Kinetic models for analysing myocardial [11C]palmitate data

[C-11]Palmitate PET can be used to study myocardial fatty acid metabolism in vivo. Several models have been applied to describe and quantify its kinetics, but to date no systematic analysis has been performed to define the most suitable model. In this study a total of 21 plasma input models comprising one to three compartments and up to six free rate constants were compared using statistical analysis of clinical data and simulations. To this end, 14 healthy volunteers were scanned using [C-11]palmitate, whilst myocardial blood flow was measured using H-2 O-15. Models including an oxidative pathway, representing production of (CO2)-C-11, provided significantly better fits to the data than other models. Model robustness was increased by fixing efflux of (CO2)-C-11 to the oxidation rate. Simulations showed that a three-tissue compartment model describing oxidation and esterification was feasible when no more than three free rate constants were included. Although further studies in patients are required to substantiate this choice, based on the accuracy of data description, the number of free parameters and generality, the three-tissue model with three free rate constants was the model of choice for describing [C-11]palmitate kinetics in terms of oxidation and fatty acid accumulation in the cell.