Parkinson's disease (PD) is a common neurodegenerative disease with a reduction of neuromelanin (NM) in the substantia nigra (SN). Using T1-weighted sequences, several magnetic resonance (MR) studies demonstrated reduced surface and volume of SN in PD and decreased signal intensity of SN. Motor system dysfunction is a main complaint in PD. Some functional magnetic resonance (fMRI) studies showed altered brain activation patterns during motor learning in PD, however long-term effects of motor learning in PD have not been yet explored with fMRI. In our study, T1-weighted images with spectral presaturation with inversion recovery (SPIR) pulse were acquired on a 3T MR scanner to investigate structural changes of SN. Structural changes of SN were quantified with manual and novel semi-automatic atlas-free local statistics signature-based segmentation methods. Receiver operating characteristic (ROC) analysis was performed to determine the sensitivity and specificity of both methods. fMRI activation patterns in PD were assessed over a period of 90 days. fMRI data was acquired during execution of finger-tapping sequence that was composed of eight finger taps. Between Day 0 and Day 28, subjects were asked to daily train finger-tapping motor sequence at home. Between Day 28 and Day 90, subjects were instructed to stop with the training at home. fMRI data was acquired on Day 0, Day 28 and Day 90. PD patients had lower surface (37.7 ± 8.0 vs. 56.9 ± 6.6 mm2) and volume (235.1 ± 45.4 vs. 382.9 ± 100.5 mm3) of SN than healthy controls. For surface, sensitivity and specificity were 91.7 % and 95 %, respectively. For volume, sensitivity and specificity were 91.7 % and 90 %, respectively. In fMRI part of the study, higher activation of parahippocampal gyrus and hippocampus were detected on Day 0 in patients with PD in comparison to healthy controls. On Day 28 and Day 90, lower activation was found in PD patients in inferior parietal lobe, posterior cingulate cortex and precuneus. On Day 28, higher activation in parahippocampal gyrus was noted as well. Manual and semi-automatic segmentation methods of SN reliably distinguish between PD patients and HC. ROC analysis shows high sensitivity and specificity of both methods. In the fMRI study, several areas with higher activation in PD could represent compensatory mechanisms in PD. In later stages of motor memory formation, lower activation may reflect general dysfunction of executive functions in PD.
|