The functional anatomy of cerebral reorganization in the human sensorimotor system

• Main Author: Ward, Nick
• Published: University College London (University of London) 2004
• Subjects: 612.8255
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.405840

Functional imaging techniques allow assessment of distributed regional brain activation in humans during the performance of specified tasks. To date, studies of the motor system in previously hemiparetic stroke patients have demonstrated task related brain activation over and above control subjects in a number of brain regions usually only recruited during more complex motor tasks. However, as most of these studies were performed in patients with good recovery, the relationship between these findings and the recovery process remained unclear. The purpose of the experiments in this thesis was to establish whether such a relationship exists by using functional magnetic resonance imaging (fMRI). All experiments employed an isometric dynamic hand grip task (with visual feedback of the force exerted). Hand grip may be usefully performed by stroke patients with poor motor function. In the first experiment, I characterised the normal functional anatomy of hand grip in 26 control subjects. The task activated regions within the recognised motor network, and in a putative human `grasping network', involving rostral ventral premotor cortex (Brodmann area 44) and intraparietal sulcus. I was then able to make comparisons between task-related activations in chronic stroke patients with a wide range of outcomes. By setting the target force as a proportion of each patient's maximum grip strength, I was able to equalise effort exerteda crossp atients. Outcome was assessedu sing nine outcomem easures,a nd a single `relative' score was calculated for each patient using a principal component analysis. This approach revealed that patients with poorer outcome activated a number of brain regions over and above those seen in the normal population. Patients with good outcome were less likely to do so. More formally, there was a negative linear correlation between task related activation in these regions and outcome. In order to observe the evolution of motor activation patterns as a function of recovery, I performed longitudinal studies using the same motor paradigm over the first 8-12 months following stroke. In all patients, there was a negativecorrelation between recovery and task-related recruitment of several brain regions, mainly within the distributed motor system. Furthermore, I was able to demonstrate a negative linear correlation between initial severity (at 10-14 days post stroke) and task related activation in the same regions. These studies demonstrate a clear relationship between the degree of recovery and task related activation (both within and across subjects) in regions such as premotor cortex, supplementary motor areas, posterior parietal cortex and cerebellum, as well as in both ipsilesional and contralesional primary motor cortex (M1). All of these regions are involved in neural circuits with projections to both the motor output part of spinal cord, as well as to M1 itself. The increased activity may represent an alternative but less efficient means of generating motor output. A reduction in activation over time suggests that increasing efficiency in neural pathways and networks in these regions underlies the improvement in performance of a simple motor task, similar to the results of motor learning studies. Lastly, by studying the same motor task across a group of normal subjects with a wide variety of ages, I was able to demonstrate that normal older subjects were also more likely to activate a more widespread motor network, in order to maintain performance. This fording is of significance, when considering that work in animal models suggests that the capacity for adaptive change is finite. There is a need therefore to establish an empirical understanding of how the brain responds to injury in relationship to recovery, and how other parameters,in particular age, have an effect on this response. 4