The quantification and variability of skeletal muscle glucose utilization (SMGU) in healthy subjects under basal (low insulin) conditions are poorly known. This information is essential early in clinical drug development to effectively interrogate novel pharmacologic interventions that modulate glucose uptake. The aim of this study was to determine test-retest characteristics and variability of SMGU within and between healthy subjects under basal conditions. Furthermore, different kinetic modeling strategies were evaluated to find the best-fitting model to assess SMGU studied by 18F-FDG.
Methods: Six healthy male volunteers underwent 2 dynamic 18F-FDG PET/CT scans with an interval of 24 h. Subjects were admitted to the clinical unit to minimize variability in daily activities and food intake and restrict physical activity. 18F-FDG PET/CT scans of gluteal and quadriceps muscle area were obtained with arterial input. Regions of interest were drawn over the muscle area to obtain time-activity curves and standardized uptake values (SUVs) between 60 and 90 min. Spectral analysis of the data and kinetic modeling was performed using 2-tissue-irreversible (2T3K), 2-tissue-reversible, and 3-tissue-sequential-irreversible (3T5KS) models. Reproducibility was assessed by intraclass correlation coefficients (ICCs) and within-subject coefficient of variation (WSCV).
Results: SUVs in gluteal and quadriceps areas were 0.56±0.09 and 0.64±0.07. ICCs (with 90% confidence intervals in parentheses) were 0.88 (0.64-0.96) and 0.96 (0.82-0.99), respectively, for gluteal and quadriceps muscles, and WSCV for gluteal and quadriceps muscles was 2.2% and 3.6%, respectively. The rate of glucose uptake into muscle was 0.0016±0.0004 mL/mL⋅min, with an ICC of 0.94 (0.93-0.95) and WSCV of 6.6% for the 3T5KS model, whereas an ICC of 0.98 (0.92-1.00) and WSCV of 2.8% was obtained for the 2T3K model. 3T5KS demonstrated the best fit to the measured experimental points.
Conclusion: Minimal variability in skeletal muscle glucose uptake was observed under basal conditions in healthy subjects. SUV measurements and rate of glucose uptake values were reproducible, with an average WSCV of less than 5%. Compared with SUV, the 3-tissue model adds information about kinetics between blood, intra- and intercellular compartments, and phosphorylation that may highlight the exact mechanisms of metabolic changes after pharmacologic intervention.
Keywords: FDG PET/CT; kinetic modeling; reproducibility; skeletal muscle.
© 2015 by the Society of Nuclear Medicine and Molecular Imaging, Inc.