Statistical voxel-based analysis of [18F]FDG PET animal studies for the estimation of glucose metabolism in stress conditions

Aim: Aim of this study was to evaluate the effect of acute stress on rat brain metabolism using PET [18F]FDG and Foot-Shock (FS) paradigm. Regional stress effect was evaluated both by a statistical voxel-based analysis using a rat-model version of Statistical Parametric Mapping (SPM) optimized in our facility. Materials & Methods: 16 male CD rats, randomly assigned into two groups, 8 rats for Control Group (CG) and 8 rats for Stressed Group (SG), were injected with [18F]FDG in a lateral tail vein. Following the FS protocol, the SG was exposed, immediately after radiotracer injection, to the shock session. One hour after [18F]FDG injection, CG and SG were then subjected to a PET brain study for 30 minutes. All animals have been then sacrificed and corticosterone and circulating glucose were measured. A rat template was created using a control rat, outside of the two groups, who performed, in addition to the PET study, an and MRI study: the PET images for this rat, co-registered to his MRI, was used as [18F]FDG/MRI template. Statistical analysis were performed setting p<0.05, a threshold of 100 voxels and rescaling data to a mean glucose consumption value of 28mol/100g/min. In particular three SPM design were used: 1) a single subject design was used both to perform a Jack-Knife analysis on CG and to evaluate each SG rat with respect to CG; 2) a compare-populations model was used to evaluate differences in SG with respect to CG; 3) a simple regression model allowed to evaluate correlations between cerebral glucose metabolism and corticosterone levels in different brain regions. Results: Jack-Knife analysis on CG showed that one rat had a pattern of altered metabolism in comparison to the other control rats, suggesting a different susceptibility to experimental handling. This rat was excluded from analysis. Single-subject analysis on stressed rats revealed a similar pattern, except for one animal. Interestingly this rat escaped from box during stress paradigm. Group analysis showed a relative reduction of regional brain metabolism at the level of thalamus, hypothalamus, entorhinal and piriform cortex, and cerebellum, while an increase was observed in motor, somato-sensory cortex, olfactory bulb regions and hippocampal subregions. A negative correlation was observed between cerebral glucose metabolism and corticosterone levels in different brain regions. Conclusion: Results of the study indicatedthat acute stress modifies regional brain functions and these modifications are associated with circulating level of stress markers like corticosterone.