Quantification of myocardial blood flow using dynamic myocardial CT perfusion compared with 82Rb PET.

ABSTRACT

PURPOSE: Absolute measures of myocardial blood flow (MBF) obtained with dynamic myocardial CT perfusion (DM-CTP) are underestimated when compared with reference standards. This is to some extent explained by incomplete extraction of iodinated contrast agent (iCA) to the myocardial tissue. We aimed to establish an extraction function for iCA, use the function to calculate MBFCT and to compare this with MBF measured with 82Rb positron emission tomography (PET). MATERIALS AND METHODS: Healthy individuals without coronary artery disease (CAD) were examined with 82Rb PET and DM-CTP. The factors a and ß of the generalized Renkin-Crone model were estimated using a non-linear least squares model. The factors providing the best fit for the data were subsequently used to calculate MBFCT. RESULTS: Of consecutive 91 individuals examined, 79 were eligible for analysis. The factors a and ß providing the best fit of the nonlinear least-squares model to the data were a â€‹= â€‹0.614 and ߠ​= â€‹0.218 (R-squared â€‹= â€‹0.81). Conversion of the CT inflow parameter (K1) values using the derived extraction function resulted in a significant correlation between MBF measured during stress using CT and PET (P â€‹= â€‹0.039). CONCLUSION: In healthy individuals, flow estimates obtained with dynamic myocardial CT perfusion during stress were, after conversion to MBF using the extraction of iodinated CT contrast agent, correlated with absolute MBF quantified with 82Rb PET.