Abstract
1078
Introduction: Alterations of brain glucose metabolism are a well-established feature of patients with disorders of consciousness (DOC). However, most studies are conducted on relatively small samples (median n=10). Here, we aimed to quantitatively synthesize available evidence from studies investigating brain metabolism alterations in DOC, and to test whether such metabolic alterations distribute according to specific resting-state brain networks.
Methods: We conducted a systematic review and meta-analysis of FDG-PET studies published up to February 2020. Coordinate-based meta-analysis was performed via activation likelihood estimation in GingerALE, setting the statistical threshold to p<0.001 (voxel-level) and p<0.05 FWE-corrected (cluster-level). Distribution of resulting hypometabolic clusters was compared to the topography of well-known resting-state brain networks, using the FIND lab’s functional connectivity atlas (including subcortical areas).
Results: Thirteen studies for a total of 278 patients (unresponsive wakefulness syndrome (UWS, n=107); minimally conscious state (MCS, n=154); unspecified UWS/MCS, n=17) and 282 healthy controls, were selected. Meta-analysis revealed consistent reductions in glucose metabolism in patients with DOC, encompassing thalamus and caudatum, precuneus and posterior/middle cingulate cortex, and right middle frontal gyrus. Based on the FIND lab’s functional connectivity atlas, the reported metabolic reductions of DOC patients fell within the default mode network (40% of the assigned voxels), the basal ganglia/thalamus network (29%) and the executive control network (18%)(Figure 1). Separate meta-analysis of UWS and MCS groups revealed areas of hypometabolism in thalamus and posterior cingulate gyrus in both groups, with additional hypometabolic clusters in fronto-parietal regions in UWS patients only, and in the caudatum in MCS patients only. Conclusion: These results suggest that patients with DOC are characterized by consistent alterations in both fronto-parietal and subcortical brain networks, in coherence with the mesocircuit hypothesis for consciousness1,2.