Jørgen Frøkiær

Objectives: Dynamic PET with 11C-acetate can be used to quantify myocardial blood flow and oxidative metabolism, the latter of which is used to calculate myocardial external efficiency (MEE). Calculation of MEE requires forward stroke volume (FSV) data. FSV is affected by cardiac loading conditions, potentially introducing bias if measured with a separate modality. The aim of this study was to develop and validate methods for automatically extracting FSV directly from the dynamic PET used for measuring oxidative metabolism. Methods: 16 subjects underwent a dynamic 27 min PET scan on a Siemens Biograph TruePoint 64 PET/CT scanner after bolus injection of 399±27 MBq of 11C-acetate. The LV-aortic time-activity curve (TAC) was extracted automatically from dynamic PET data using cluster analysis. The first-pass peak was derived by automatic extrapolation of the down-slope of the TAC. FSV was then calculated as the injected dose divided by the product of heart rate and the area under the curve of the first-pass peak. Gold standard FSV was measured in the left ventricular outflow tract by cardiovascular magnetic resonance using phase-contrast velocity mapping within two weeks of PET imaging. Results: Cluster analysis could be performed successfully in all patients. Heart rate varied by 5±4 beats/min between scans. FSV-PET was highly correlated with FSV-MRI (PET=0.92*MRI+18 mL, r = 0.88, p < 0.001) although obtained values were significantly higher (bias of 14±13%, 95% confidence interval -11:40%, p = 0.001). Conclusions: FSV can be obtained automatically and reliably using dynamic 11C-acetate PET/CT and cluster analysis, although a small overestimation is observed when compared to FSV determined from MRI. This method could potentially be generalized to other tracers, although this requires further validation.