Deep brain stimulation (DBS) of the subthalamic nucleus (STN) and DBS of the pedunculopontine nucleus (PPN) have been shown to be effective surgical therapies for Parkinson’s disease (PD). To better understand the PPN and STN as DBS targets for PD, this research compares the anatomy, electrophysiology, and motor control roles of these nuclei. PPN and STN connections were examined in vivo in human subjects and in the non-human primate using probabilistic diffusion tractography. Both the PPN and STN were connected with each other and with the motor cortex (M1) and basal ganglia. After studying these anatomical connections in primates, their functional significance was further explored in an anesthetized rat model of PD. Examination of the electrophysiological relationship between the PPN and basal ganglia in the presence of slow cortical oscillatory activity suggested that excitatory input from the STN may normally modulate PPN spike timing but that inhibitory oscillatory input from the basal ganglia output nuclei has a greater effect on PPN spike timing in the parkinsonian brain. To examine transmission and modulation of oscillatory activity between these structures at higher frequencies, LFP activity was recorded from the PPN and STN in PD patients performing simple voluntary movements. Movement-related modulation of oscillatory activity predominantly occurred in the α (8-12 Hz) and low β (12-20 Hz) frequencies in the STN but in the high β (20-35 Hz) frequencies in the PPN, supporting observations from rodent studies suggesting that oscillatory activity is not directly transmitted from the STN to the PPN in PD. Finally, to better understand the roles of the STN and PPN in large-scale movement, the effects of STN and PPN DBS on gait abnormalities in PD patients were studied. DBS of the STN appeared to improve gait by optimising executive gait control while DBS of the PPN appeared to restore autonomic gait control. These results have several implications for DBS patient selection, surgical targeting, and for understanding the mechanisms underlying DBS efficacy.

This thesis is concerned with the neurocircuitry that underpins the cardiovascular response to exercise, which has thus far remained incompletely understood. Small animal studies have provided clues, but with the advent of functional neurosurgery, it has now been made possible to translate these findings to humans. Chapter One reviews the background to the studies in this thesis. Our current understanding of the cardiovascular response to exercise is considered, followed by a discussion on the anatomy and function of various brain nuclei. In particular, the rationale for targeting the periaqueductal grey (PAG) and the subthalamic nucleus (STN) is reviewed. Chapter Two reviews the use of deep brain stimulation (DBS), in which deep brain stimulating electrodes are implanted into various brain nuclei in humans, in order to treat chronic pain and movement disorders. This technique not only permits direct electrical stimulation of the human brain, but also gives the opportunity to record the neural activity from different brain regions during a variety of cardiovascular experiments. This chapter also gives a detailed methodological description of the experimental techniques performed in the studies in this thesis. Chapter Three identifies the cardiovascular neurocircuitry involved in the exercise pressor reflex in humans using functional neurosurgery. It shows for the first time in humans that the exercise pressor reflex is associated with significantly increased neural activity in the dorsal PAG. The other sites investigated, which had previously been identified as cardiovascular active in both animals and humans, seem not to have a role in the integration of this reflex. Chapter Four investigates whether changes in exercise intensity affect the neurocircuitry involved in the exercise pressor reflex. It demonstrates that the neural activity in the PAG is graded to increases in exercise intensity and corresponding increases in arterial blood pressure. This chapter also provides evidence to suggest that neural activity in the STN corresponds to the cardiovascular changes evoked by the remote ischaemic preconditioning stimulus in humans. Chapter Five identifies the cardiovascular neurocircuitry involved during changes in central command during isometric exercise at constant muscle tension using muscle vibration. It shows that, in humans, central command is associated with significantly decreased neural activity in the STN. Furthermore, the STN is graded to the perception of the exercise task, i.e. the degree of central command. The other sites investigated appear not to have as significant a role in the integration of central command during the light exercise task that was undertaken. Chapter Six studies the changes in muscle sympathetic nerve activity (MSNA) during stimulation of various brain nuclei in humans. Regrettably, the results presented in this chapter are not convincing enough to support the hypothesis that stimulation of particular subcortical structures corresponds to changes in MSNA. However, the cardiovascular changes that were recorded during stimulation of the different subcortical structures are congruous with previous studies in both animals and humans. Chapter Seven presents a brief summary of the findings in this thesis.

L’efficacité de la stimulation cérébrale profonde subthalamique dans certains cas de maladie de Parkinson est maintenant bien établie. Toutefois, des progrès restent possibles, à la fois en terme de contrôle du geste chirurgical et en terme de définition de la cible chirurgicale. Dans la première partie de ce travail, nous nous sommes intéressés à l’optimisation du contrôle de l’implantation des électrodes de stimulation cérébrale profonde. Nous avons tout d’abord analysé rétrospectivement les résultats obtenus en réalisant une imagerie tridimensionnelle per-opératoire pour le contrôle de positionnement des électrodes. Nous nous sommes ensuite intéressés à la possibilité d’utiliser un repère de visée radiologique per-opératoire. Nous avons revu pour cela une série de patients ayant subi une réimplantation d’électrodes, pour lesquels l’électrode déjà en place était utilisée comme point de repère à la fois pour définir la cible de la réimplantation et pour contrôler radiologiquement l’implantation de la nouvelle électrode. Dans la seconde partie, nous avons travaillé à l’optimisation de la cible subthalamique. Nous avons tout d’abord évalué la pertinence du repérage du faisceau mamillo-thalamique sur des coupes IRM axiales comme marqueur de la coordonnée y du bord antérieur du noyau subthalamique. Ensuite, nous avons tâché de proposer une normalisation tridimensionnelle de l’espace stéréotaxique à partir de données recueillies dans une série de volontaires sains. Enfin, pour une série de patients opérés avec un bon résultat, nous avons cherché à corréler la position des contacts actifs en stimulation chronique avec des points de repères profonds visibles en IRM. Nous avons pu proposer ainsi une cible normalisée dont les coordonnées sont : x = 0,44xbord latéral du V3 + 10,71mm; y = 0,69xfaisceau mamillothalamique + 1,62 mm ou 0,34 distance CACP + 2,52 mm; z = 0,72 hauteur du thalamus – 16 mm. Cette cible sera évaluée dans une future étude prospective. / The clinical efficacy of subthalamic deep brain stimulation is now well established. Nevertheless, progress is possible, regarding especially (1) the accuracy of electrodes implantation and (2) the definition of the surgical target. In the first part of this work, we worked on the optimization of DBS electrodes implantation. First, we analyzed retrospectively the results obtained by using intra-operative 3D imaging for the control of microelectrodes and definite leads placement. Thereafter, we considered the possibility to use a radiological landmark for intraoperative controls. To this end, we studied the cases of patients who underwent reimplantation of DBS electrodes. The initial electrode (still implanted) was used as a landmark: (1) for the deifntion of the reimplantation target and (2) for the radiological control of the new lead positioning. In the second part, we worked on the optimization of the surgical target. First, we assessed the interest of the mamillothalamic tract as a landmark of the anteroposterior coordinate of the anterior border of the STN in MR axial images. Thereafter, we tried to identify MR landmarks for tridimensionnal normalization of the stereotactic space. Finally, we tried to correlate the coordinates of active contacts with MR-defined landmarks in a series of patients that had been operated with good clinical results. Based on our results, we can propose the following coordinates for a new normalized subthalamic target : x = 0.44xlat edge 3rd ventricle + 10.71mm; y = 0.69xmamillo-thalamic tract + 1.62 mm or 0.34 ACPC length + 2.52 mm; z = 0.72xthalamus height – 16 mm. We will assess this target in a future prospective study.

Introdução: A disfunção da marcha é um dos maiores comprometimentos funcionais do paciente com a doença de Parkinson (DP). A estimulação cerebral profunda do núcleo subtalâmico tem mostrado melhora da marcha e equilíbrio. Esse efeito pode ser mantido e potencializado por programas de reabilitação motora específicos, como o treino em esteira sem e com suporte de peso corporal. No entanto, faltam estudos desses treinos em pacientes com a DP em uso desta estimulação. Objetivo: Comparar parâmetros cinemáticos lineares e angulares da marcha de pacientes com a DP em uso de estimulação cerebral profunda bilateral do núcleo subtalâmico, antes e após dois treinamentos: esteira sem e com suporte de peso corporal, associados à cinesioterapia convencional. Métodos: 12 pacientes (60,9 ± 10,6 anos; 20 ± 7 anos de doença e 20 ± 4 meses de tempo de cirurgia) completaram ambos os treinos em estudo cruzado fixo. Os pacientes passaram por 8 semanas de treino de marcha em esteira sem suporte de peso corporal e programa de cinesioterapia convencional, seguidas por 6 semanas de período sem intervenção. Posteriormente, realizaram 8 semanas de treino de marcha em esteira com suporte de peso corporal e o mesmo programa de cinesioterapia regular. As intervenções tiveram frequência de duas vezes por semana e duração de 90 minutos por sessão. A análise cinemática da marcha envolveu oito câmeras infravermelhas que detectaram 19 marcadores reflexivos nos membros inferiores dos pacientes. A análise estatística utilizou o teste Wilcoxon e foi adotado valor de p <= 0,05 como estatisticamente significante. Resultados: Ambos os treinos não mostraram diferenças significativas nos parâmetros lineares. Após o treino com suporte, observou-se aumento significativo dos seguintes parâmetros angulares: amplitude de movimento da pelve (inclinação, obliquidade e rotação); amplitude de movimento do quadril (abduçãoadução e rotação); % da fase de balanço que corresponde à flexão máxima do joelho e amplitude de movimento da progressão do pé. Conclusão: O treino em esteira com suporte de peso corporal mostrou capacidade de promover mudanças em parâmetros cinemáticos angulares da marcha. As implicações do treino em suspensão podem ter sido somadas aos efeitos neurofisiológicos da estimulação cerebral profunda e então desencadeado a melhora da mobilidade dos membros inferiores durante a marcha / Introduction: Gait disturbance is one of the hallmark features of Parkinson\'s disease (PD). Subthalamic nucleus deep brain stimulation (DBS) has shown improvements in gait and balance, and this effect can be maintained and enhanced by specific motor rehabilitation programs, such treadmill training without and with body weight support. However, at present there is a paucity of research on these combined interventions in PD with of this stimulation. Objective: To compare training-induced changes in gait linear and angular kinematic parameters among patients with PD who have used bilateral subthalamic nucleus DBS, and a combined intervention of conventional physical therapy with either treadmill training with body weight support or without support. Methods: 12 patients (age: 60.9 ± 10.6 years; disease duration: 20 ± 7 years; and time since DBS surgery: 20 ± 4 months) completed both training protocols in a fixed cross-over design. All patients received 8 weeks of treadmill training without body weight support in conjunction with conventional physical therapy, followed by a 6 weeks wash out period of no training. Thereafter, all patients received 8 weeks of body weight support treadmill training, in conjunction with the same conventional physical therapy. Both interventions had a frequency of two times per week, and duration of 90 minutes per session. Gait kinematic analysis involved eight infrared cameras that detected 19 reflective spherical markers attached to the limb lower of patients. Statistical analysis used the Wilcoxon and was adopted the value of p <= 0,05 as statistically significant. Results: Both the training no showed significant differences in linear parameters. After the body weight support training, observed there was a significant increase in following angular parameters: pelvis\' range of motion (tilt, obliquity, rotation); hip\'s range of motion (abduction-adduction and rotation); % Knee maximal flexion on Swing phase and foot progression\' range of motion. Conclusion: Treadmill training with body weight support showed an ability to promote changes in gait angular kinematic parameters. The implications of this training may have been added to the neurophysiological effects of DBS and then triggered the improved of mobility of lower limbs during gait

INTRODUÇÃO: Os efeitos da levodopa (LD) e da estimulação cerebral profunda (ECP) de núcleo subtalâmico (STN) sobre o equilíbrio e sintomas axiais são até o momento controversos. OBJETIVOS: Avaliar quantitativamente os efeitos da ECP de STN e da LD sobre o equilíbrio estático em pacientes com DP operados, em comparação com a LD em pacientes não operados. MÉTODOS: Trinta e um pacientes submetidos a ECP de STN entre 3 meses e 1 ano e meio antes da avaliação e 26 controles portadores de DP não operados, estágios Hoehn e Yahr 2 a 4 foram avaliados usando UPDRS para avaliação clínica e plataforma de força para avaliar oscilações posturais. O primeiro grupo foi avaliado com ECP e sem medicação, com ECP e com medicação e sem ECP e sem medicação. O segundo grupo foi avaliado com e sem medicação. Cada paciente foi avaliado com os olhos abertos e fechados. O deslocamento do centro de pressão anteroposterior, laterolateral, a área, velocidade e deslocamento total linear foram medidos pela plataforma de força. Os dados paramétricos foram comparados usando o teste t de Student e os dados não-paramétricos foram comparados pelo teste de Kruskal-Wallis. A avaliação clínica consistiu na parte 3 da escala UPDRS e na escala Hoehn e Yahr. Nível de significância estatística considerada foi p=0,05. RESULTADOS: Os pacientes não operados oscilaram mais quando sob efeito da levodopa do que sem medicação. No grupo operado, a maior oscilação é no grupo com ECP desligada e sem medicação. Tende a reduzir sob efeito da ECP apresenta redução significativa sob efeito simultâneo de ECP e levodopa. CONCLUSÃO: A associação da ECP de NST com medicação tem impacto positivo sobre o controle postural. O efeito da ECP de NST reverte o efeito negativo da levodopa sobre as oscilações observadas em pacientes não operados / INTRODUCTION: The effects of bilateral subthalamic (STN) DBS and medication on balance and on axial symptoms in PD have been so far inconsistent. OBJECTIVE: To assess quantitatively the effects of DBS on static balance in PD. METHODS: Thirty-one patients submitted to STN DBS over 3 months before and 26 non-operated controls with PD on Hoehn & Yahr stage \"on\" 2 to 4 were evaluated using UPDRS and a force plate to measure sway. The first group was evaluated on-DBS/off-medication, on-DBS/on-medication and off-DBS/off-medication. The second group was evaluated on and off medication. Each group was assessed with eyes open and then closed. Antero-posterior, laterolateral postural displacements of the center of pressure (COP), as well as 95% sway area, path length and speed of oscillation were analyzed and compared using t-Student test for parametrical data and Kruskal-Wallis test for non-parametrical data. Level of significance was set to p < 0.05. Clinical assessment consisted of UPDRS part 3 and Hoehn & Yahr scores for each of the conditions. RESULTS: Control patients tended to oscillate more in the on medication condition than off medication. DBS patients tended to oscillate more in the off-DBS/off medication condition, with a tendency to decrease the sway when on DBS/off medication with additional decrease when on DBS/on medication. CONCLUSION: Association of bilateral STN DBS and medication positively influences postural control in PD and surgery reverses the tendency of medication to increase body sway in non-operated patients

A estimulação cerebral profunda do núcleo subtalâmico (NST) é um tratamento bem estabelecido para os sintomas refratários à medicação em paciente com doença de Parkinson avançada. Além do procedimento de implante, a programação dos eletrodos é uma etapa fundamental para atingir os resultados desejados. A primeira etapa da programação é estabelecer os limiares para efeitos adversos. Contudo, a correlação entre a localização do eletrodo e o limiar para efeitos adversos associados à estimulacao das estruturas adjacentes ainda não é bem estabelecida. Características neuroanatômicas e a localização dos eletrodos foram identificadas utilizando-se um programa de planejamento de cirurgia estereotáxica, enquanto os parâmetros de estimulação e os efeitos adversos foram obtidos dos prontuários médicos. As correlações entre estas variáveis foram testadas através de análises univariadas e análises multivariadas. Estimulação monopolar produziu efeitos adversos capsulares (EA-C) em 208 dos 316 contatos (65,8%) e efeitos adversos não-capsulares (EA-NC) em 223 dos 316 contatos (70,6%). A ocorrência de EA-C esteve associada com o número do contato (p = 0,009) e com a coordenada \"Z\" (p = 0,03), enquanto o limiar de voltagem para EA-C esteve correlacionado com o ângulo da cápsula interna (p = 0,035). A ocorrência de EA-NC esteve associada com o número do contato (p = 0,005), \"X\" (p = 0,03), \"Y\" (p = 0,004) e com a distância para o núcleo rubro (p = 0,001 e p = 0,003). Houve correlação entre o limiar de voltagem para EA-NC e o ângulo da cápsula interna (p = 0,006), o ângulo coronal do eletrodo (p = 0,02), \"X\" (p = 0,001), \"Y\" (p < 0,001), \"Z\" (p < 0,001) e com as distâncias para a cápsula interna (p = 0,02) e para o núcleo rubro (p = 0,004 e p < 0,001). EA-C estiveram associados com os contatos mais distais do eletrodo e com localização mais profunda, bem como com maior angulação da cápsula interna. EA-NC estiveram associados com os contatos mais distais do eletrodo, localizados mais medial, posterior e inferiormente e mais próximos do núcleo rubro. Ademais, houve associação entre EA-NC e eletrodos implantados com maior ângulo coronal, bem como com maior angulação da cápsula interna. Estes achados poderão ser úteis no desenvolvimento de novas estratégias para o planejamento do implante de eletrodos de estimulação cerebral profunda. / Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is a well-established treatment for medically refractory motor symptoms of patients with advanced Parkinson\'s disease. Programming of the device is as relevant to patient outcome as accurate implantation of the electrodes. The first step of DBS programming is to identify the thresholds to side effects. However, the relationship between lead location and the threshold to adverse effects is not fully understood. Anatomical measurements and electrode location were evaluated on a stereotactic surgical planning software, whereas stimulation parameters and side effects were obtained from medical records. Correlations among these variables were tested using univariate and multivariable analyses. Monopolar stimulation elicited capsular side effects (CSEs) in 208 of 316 contacts (65.8%) and noncapsular side effects (NCSEs) in 223 of 316 contacts (70.6%). The occurrence of CSEs was correlated with contact number (p = 0,009) and with the \"Z\" coordinate (p = 0,03), whereas voltage threshold to CSEs exhibited correlation with the internal capsule angle (p = 0,035). The occurrence of NCSEs was correlated with contact number (p = 0,005), \"X\" (p = 0,03), \"Y\" (p = 0,004), and the distance to the red nucleus (p = 0,001 and p = 0,003). There was correlation between voltage threshold to NCSEs and the internal capsule angle (p = 0,006), electrode\'s coronal angle (p = 0,02), \"X\" (p = 0,001), \"Y\" (p < 0,001), \"Z\" (p < 0,001), and the distances to the internal capsule (p = 0,02) and to the red nucleus (p = 0,004 and p < 0,001). CSEs were associated with more distally contacts, with deeper localization, as well as with greater internal capsule angles. NCSEs were associated with more distally contacts, with localization more medial, posterior and inferior, and closer to the red nucleus. Moreover, there was a correlation between NCSEs and electrodes implanted with greater coronal angles, as well as with greater internal capsule angles. These findings can be useful to inform novel targeting strategies for deep brain stimulation lead implantation.

La maladie de Parkinson est une maladie neurodégénérative secondaire à la mort des neurones dopaminergiques, entraînant un dysfonctionnement des ganglions de la base (GB), ensemble de noyaux sous corticaux impliqués dans le contrôle de la motricité. Les symptômes moteurs sont améliorés par le traitement dopaminergique et par la stimulation cérébrale profonde (SCP) du noyau subthalamique (NST). Il a été montré que le changement d’activité bêta est corrélé avec l’amélioration motrice des symptômes de bradykinésie et rigidité. L’hypothèse serait que les oscillations bêta pourraient être un physiomarqueur spécifique des symptômes moteurs de la MP. L’objectif de cette thèse est de comprendre l’activité du NST et de trouver un physiomarqueur de la MP : au repos et pendant un paradigme Go/Nogo.Dans une première partie, pour 55 patients, nous avons modélisé la puissance avec un modèle linéaire à effets mixtes dans six bandes de fréquences selon la position des électrodes, la présence de complications liées au traitement dopaminergique, de la sévérité des symptômes, et du changement de ces symptômes suite à la prise du traitement dopaminergique. Le changement le plus important entre OFF- et ON-DOPA est observé dans la bande low-bêta. Cependant, nous n’avons pas trouvé de corrélation entre la sévérité de la bradykinésie, rigidité, ou du score axial avec l’activité low-bêta. Mais, nous avons trouvé que la sévérité de la rigidité était associée aux fréquences supérieures à 20Hz. Nous avons également trouvé une corrélation positive entre la sévérité du tremblement dans toutes les bandes de fréquences, qui est très importante pour la bande thêta. Etonnement, nous avons trouvé que les complications liées au traitement dopaminergique pouvaient induire des informations dans toutes les bandes. Cet effet est très important pour la bande thêta. Nos résultats au repos suggèrent que l’activité bêta n’est pas le physiomarqueur idéal de la MP.Pendant le paradigme Go/Nogo, pour 15 patients, nous avons modélisé la puissance avec un modèle linéaire à effets mixtes pour toutes les fréquences selon la position des électrodes, la sévérité des symptômes, les données comportementales et cinétiques. Nous avons trouvé une spécificité temporelle et spatiale des activités du NST au cours du paradigme. Nos résultats montrent que le NST est impliqué dans le contrôle des différentes étapes du mouvement volontaire du membre supérieur et que le défaut de dopamine dans le circuit sensorimoteur des GB génère une série de changements oscillatoires au sein du NST lors du mouvement. Cependant, l'activité thêta est apparue comme une activité caractéristique en OFF-DOPA fortement corrélée aux paramètres comportementaux et cinétiques. Cette activité pourrait être le physiomarqueur de la MP pendant l'exécution d'un mouvement, son augmentation pourrait induire une réponse motrice et un mouvement rapide. / The Parkinson's disease is a secondary neurodegenerative disease in the death of the dopaminergic neurons, pulling a dysfunction of the ganglions of the base (GB), the set pits under cortical implied in the control of the motricity...

Le but de cette étude était d'évaluer le rôle des boucles des ganglions de la base et des voies dopaminergiques sur les mécanismes « bottom-up » et « « top-down » du contrôle de l'attention visuelle (AV). Nous avons comparé les performances sur 3 tâches informatisés, appropriées à l'étude de la capture attentionnelle (CA), des mécanismes de sélection de la réponse motrice et d'initiation du mouvement, de deux groupes de patients avec maladie de Parkinson (MP) - un groupe étant évalué dans trois différentes conditions de stimulation électrique (sans stimulation, ou stimulation sélective de la partie sensorimotrice, SM, ou de la partie associative, AS, du noyau subthalamique, NST), l'autre groupe étant évalué dans deux différentes conditions de traitement médical (avec ou sans levodopa) - avec celles d'un groupe des sujets contrôles. Nos résultats suggèrent dans la MP un affaiblissement des mécanismes « top-down » de contrôle de l'AV, ce qui pourrait aussi expliquer indirectement l'augmentation de la CA. Le traitement dopaminergique est efficace dans le rétablissement des mécanismes « top-down » de l'AV, suggérant une implication des voies dopaminergiques dans ce domaine cognitif. Ces voies semblent aussi jouer un rôle dans les mécanismes « bottom-up » de l'attention, comme l'a suggéré le renforcement de la CA sous traitement dopaminergique. La stimulation du NST a montré un effet similaire à celui obtenu par un traitement dopaminergique, en favour d'une implication directe des boucles des ganglions de la base dans le contrôle de l'AV. Nos résultats ont mis en évidence une spécialisation fonctionnelle de différents sous-territoires du NST en ce qui concerne les mécanismes de « top-down ». La stimulation SM produit des effets marqués sur les processus d'initiation de mouvement et des effets positifs sur les mécanismes endogènes de l'AV, alors que la stimulation de la partie AS semble être plus particulièrement efficace dans l'amélioration des mécanismes de sélection de cible. / We aimed to investigate the possible role of cortico-basal ganglia loops and dopaminergic pathways in the mechanisms of top-down and bottom-up control of visual attention (VA). We compared the performances on 3 computerized tasks, respectively suitable to study attentional capture (AC), motor response selection and movement initiation, of two groups of patients with Parkinson's disease (PD), one evaluated in different sets of electrical stimulation (without stimulation, or selective stimulation of the sensorimotor, SM, or associative, AS, parts of the subthalamic nucleus, STN), the other in different conditions of medication (with or without levodopa), with those of a group of controls. Our results showed that in PD there is a weakening of the mechanisms underlying the top-down control of VA, which also would account indirectly account for the enhancement of AC. Dopaminergic treatment proved to be effective in restoring the top-down mechanisms of VA, suggesting an involvement of dopaminergic pathways in this cognitive domain. These pathways seem to play a role also in the bottom-up mechanisms of attention, as suggested by the enhancement of AC under dopaminergic treatment. The STN-stimulation showed a similar effect to that obtained by dopaminergic treatment, establishing a direct involvement of the basal ganglia loops in VA control. Our results highlighted a functional specialization of different sub-territories of the STN in relation to the top-down mechanisms. SM stimulation produced marked effects on the movement initiation processes and appreciable positive effects on endogenous VA mechanisms, while AS stimulation seems to be especially effective in improving the mechanisms of target selection.

Par sa connectivité directe avec le cortex, le noyau subthalamique (NST) représente une des structures d'entrée du système des ganglions de la base, et se trouve impliqué dans différents aspects du comportement (contrôle moteur, cognitif et limbique). Néanmoins, les corrélats électrophysiologiques de ces processus restent débattus. Les études effectuées dans le cadre de cette thèse visent à éclaircir le rôle possible du NST dans trois fonctions exécutives, à savoir l'inhibition réactive (suppression d'un mouvement programmé), l'inhibition proactive (préparation à inhiber son mouvement) et l'attention soutenue. Pour ce faire, les activités extracellulaires et/ou en potentiels de champs locaux du NST ont été enregistrées chez 28 patients parkinsoniens pendant qu'ils effectuaient des taches cognitives, visant à dissocier les corrélats de ces différentes fonctions exécutives. Dans une première étude, les activités en potentiel de champs locaux du NST lors de l'inhibition réactive et proactive ont été étudiées à l'aide d'un paradigme modifié du « stop signal ». L'inhibition réactive se caractérise par une augmentation rapide de synchronisation relative de l'activité du NST dans la bande de fréquence β (13-35 Hz), tandis que l'inhibition proactive se caractérise par la maintenance tonique d'un niveau élevé d'activité β qui prédit les performances des patients lors de l'inhibition réactive. Dans la seconde étude, nous avons montré qu'une population neuronale (n=7 neurones) augmente rapidement sa fréquence de décharge lors de l'inhibition réactive. Enfin, dans la troisième étude, nous avons utilisé un paradigme permettant de moduler le niveau attentionnel requis pour réaliser un comportement simple. Nos résultats indiquent qu'une baisse d'activité β est observée uniquement lorsque le sujet maintient une attention soutenue pour encoder, retrouver en mémoire une information afin de produire une réponse. L'ensemble de ces résultats nous ont permis d'apporter des preuves électrophysiologiques de l'implication du NST dans ces différentes fonctions et de clarifier la dynamique temporelle des activités neuronales supportant ces processus. Ils suggèrent ainsi l'hypothèse d'une implémentation de différents aspects du contrôle exécutif dans le NST via des mécanismes communs et interactifs dont la dynamique temporelle permettrait la modulation fine du comportement. / The subthalamic nucleus (STN) is an input structure of the basal ganglia implicated in many behavioral processes (motor, cognitive and limbic control). However the electrophysiological correlates of these processes remain unclear. This thesis aims to clarify the role of the STN during 3 executive functions: reactive inhibition (suppression of a prepotent move), proactive inhibition (preparation to inhibit a move) and sustained attention. To this end, extracellular and local field potential activities were recorded in 28 patients with Parkinson's disease while they performed cognitive tasks, aiming to dissociate the neural correlates of these executive functions In a first study, local field potentials β (13-35 Hz) activity was recorded in the STN during reactive and proactive inhibition. Reactive Inhibition was related to a relative increase of β activity, while proactive inhibition was related to maintenance of a tonic level of β activity predictive of reactive inhibitory performances. In a second study, we showed that reactive inhibition is related to a phasic increase of firing rate in a neuronal subpopulation (n=7 neurons). In a third study, we recorded Local field potentials in the STN while patients performed a sustained attention-demanding task (combining a visual search and a delayed match-to-sample paradigm) and found a systematic suppression of 15-35 Hz activity during each repetition of the task directly related to the amount of attention allocated by the participants. Altogether, these results present electrophysiological evidences of the implication of the STN in these functions and clarify the temporal dynamics of neuronal activities supporting these processes. These results may suggest an implementation of various executive functions in the STN via common and interactive mechanisms which temporal dynamics would mediate behavioral control.

Gamma oscillations of local field potentials (LFP) in the basal ganglia and thalamus had not been studied during saccades. Eleven patients were studied during deep brain stimulation (DBS); 6 were in the subthalamic nucleus (STN); 3 in the globus pallidus interna (GPi); and 2 in the thalamic ventralis intermedius nucleus (Vim). Patients performed horizontal saccades to visual targets while LFPs from DBS electrodes, scalp electroencephalogram (EEG), and electrooculogram (EOG) were recorded. Wavelet spectrograms were generated and saccade onset and event-related gamma synchronizations (ERS) were compared to baseline without eye motion. ERS were recorded at and after saccade onset in the STN, GPi and Vim, EEGs and EOGs; but were absent during target light illumination without saccades. ERS were symmetric in all DBS contacts and appeared identical in DBS LFPs, frontal EEGs and EOGs. These findings indicate their origin from extraocular muscle spike potentials rather than brain neural activity.