Personalized metabolic connectome analysis reveals glucose heterogeneity in ischemic cerebrovascular disease.

PURPOSE: This study aimed to establish individualized metabolic networks for patients with ischemic cerebrovascular disease (ICVD) using cerebral glucose metabolism data and to analyze topological alterations before and after surgical intervention.

METHODS

We enrolled 31 surgically treated ICVD patients with unilateral cerebral infarction (21 with complete postoperative follow-up) and 17 normal controls (NC). We constructed individualized brain metabolic networks from [18F]-fluoro-2-deoxy-D-glucose positron emission tomography ([18F]FDG PET) scans using Kullback-Leibler divergence. In the NC group, subnetworks were defined as the left intrahemispheric, right intrahemispheric and interhemispheric networks. In the ICVD group, after hemispheric alignment based on the surgical side, subnetworks were redefined as the surgical intrahemispheric, non-surgical intrahemispheric, and interhemispheric networks. Then, graph-theoretical parameters were calculated to derive the metabolic connectivity expression score (MCES). Finally, we used Kruskal-Wallis tests, paired t-tests, and Spearman correlation analysis to assess topological differences and their relationship with National Institutes of Health Stroke Scale (NIHSS) scores.

RESULTS

The preoperative ICVD group showed significantly altered metabolic connectivity compared with controls, with higher global MCES values (0.83 vs. 0.32, p < 0.0001). Ipsilesional subnetworks exhibited reduced connectivity, whereas contralesional subnetworks showed increased connectivity. Subnetwork analysis showed that the interhemispheric network in the ICVD group had a mean MCES of 0.91, which was significantly higher than that of the interhemispheric network in the NC group (0.17, p < 0.0001). Postoperatively, network connectivity showed partial recovery, particularly within the interhemispheric network. MCES was significantly correlated with NIHSS scores in the global, non-surgical intrahemispheric, and interhemispheric networks (|r|=0.58 ~ 0.65, p < 0.05), but not in the surgical intrahemispheric network.

CONCLUSION

ICVD is associated with disrupted metabolic connectivity that exhibits early postoperative remodeling following surgical revascularization. Changes in interhemispheric network topology may provide insight into early postoperative metabolic network reorganization.