Methodological aspects of quantitative cardiac molecular imaging

Sammanfattning: The objective of this research was to facilitate the use of quantitative cardiac molecular imaging by developing and validating methods and applications. More specifically:we determined the optimal tracer kinetic model for analysis of 11C-PIB and evaluated the performance of two simpler measures, retention index (RI) and standardized uptake value (SUV), in the quantification of cardiac 11C-PIB uptake in amyloidosis. An irreversible two-tissue (2Tirr) model best described the 11C-PIB uptake in cardiac amyloidosis. RI and SUV showed high correlation with quantitative results from this kinetic model and also a better discrimination between amyloidosis patients and controls than a 2Tirr model with population averaged metabolite correction. RI and SUV are furthermore more feasible for use in clinical routine and therefore the preferred measure to use in PET diagnosis of cardiac amyloidosis. We also tested the feasibility of a semiautomatic software to analyze RI and visualize cardiac uptake of 11C-PIB in amyloidosis. The RI values were comparable with RI based on manual segmentation, showing significantly higher 11C-PIB RI in amyloidosis patients than in healthy volunteers. A fast and accurate semiautomatic analysis process is thus feasible to use for PET in cardiac amyloidosis instead of the laborious manual analyses that were used so far.Furthermore, we assessed the quantitative accuracy of cardiac perfusion measurements with 15O-water PET in a digital time-of-flight PET-MR scanner. A high correlation and agreement between PET-MR based and PET-CT based MBF was found; cardiac perfusion measurements with 15O-water can therefore be performed accurately with the fully integrated Signa PET-MR scanner. Finally, we assessed the quantitative accuracy of cardiac perfusion measurements using dynamic contrast-enhanced MRI with simultaneous 15O-water PET as reference at rest and during adenosine-induced hyperemia with a fully integrated PET-MR scanner. The correlations between global and regional MRI- and PET-based MBF values were good and the biases were negliable for both global and regional MBF comparisons, but the limits of agreement were wide for both global and regional MBF, with larger variability for high MBF-values indicating that MRI-based quantitative MBF measurement based on widely available acquisition protocols is not yet ready for clinical introduction.