Abstract
P1388
Introduction: Nonalcoholic steatohepatitis (NASH) is a metabolic disease that affects many patients worldwide. PET with 18F-fluorodeoxyglucose (FDG) has the potential to characterize the disordered energy metabolism in NASH but has not been feasible to show this ability using standard static imaging or dynamic imaging with a standard temporal resolution (e.g., 10-20s/frame). Early work using dynamic FDG-PET showed measuring the change in glucose transport but not in glucose metabolism [1]. Herein, we demonstrate the use of high-temporal resolution (2s/frame) dynamic imaging and kinetic modeling enabled on total-body PET to characterize organ metabolism in healthy subjects and those with NASH.
Methods: Fourteen healthy subjects and ten with nonalcoholic fatty liver disease (NAFLD) were included in this study and scanned on a uEXPLORER total-body PET/CT system for one hour. Histologically-proven NASH was defined by NAFLD activity score (NAS) greater than 4. Prior IRB approval and informed consent were obtained. The data were reconstructed into a total of 66 time frames: 30x2s, 12x10s, 6x30s, 12x120s and 6x300s. Regions of interest (ROI) were placed in the descending aorta to extract an image-derived blood input function, and in the liver, gray matter, and spleen to extract tissue time activity curves (TACs). Both the conventional two-tissue (2T) model, which combines the interstitial space and intracellular space into a free-state space (Fig. 1), and the proposed three-tissue (3T) model, which separately models the interstitial space, were used to fit the high-temporal resolution TACs. Furthermore, the effect of dual blood supply was modeled with the 2T and 3T models for liver kinetic modeling. Standardized uptake value (SUV) at one hour post injection and the net influx rate Ki estimated with the 2T and 3T models were used to characterize glucose metabolism and compared between the non-NASH group (19 subjects, including healthy subjects and NAFLD with a NAS ≤ 4) and NASH group (5 subjects) using statistical T-tests.
Results: The 3T model showed an improved TAC fitting in all organs as compared to the 2T model (Fig. 2A, Fig. 3A, Fig. 4A) and had a lower Akaike information criterion (AIC) value in all subjects. The 3T Ki of the liver was higher in NASH than in non-NASH (P = 0.035), which may reflect the increased energy metabolism of hepatic macrophages in NASH. Liver SUV and 2T Ki were not statistically different between the two groups (P > 0.3) (Fig. 2). A metabolic difference was also observed in the spleen between NASH and non-NASH, but only by 3T Ki, not by SUV or 2T Ki (Fig. 3). The increased metabolism in the spleen of patients with NASH can be partly explained by the fact that spleen is one of the major immune organs and its activity may be stimulated by distant inflammation. In the gray matter, SUV and 2T Ki did not differ significantly between NASH and non-NASH (P > 0.1), but the improved fitting by the 3T model led to a significantly lower Ki in NASH (P = 0.005) (Fig. 4). The result suggests a reduced energy metabolism in the brain of patients with NASH, in align with the hypothesis that neurodegeneration may occur in NAFLD and NASH [2].
Conclusions: This study explored total-body FDG-PET quantification of glucose metabolism in the liver, spleen and brain by using high-temporal resolution dynamic imaging. FDG-PET has the ability to characterize the changes of metabolism in NASH but it is only possible with a 3T compartmental model that accounts for the interstitial space, not by a conventional 2T model or SUV. The reported results suggest that NASH results in increased energy metabolism in the liver and spleen but decreased metabolism in the brain.
References.
[1] Wang et al, Phys. Med. Biol. 63(2018) 155004.
[2] Mondal et al, Journal of Neuroinflammation 201(2020).
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.