A standardized blood sampling scheme in quantitative FDG-PET studies

Area of Science:

• Neuroimaging
• Metabolic Research
• Positron Emission Tomography (PET)

Background:

• Quantitative estimation of brain glucose metabolism (rCMRGlc) via PET requires precise arterial input curves.
• Traditional methods necessitate frequent arterial blood sampling (≤5s intervals) to capture peak radioactivity.

Purpose of the Study:

• To develop and validate a standardized, reduced-sampling method for accurately defining the arterial input curve for rCMRGlc quantification.
• To assess the consistency of rCMRGlc values derived from fitted versus measured input curves.

Main Methods:

• A standardized 13-time-point sampling scheme was implemented.
• The standardized input curve was fitted using polynomial functions (rising phase) and spectral analysis (decreasing phase).
• rCMRGlc was calculated in 32 regions of interest in 20 volunteers using measured, standardized, and fitted input curves.

Main Results:

• The standardized input curve accurately defined the radioactivity delivery.
• rCMRGlc estimations using the fitted input curve showed minimal relative error (1-2%) compared to measured curves.
• The input-curve-fitting technique proved robust, independent of peak time occurrence.

Conclusions:

• A reduced arterial blood sampling scheme with input curve fitting provides accurate rCMRGlc quantification.
• This technique simplifies the input curve determination process in PET studies of brain metabolism.
• The method enhances the feasibility of studying brain glucose metabolism with less invasive procedures.