Time-resolved angiography with stochastic trajectories for dynamic contrast-enhanced MRI in head and neck cancer: Are pharmacokinetic parameters affected?

Purpose: To investigate the effects of different time-resolved angiography with stochastic trajectories (TWIST) k-space undersampling schemes on calculated pharmacokinetic dynamic contrast-enhanced (DCE) vascular parameters. Methods: A digital perfusion phantom was employed to simulate effects of TWIST on characteristics of signal changes in DCE. Furthermore, DCE-MRI was acquired without undersampling in a group of patients with head and neck squamous cell carcinoma and used to simulate a range of TWIST schemes. Errors were calculated as differences between reference and TWIST-simulated DCE parameters. Parametrical error maps were used to display the averaged results from all tumors. Results: For a relatively wide range of undersampling schemes, errors in pharmacokinetic parameters due to TWIST were under 10% for the volume transfer constant, K trans, and total extracellular extravascular space volume, Ve . TWIST induced errors in the total blood plasma volume, Vp , were the largest observed, and these were inversely dependent on the area of the fully sampled k-space. The magnitudes of errors were not correlated with K trans, Vp and weakly correlated with Ve . Conclusions: The authors demonstrated methods to validate and optimize k-space view-sharing techniques for pharmacokinetic DCE studies using a range of clinically relevant spatial and temporal patient derived data. The authors found a range of undersampling patterns for which the TWIST sequence can be reliably used in pharmacokinetic DCE-MRI. The parameter maps created in the study can help to make a decision between temporal and spatial resolution demands and the quality of enhancement curve characterization.