Dynamic FDG PET imaging of the brain in a mouse model of dystonia

1554 Objectives Primary dystonia is a common but poorly understood movement disorder characterized by involuntary muscle contractions leading to abnormal movements and postures. A fraction of Rcn2 knockout mice (30%) develop early onset overt dystonia though most of them are apparently normal. The purpose of this study was to detect potential functional or anatomic abnormalities in the brain of dystonic mice with dynamic fluorine-18 fluorodeoxyglucose (FDG) PET. Methods FDG was administered to Rcn2 knockout mice with or without dystonia and imaged for 60 min. The dynamic FDG PET data were analyzed in a 3-compartment kinetic model and compared to the standard of care standardized uptake value (SUV) method. A model corrected blood input function (MCIF) was obtained from the left ventricular blood pool, which was corrected for partial volume averaging and spill-over contamination1. The MCIF was utilized in a 3-compartment kinetic model to obtain the kinetic rate parameters, K1-K4, and hence the net cerebral FDG influx constant, Ki (ml/min/g). Results Standard SUV analysis (imaging after 30 minutes post FDG injection for 20 minutes) did not show differences between dystonic and non-dystonic mice in the brain. In contrast, dynamic imaging indicated differences in the brain time activity curves for the dystonic mice compared to the non-dystonic ones, especially at the early time points. The elevated Ki and K1 measured in dystonia mice relative to controls were suggestive of leaky vasculature in the dystonic mouse brain, which was confirmed by histology. The average Ki for dystonic mouse brain were higher (0.019±0.007 ml/min/g) than the control brain (0.007±0.004 ml/min/g). Conclusions Dynamic FDG PET imaging in a compartment model was able to differentiate the dystonic brain from the non-dystonic brain while the SUV analysis was not. Our finding also suggested vascular leakage in the brain of mice with dystonia.