Glucodynamic Network Transitions of the Human Brain

Variability in neuronal firing patterns enables the brain to flexibly transition between network states, a hallmark of adaptive information processing. Yet the energetic basis of this flexibility remains unknown. Here, we report that the capacity of the brain to shift between network states is contingent on the moment-to-moment variability of glucose metabolism. Using simultaneous resting-state functional FDG-PET and BOLD-fMRI in 78 healthy adults (20-86 years), we quantified brain metabolic and haemodynamic variability, complexity, and dynamic network connectivity. Individuals with greater variability and complexity of glucose metabolism demonstrated a broader repertoire of network states, marked by globally connected and influential central hubs. Individuals who achieved a broader range of network states had better cognitive performance. In older adults, widespread reductions in dynamic metabolic connectivity were observed. Glucodynamic parameters predicted age and cognitive performance more robustly than haemodynamic measures. These findings reveal that functional network dynamics are emergent properties of the brains metabolic architecture and that the organisation of large-scale neural circuits is constrained by energetic principles. This understanding may provide critical new insights into the metabolic basis of brain ageing, neurodegeneration and psychiatric conditions.