Investigating the regional correlation of subcortical structures as imaged by DAT SPECT with clinical phenotypes in Parkinson’s disease

Objectives To apply a novel image analysis method to SPECT imaging of Parkinson’s disease (PD) that estimates population-based templates for each structure individually to investigate the regional correlation of subcortical structures with PD clinical phenotypes.

Methods A surface registration algorithm was used, under the broad class of Large Deformation Diffeomorphic Metric Mapping. We analyzed 128 subjects: 48 healthy controls (HCs) and 80 PDs. A common coordinate system for the population based on MR images was obtained, and DAT SPECT intensities across subjects were compared. This method has the advantage of estimating population-based templates from patients for each structure individually rather than using a predetermined collective atlas for all regions based on HCs, as is customary. Pattern recognition using Principal Component Analysis (PCA) was applied to SPECT intensities to obtain sub-regions with highest variance across subjects. Since all structures are registered to a common template, we specifically investigated the correlation of any sub-regions in subcortical structures especially caudate & putamen with a range of disease symptoms/scores, e.g. the unified PD rating scale (UPDRS), gait, tremor, bradykinesia and rigidity.

Results Results showed excellent discrimination between HCs and PDs. The main finding was that our effective normalization methodology enabled pattern recognition analysis to generate new markers of disease progression (PC components), and allowed extraction of regional correlation of subcortical structures such as caudate and putamen with PD clinical features. Higher negative correlations were observed with rigidity and bradykinesia than with tremor.

Conclusions The presented sub-structure analysis has the potential to provide novel insights into the relationship between dopaminergic alterations and PD phenotypes, including regional involvement of subcortical structures.