Longitudinal volumetry and shape analysis of the striatum in a cohort with leukoaraiosis : a pilot study

Background: Diffuse white matter changes (or 'leukoaraiosis') on brain MRI are a common finding, particularly in the elderly, and are linked to cognitive impairment, gait problems, depression and transition to disability. The areas of the brain that are affected by leukoaraiosis have the potential to disrupt fronto-striatal brain circuits that govern many cognitive functions, but no studies have looked at the striatum in this group - studies in a number of neurodegenerative diseases utilising striatal size measurement are outlined as a template for a study in a leukoaraiosis group. Methods: This study analysed caudate nucleus and putamen volumes and shapes in relation to the severity of leukoaraiosis, cognitive, motor and mood-related outcomes, using MRI and clinical data from a subset of the the European multicentre LADIS (Leukoaraiosis and Disability in the Elderly) study. Striatal volumes and shapes were derived from manual tracing of the structures on T1-weighted MRI scans. The scans of 66 subjects were traced, including scans taken at study entry and three years later on the same subjects. Results: There was no significant correlation between severity of leukoaraiosis and striatal volumes, nor was there significant longitudinal change in striatal sizes over three years. Regression analysis of clinical outcomes showed a significant correlation between smaller caudate nuclei and poorer performance on measures of executive functioning both at study baseline (beta=0.335, p=0.017) and at 3-year follow-up (beta=0.423, p=0.002), although no specific areas of the caudate were notably changed on shape analysis. There was also a significant correlation between caudate size and performance on baseline measures of gait and balance (beta=0.339, p=0.004), with shape analysis showing left caudate changes corresponding with inputs of the dorsolateral prefrontal, premotor and motor cortex. This effect was not seen at follow-up. Analysis of striatal sizes with measures of depression and disability did not show any significant correlation. Discussion: This study shows that the structure (size and shape) of the caudate nucleus is altered in tandem with measures of cognition, gait and balance. The possibilities of direct damage via the leukoaraiosis disease process, 'deafferentation' from severing of afferent white matter tracts, and other explanations are explored. Limitations of the study include a less impaired cohort selected due to study design, as well as potential low power, both of which would predispose to Type 2 error. Conclusion: Striatal measurement has promise as a biomarker of different cognitive functions in clinical populations. Shape analysis in particular has potential to isolate different circuits which are dysfunctional in disease, allowing further progress to be made on determining their pathophysiology. Further avenues for research include automated tracing of larger datasets of MRI scans to improve statistical power