Dementia is very common in the late stage of patient with Parkinson's disease (PD). We aim to explore its underlying pathogenesis and identify candidate biomarkers using untargeted metabolomics analysis. Consecutive PD patients and healthy controls were recruited. Clinical data were assessed and patients were categorized into Parkinson's disease without dementia (PDND) and Parkinson's disease dementia (PDD). Fast plasma samples were obtained and untargeted liquid chromatography-mass spectrometry-based metabolomics analysis was performed. Based on the identified differentially-expressed metabolites from the metabolomics analysis, multivariate linear regression analyses and receiver operating characteristic (ROC) curves were further employed. According to the clinical data, the mean ages of PDND and PDD patients were significantly higher than those of healthy controls. The incidence of hypercholesterolemia was decreased in PDD patients. PDD patients also had lower levels of triglyceride, low-density lipoprotein cholesterol, and apolipoprotein B. There were 24 and 57 differentially expressed metabolites in PDD patients when compared with the healthy controls and PDND patients from the metabolomics analysis. Eleven lipid metabolites were simultaneously decreased between these two groups, and can be further subcategorized into fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, and prenol lipids. The plasma levels of the eleven metabolites were positively correlated with MMSE score and can be candidate biomarkers for PDD patients with areas under the curve ranging from 0.724 to 0.806 based on the ROC curves. Plasma lipoproteins are significantly lower in PDD patients. A panel of eleven lipid metabolites were also decreased and can be candidate biomarkers for the diagnosis of PDD patients. Lipid metabolic dysregulation is involved in the pathogenesis of Parkinson's disease dementia.
Keywords: Apolipoprotein; Dementia; Lipid; Metabolomics; Parkinson’s disease.