Voxel-based analysis of normal cerebral [18F]FDG uptake during childhood using statistical parametric mapping.

The changing pattern of relative cerebral (18)F-fluoro-2-deoxy-D-glucose (FDG) uptake that occurs during normal childhood development is not completely understood. Using SPM8 we undertook a voxel-based analysis of dedicated cerebral FDG scans in 28 children ranging in age from 11 months to 16 years to examine the effects of age on regional FDG uptake. The subjects included were children with suspected or proven extracranial malignancies without central nervous system metastases and no previous or current therapies or medical conditions likely to interfere with cerebral metabolism. The included cerebral FDG scans were considered to represent normal cerebral FDG distribution in a child of their age at the time of the scan. When normalised to whole brain mean uptake, the voxel-based analysis showed increasing FDG uptake with age in the premotor and prefrontal cortices, insula cortex, cingulate cortex, basal ganglia, thalamus, cerebellum and in small areas of the inferior temporal lobes and left Heschl's gyrus. These findings correlate with previous published analysis of the same data that used qualitative and maximal standardised uptake value (SUV(max)) analysis techniques. This data provides more regionally specific information and further supports the conclusion that relative cerebral FDG uptake in children has not reached a typical adult pattern by approximately one year of age but in fact changes throughout childhood. The results speak to the importance of using age-matched data or adjusting for age in the statistical analysis of studies comparing paediatric cerebral FDG scans to a control dataset to avoid bias due to different age distributions in the groups of subjects studied. The areas of increasing FDG uptake with age probably relate to underlying neuronal processes linked to normal neurodevelopment including key resting state networks.