Altered spontaneous brain activity in type 2 diabetes mellitus: an activation likelihood estimation Meta-analysis

Objective This meta-analysis aims to identify consistent results of vulnerable brain regions through the existing resting-state fMRI studies, thus exploring the changes of neural spontaneous brain activity in type 2 diabetes mellitus (T2DM) patients. Methods A systematic analysis of studies on brain resting-state changes in T2DM patients using ALFF, fALFF and ReHo analysis . The meta-analysis was based on the activation likelihood estimation method, using the software of Ginger ALE 2.3. Results Ten studies from 7 references (188 T2DM patients and 170 healthy controls) were included. Based on the analysis of ALFF and ReHo data, this meta-analysis identi fied the robust reduction of resting-state spontaneous brain activity in T2DM patients, including the left lingual gyrus of occipital lobe, right cerebellum posterior lobe, left postcentral gyrus and right insula (cluster size= 800, 488, 368, 256 mm3, P<0.05 after FDR correction), while no increased spontaneous brain activation was found in any regions. The meta-analysis from ReHo studies showed reduced resting-state spontaneous brain activity in the left lingual gyrus of occipital lobe, left postcentral gyrus, right insula and posterior cingulum/lingual gyrus (cluster size=832, 368, 280, 232 mm3, P<0.05 after FDR correction), while no increased spontaneous brain activation was found in any regions. Conclusion This meta-analysis study using the activation likelihood estimation method demonstrated that the resting-state spontaneous brain anomalies in T2DM patients might contribute to exploring machenism underlying diabetic encephalopathy. Key words: Diabetes mellitus, type 2; Magnetic resonance imaging; Meta-analysis; Activation likelihood estimation method