Quantitative susceptibility mapping and cognitive impairment in Parkinson ’s disease.
Thesis DisciplineMedical Physics
Degree GrantorUniversity of Canterbury
Degree NameMaster of Science
Excess iron deposition in the substantia nigra (SN) has been linked to Parkinson’s disease (PD). Increased iron deposition may also be associated with PD duration and worsening motor impairment. Iron deposition in the SN and the almost inevitable progression to dementia in PD has not yet been investigated. Furthermore the link between PD with normal cognition (PD-N), those with mild cognitive impairment (PD-MCI) and dementia (PDD) and excess iron deposition in key grey matter nuclei of the basal ganglia (BG) other than the SN may be even more pertinent. In this thesis, I use a non-invasive, magnetic resonance imaging (MRI) methodology called quantitative susceptibility mapping (QSM) to indirectly estimate iron deposition in these grey matter nuclei and determine whether QSM values are associated with cognitive impairments in PD.
A total of 121 subjects were recruited for this study. Extensive neuropsychological testing was used to characterize participants. Of these 31 were PD-N, 56 were PD-MCI, 10 were PDD and 24 were age and sex matched healthy controls (HCs). Subjects were imaged using a multi-echo spoiled gradient echo. QSM was reconstructed from the raw images using non-linear morphology enabled dipole inversion and Laplacian boundary value background field removal. QSM values were extracted from the BG and red nucleus (RN) using two separate segmentations. Bayesian multi-level regression models were used to (1) test for differences between HCs and PD as a whole in each region of interest (ROI), and (2) to investigate any association between QSM values and cognitive impairment. Finally a whole brain analysis was performed to assess group differences on a voxel-by-voxel association between QSM values and cognitive score.
Relative to controls, the PD group showed significantly higher QSM values in the SNc. There were no significant group differences in QSM values across the other basal ganglia structures investigated. For the category analysis the left SNc for PD-N was the only significant difference observed. There was a weak positive correlation between the right RN and cognitive score with no other nuclei having any significant correlation. Whole brain analyses revealed no significant differences between the groups or association with cognition.
In this thesis, I confirmed that increased QSM, and hence iron accumulation, in SNc in PD is a robust finding, consistent with previous imaging and pathological studies on PD, however this finding was limited to the ROI analysis. It appears from the evidence we have gathered that iron deposition does not affect cognitive functioning. In conclusion, this thesis established QSM in a population of PD patients. While the group association agreed with past studies this thesis hints a number of possible improvements and new directions to further investigate cognition and QSM.