Accuracy of shortened acquisition protocols for the quantitative assessment of the 2-tissue compartment model using dynamic PET F-18-deoxyglucose (FDG) studies

Objectives The assessment of the FDG kinetics demands the use of a 2-tissue compartment model. The classical approach is based on a dynamic data acquisition for 60 minutes following the FDG injection. We evaluated the accuracy from shortened acquisition protocols, which are more suitable for routine applications.

Methods Dynamic PET images were acquired for 60 minutes using a 28 frames protocol over the region of the primary tumor in 421 patients and 1131 VOIs were placed over the malignant lesions and reference tissues. We applied a 2-tissue compartment model to the VOI data of the dynamic studies. The results obtained from the 60 minute data acquisitions were used for reference. Then data sets were shortened to a 10-minute and a 20-minute dynamic study, including also the 60 minute acquisition. The 2-tissue compartment parameters were calculated also for these data sets.

Results The correlation analysis revealed for vb a correlation of r=0.989 for both, the 10-minute and 20-minute data acquisition. For K1 we noted a correlation of r=0.998 and r=0.992 for the 20-minute and 10-minute data acquisition. The correlation coefficients were lower for k3 with r=0.927 and r=0.888 for the 20-minute and 10-minute data sets

Conclusions Quantitative PET FDG studies require a dynamic data acquisition usually for 60 minutes. However, for routine purpose the dynamic data acquisition can be limited to a 10-minute dynamic series and combined with a whole body acquisition 60 minutes after tracer application. The results obtained with 2-tissue compartment fitting are acceptable e.g. for therapy follow up studies.