P16 New Insights Into Brain Energy Metabolism in Type C Hepatic Encephalopathy: A Dual 18F-FDG PET and 9.4T 1H MRS Preclinical Study

Background and Aim: Type C hepatic encephalopathy (HE) is a severe neuropsychiatric decline which occurs as a consequence of chronic liver disease. Energy metabolism alterations have been suggested in HE, but in vivo studies have reported conflicting results. Here, we propose an atlas-based co-registration method of quantitative preclinical 18F-FDG PET maps with localized 1H MR spectroscopy (MRS), providing regional and quantitative information on both brain glucose uptake and neurometabolic profiles. Methods: MRS experiments on a 9.4T MRI system (SPECIAL sequence, TE=2.8ms) using BDL rats, model of type C HE, were performed at week 0 (used as self-control) and 6 post-surgery on 2 brain regions, the hippocampus (week 0/6: n=4/9) and the cerebellum (week 0/6: n=3/4). PET experiments on BDL (n=10) and SHAM (n=8) rats were conducted at week 6, using a LabPET-4 preclinical scanner. Quantitative brain maps of glucose cerebral metabolic rates (CMRglc) were estimated using Sokoloff’s approach and registered to an MRI atlas for regional comparison with 1H MRS metabolites’ concentrations. Results: A significant increase in brain glutamine in both regions, stronger in the cerebellum than in the hippocampus (115% versus 73%), and a decrease in glutamate and in the main osmolytes (Ins, Tau, tCr) were observed using 1H MRS. Moreover, Asc decreased in the hippocampus and GABA in the cerebellum. After co-registration, a 2-fold lower CMRglc was observed in BDL versus SHAM rats in both brain regions. In contrast, the standardized uptake value (SUV), a more common metric in PET studies, showed no significant difference between the 2 groups for any of the brain regions because it overlooks inter-subject differences in systemic metabolism. Conclusion: We report the first multimodal 18F-FDG PET and 1H MRS in vivo study in the type C HE BDL rat model. In addition, the refined FDG PET analysis proposed here enables a quantitative and regional measurement of the cerebral metabolic rate of glucose. Our results suggest local alterations of energy metabolism in the cerebellum and in the hippocampus of BDL rats, concomitant with a previously reported glutamine-associated metabolic impairment, and emphasize the importance of PET quantification methods that account for systemic metabolism differences (see Figure 1).Figure 1.: Atlas-based co-registration of 1H MRS and 18F-FDG PET. A. Atlas regions with the circled labels corresponding to the cerebellum and hippocampus. B. 18F-FDG PET CMRglc values averaged over the atlas labels of the cerebellum and hippocampus for BDL and SHAM rats, with an example of brain segmentation on the PET image, ****: P<0.0001, Student’s t-test. C. 1H MRS metabolites quantifications in the cerebellum and hippocampus with an example of voxel location, % change of mean metabolites’ concentrations between week 0 (before surgery) and week 6 for BDL rats, only statistically significant changes are displayed (P<0.05, Student’s t-test).