This work investigated whether (18)F-FDG PET standardized uptake value (SUV) is stable over time in the normal human liver.
Methods: The SUV-versus-time curve, SUV(t), of (18)F-FDG in the normal human liver was derived from a kinetic model analysis. This derivation involved mean values of (18)F-FDG liver metabolism that were obtained from a patient series (n = 11), and a noninvasive population-based input function was used in each individual.
Results: Mean values (±95% reliability limits) of the (18)F-FDG uptake and release rate constant and of the fraction of free tracer in blood and interstitial volume were as follows: K = 0.0119 mL·min(-1)·mL(-1) (±0.0012), k(R) = 0.0065·min(-1) (±0.0009), and F = 0.21 mL·mL(-1) (±0.11), respectively. SUV(t) (corrected for (18)F physical decay) was derived from these mean values, showing that it smoothly peaks at 75-80 min on average after injection and that it is within 5% of the peak value between 50 and 110 min after injection.
Conclusion: In the normal human liver, decay-corrected SUV(t) remains nearly constant (with a reasonable ±2.5% relative measurement uncertainty) if the time delay between tracer injection and PET acquisition is in the range of 50-110 min. In current clinical practice, the findings suggest that SUV of the normal liver can be used for comparison with SUV of suspected malignant lesions, if comparison is made within this time range.