RBD related pattern based on 18F-FDG PET imaging using PCA analysis

Objectives Principal components analysis (PCA) has been used with 18F-FDG PET to establish and quantify disease related spatial convariance pattern. Although parkinson’s disease related pattern (PDRP) was established using PCA, the connection between PDRP and rapid eye movement (REM) sleep behavior disorder (RBD), the prodromal phase of PD, remains unknown. To observe the PDRP expression in preclinical stage of PD, we quantified PDRP scores in patients with RBD.We also used strictly defined statistical criteria to identify RBD-related pattern(RBDRP)to detect the possible brain network association between RBD and PD in this 18F-FDG PET study.

Methods 10 patients with RBD (age 62.1±6.8 years) and 10 age-matched (age 53.6±12.5 years) normal individuals were studied with resting brain 18F-FDG PET. Topographic profile rating (TRP) algorithm based on PDRP pattern and principal components analysis (PCA) were used to quantify PDRP scores in RBD patients and to detect specific metabolic pattern associated with RBD (RBDRP), respectively.

Results The expressions of PDRP in patients with RBD were increased compared with normal individuals, though the elevations were not as significant as those in PD patients. We also found that the features of RBDRP by PCA were demonstrated as bilaterally increased metabolism in the premotorcortex, putamen, thalamus and cerebellum, and decreased metabolism in the occipital cortex. RBDRP values for RBD patients were significantly elevated relative to those for normal individuals (P<0.001).

Conclusions These findings suggest that RBD is probably associated with PD and cerebral glucose metabolism features of RBDRP based on 18F-FDG PET imaging may contribute to differential diagnosis between RBD and normal individuals.

Research Support National Natural Science Foundation of China NO.81171189