Do glial-reactivity and cerebral blood flow modulate cerebral glucose metabolism in Alzheimer’s disease?

Abstract

Background Alzheimer’s disease is a devastating neurodegenerative disorder with a complex pathogenesis. One main pathological feature utilised in diagnosis is neurodegeneration or neuronal injury, which is reflected in reductions in cerebral glucose metabolism measured by [18F]Fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET). Here we evaluated the involvement of glial reactivity measured with magnetic resonance spectroscopy (MRS) and cerebral blood flow measured with arterial spin labelling (ASL) on [18F]FDG PET as a measure of cerebral glucose metabolism. Method 123 people living with early Alzheimer’s disease who completed baseline evaluations on the evaluating liraglutide in Alzheimer’s disease trial were enrolled. Participants completed [18F]FDG PET scans with arterial input, T1 weighted MRI, single-voxel 1HMRS, and pulsed ASL scans at Imperial College London Clinical Imaging Facility. The Totally Automatic Robust Quantitation in NMR (TARQUIN) package was used to process MRS scans and identify the concentration of myo-inositol within the posterior cingulate cortex (PCC), a marker of glial activation. Oxford-ASL was utilised to process ASL and quantify cerebral blood flow in the PCC. Finally, spectral analysis was performed on the [18F]FDG PET scans to assess the cerebral metabolic rate of glucose in the PCC. Result Pearson’s correlations were performed between the cerebral metabolic rate of glucose, cerebral blood flow and glial activity measured by the level of myo-inositol in the PCC. Increased cerebral glucose metabolism was correlated with higher myo-inositol in this sample of Alzheimer’s disease participants. In contrast, cerebral blood flow was not associated with cerebral glucose metabolism. Conclusion Here we demonstrate that increased glial reactivity contributes to [18F]FDG PET signal in the early stages of Alzheimer’s disease. In response to early neuronal injury, astrocytes and microglia may become activated and enhance regional rates of glucose consumption. Hence, the contribution from these cells in addition to neurons should be considered in interpreting [18F]FDG PET as a measure of cerebral glucose metabolism. Interestingly, cerebral blood flow did not influence glucose metabolism. Microglia and astrocyte reactivity may contribute to an increase the cerebral glucose metabolism while neuronal loss and synaptic function may contribute to lower glucose metabolism measured by [18F]FDG in the early stages of Alzheimer's disease.