Differentiation of Cerebellum-Type and Parkinson-Type of Multiple System Atrophy by Using Multimodal MRI Parameters

Abstract

Recent studies have demonstrated the structural and functional changes in patients with multiple system atrophy (MSA). However, little is known about the different parameter changes of the most vulnerable regions in different types of MSA. In this study, we collected resting-state structure, perfusion, and patients with functional magnetic resonance imaging (fMRI) data of cerebellum-type of MSA (MSA-c) and Parkinson-type of MSA (MSA-p). First, by simultaneously using voxel-based morphology (VBM), arterial spin labeling (ASL), and amplitude of low-frequency fluctuation (ALFF), we analyzed the whole brain differences of structure, perfusion, and functional activation between patients with MSA-c and MSA-p. Second, we explored the relationships among structure, perfusion, function, and the clinical variables in patients with MSA. Finally, we extracted the MRI parameters of a specific region to separate the two groups and search for a sensitive imaging biomarker. As a result, compared with patients with MSA-p type, patients with MSA-c type showed decreased structure atrophy in several cerebella and vermis subregions, reduced perfusion in bilateral cerebellum_4_5 and vermis_4_5, and an decreased ALFF values in the right lingual gyrus (LG) and fusiform (FFG). Subsequent analyses revealed the close correlations among structure, perfusion, function, and clinical variables in both MSA-c and MSA-p. Finally, the receiver operating characteristic (ROC) analysis showed that the regional cerebral blood flow (rCBF) of bilateral cerebellum_4_5/vermis_4_5 could differentiate the two groups at a relatively high accuracy, yielding the sensitivity of 100%, specificity of 79.2%, and the area under the curve (AUC) value of 0.936. These findings have important implications for understanding the underlying neurobiology of different types of MSA and added the new evidence for the disrupted rCBF, structure, and function of MSA, which may provide the potential biomarker for accurately detecting different types of patients with MSA and new ideas for the treatment of different types of MSA in the future.