Cardiac PET/CT with Rb-82: optimization of image acquisition and reconstruction parameters.

BACKGROUND

Our aim was to characterize the influence of time-of-flight (TOF) and point spread function (PSF) recovery corrections, as well as ordered subset expectation maximization (OSEM) reconstruction parameters, in Rb PET/CT quantification of myocardial blood flow (MBF) and myocardial flow reserve (MFR). Rest and stress list-mode dynamic Rb PET acquisition data from 10 patients without myocardial flow defects and 10 patients with myocardial blood flow defects were reconstructed retrospectively. OSEM reconstructions were performed with Gaussian filters of 4, 6, and 8 mm, different iterations, and subset numbers (2 × 24; 2 × 16; 3 × 16; 4 × 16). Rest and stress global, regional, and segmental MBF and MFR were computed from time activity curves with FlowQuant software. Left ventricular segmentation using the 17-segment American Heart Association model was obtained.

RESULTS

Whole left ventricle (LV) MBF at rest and stress were 0.97 ± 0.30 and 2.30 ± 1.00 mL/min/g, respectively, and MFR was 2.40 ± 1.13. Concordance was excellent and all reconstruction parameters had no significant impact on MBF, except for the exclusion of TOF which led to significantly decreased concordance in rest and stress MBF in patients with or without perfusion defects on a coronary artery basis and in MFR in patients with perfusion defects.

CONCLUSIONS

Changes in reconstruction parameters in perfusion Rb PET/CT studies influence quantitative MBF analysis. The inclusion of TOF information in the tomographic reconstructions had significant impact in MBF quantification.