwellen, sondern auch die Umsetzung der gewonnenen Erkenntnisse in effizientere therapeutische Anwendungen. Speziell Oszillationen im Alpha-Frequenzbereich (8-14Hz) spielen eine bedeutsame Rolle, indem sie den Informationsfluss im Gehirn durch Hemmung aktuell irrelevanter Areale steuern, und zwar laut einer führenden Theorie als “asymmetrisch gepulste Inhibition” mit einem Maximum der Hemmung während der Hochpunkte (“Peaks”) und während hoher “Power” (∼ Amplitude). Der “µ-Rhythmus”, Wellen in alpha-Frequenz über dem sensomotorischen Kortex, scheint für diese Areale eine analoge Rolle wie das okzipitale Alpha für den visuellen Kortex zu spielen. Die CSE lässt sich durch die Amplitude der ausgelösten kontralateralen Muskelzuckungen (MEPs im EMG) quantifizieren. Im Vorexperiment erforschten wir den Einfluss der Power der µ-Wellen auf die CSE. 16 Teilnehmer wurden in einer Sitzung mit Einzelpuls-TMS des linken M1 stimuliert. Die Pulse wurden durch die momentane Power ausgelöst, 10 Dezile des individuellen µ-Powerspektrums wurden in pseudorandomisierter Reihenfolge angesteuert, verteilt auf 4 Stimulationsblöcke. Nach Berücksichtigung der “Inter-Trial-Intervalle” (ITIs, bekannter “Confounder”) und Normalisierung pro Block zeigten unsere Daten eine schwache positiv-lineare Korrelation zwischen µ Power und MEP-Amplitude, welche somit im Widerspruch zur angenommenen hemmenden Wirkung von µ steht, aber mittlerweile in mehreren anderen Studien repliziert wurde. Diese Diskrepanz kann z.B. durch eine tatsächlich fazilitatorische Wirkung erklärt werden, oder auch durch eine anatomisch dem sensorischen Kortex (S1) zuzuordnende Quelle der angesteuerten µ-Wellen, was über hem- 83mende Interneurone von S1 auf M1 zu einer ‘Vorzeichenumkehrung’ der Effektrichtung führen könnte. Weiterhin wird eine Abhängigkeit der ‘erregbarsten’ Power-Werte von der Stimulusstärke diskutiert. Im Hauptexperiment sollte mit ‘paarig-assoziativer Stimulation’ (PAS) (intervallsensitive Kombination von Elektrostimulation des rechten Nervus medianus mit TMS des linken M1) positive Plastizität (die Intervention überdauernde Stärkung von Synapsen) induziert werden. Dem ging ein umfangreiches “Screening” zur Identifikation geeigneter Probanden mit ausgeprägtem µ-Rhythmus (für präzise EEGTriggerung) voraus. Letztlich absolvierten 16 Teilnehmer je 4 Sitzungen (eine pro Trigger-Bedingung). Unsere Hypothese war hierbei, mehr Plastizität nach Stimulation während der Tiefpunkte (“Troughs”) als während der Peaks zu erzielen, also mehr synaptische ‘Formbarkeit’ während höherer Erregbarkeit. In Anbetracht der schwachen Ergebnisse des Vorexperiments sowie einer widersprüchlichen Beweislage bezüglich einer fazilitatorischen oder inhibitorischen Funktion wurden hohe und niedrige Power nicht explizit miteinander verglichen. TMS während PAS wurde durch (1) µ-Peaks, (2) µ-Troughs, (3) mittlere µ-Power und (4) open-loop getriggert. (3) und (4) dienten jeweils als Kontrollbedingung. PAS konnte, unabhängig von der EEG-Bedingung, keine signifikante Veränderung der MEP-Amplituden vom Ausgangswert hervorrufen. Die fehlende Wirkung könnte durch intra- und interindividuelle Schwankungen gewisser Parameter zwischen den Sitzungen erklärt werden (z.B. MEP-Ausgangswerte, absolute µ-Power während PAS), die sich jedoch nicht als systematische Confounder zwischen EEG-Bedingungen herausstellten. Die, im Gegensatz zu open-loop-Studien, schwankenden ITIs während der PAS könnten die Wirkung ebenfalls beeinträchtigt haben. Weiterhin waren zwei verschiedene Kortexareale (S1 und M1) am Protokoll beteiligt, was die Identifikation einer relevanten EEG-Eigenschaft erschwerte. Gegenwärtig rufen Plastizitäts-induzierende TMS-Protokolle in der Forschung und in Studien mit Schlaganfallpatienten schwankende und zeitlich begrenzte Wirkungen hervor. Durch EEG-Triggerung und / oder die Kombination mit klassischer Physiotherapie könnte eine verbesserte Effektivität und somit eine routinemäßige Anwendung erreicht werden. Trotz unserer negativen Ergebnisse bleibt EEG-getriggerte TMS ein vielversprechendes Instrument in Forschung und Klinik.

Abstract:

This thesis presents two experiments employing real-time EEG-triggered transcranial magnetic stimulation (TMS) on healthy volunteers to investigate the role of sensorimotor 8-14Hz µ rhythm in EEG at rest on corticospinal excitability and induction of positive plasticity. We intended to identify brain states favorable to induction of positive plasticity to inform development of more efficient TMS protocols for clinical application e.g. in stroke patients. Applying TMS triggered by pre-determined EEG brain states in real time (opposed to open-loop TMS with post-hoc trial sorting) offers not only more precise research into the role of certain brain waves, but also translation into more efficient therapies. The membrane potential of superficial cortical neurons fluctuates rhythmically, visible as oscillations in surface EEG. Different brain areas seem to communicate through these synchronized fluctuations. ‘Brain waves’ therefore convey valuable information about the excitability of said areas. Oscillations in the alpha frequency range (8-14Hz) play a crucial role in this, gating information by inhibiting brain areas irrelevant to the current task. According to an influential hypothesis, this function is exerted as an ‘asymmetric pulsed inhibition’, with a maximum of inhibition during the peaks and during high alpha power (∼ amplitude). Sensorimotor alpha frequency waves (µ rhythm) play a similar role as the well-researched occipital alpha does for the visual cortex. The primary motor cortex (M1) provides a quantifiable measure of (corticospinal) excitability, the amplitude of TMS-elicited contralateral muscle twitches (appearing as MEPs in the EMG). The first experiment investigated the role of µ power for M1 excitability. 16 participants underwent one session of single-pulse TMS of the left M1, triggered by overall 10 individual power deciles in pseudorandomized order, partitioned into 4 ‘blocks’ of stimulation over time. The data revealed, after stratification for confounding inter-trial-intervals (ITIs) and normalization to block average, a weak positive linear relationship contrary to the proposed inhibitory role of µ, which has however since been replicated several times in other studies. This discrepancy can be explained e.g. by an in fact facilitatory nature of µ, by a postcentral and thus sensory cortical (S1) source of the targeted oscillations, reversing the inhibitory effect in sign to a facilitatory one through S1-to-M1 feedforward inhibition, or by a shift of most excitable power values dependent on stimulus strength. For the main experiment, we applied a paired associative stimulation (PAS) pro- 81tocol intended to induce positive plasticity (strengthening of synaptic connection outlasting the intervention), combining electrical stimulation of the right median nerve at the wrist with a TMS of the left M1 in a temporally sensitive manner. After an extensive screening to pre-select suitable subjects with a sufficiently strong µ rhythm (to ensure accurate performance of the real-time EEG targeting), 16 participants completed 4 sessions (one condition each). We expected to induce more positive plasticity during more excitable brain states, i.e., µ troughs rather than µ peaks. In light of our findings on µ power from the first experiment (weak influence as compared to ITIs and intrinsic variability over time) and overall contradictory evidence as to its (facilitatory versus inhibitory) role, high vs. low power were not explicitly compared. TMS during PAS was applied at (1) µ peaks, (2) µ troughs, (3) at medium µ powers and (4) open-loop. (3) and (4) both served as controls. The intervention failed to evoke a significant change in MEP amplitudes from baseline irrespective of condition. Possible explanations can be found in the intra- and interindividual variability of decisive parameters across sessions (e.g. baseline amplitudes and absolute µ powers during PAS), which however did not significantly depend on the targeted condition and were thus not true confounders. The number of sessions might still have introduced a further measure of variability. Varying PAS ITIs (due to EEG-triggering) could have also impeded plasticity induction, and the involvement of two cortical regions (S1 and M1) might have complicated the identification of one relevant brain state. Currently, plasticity-inducing TMS protocols in research and clinical trials evoke variable and transient effects. Improvements to enable routine application might come from EEG-triggering and/or combining with traditional motor training (physiotherapy). Regardless of our nil results in plasticity induction, EEG-triggered TMS remains a promising instrument in research and therapy.

Das Dokument erscheint in: