On the accuracy and precision of cardiac magnetic resonance T2 mapping: A high-resolution radial study using adiabatic T2 preparation at 3 T.

The goal of this study was to characterize the accuracy and precision of cardiac T2 mapping as a function of different factors including low signal-to-noise ratio (SNR), imaging in systole, and off-resonance frequencies. Bloch equation and Monte Carlo simulations were used to determine the influence of SNR and the choice of T2 preparation echo time (TET2prep ) increments on the accuracy and precision of high-resolution radial cardiac T2 mapping at 3.0 T. Healthy volunteers were scanned to establish the difference in precision and inter- and intraobserver variability between T2 mapping in diastole and systole, as well as the effect of SNR and off-resonance frequencies on the accuracy of T2 maps. The simulations demonstrated that a TET2prep increment of ∼0.75 times the T2 value of interest optimally increases the precision of the T2 fit. Systolic T2 maps were found to have a higher precision (P = 0.002), but similar inter- and intraobserver variability compared with diastolic T2 maps, whereas off-resonance frequencies beyond ± 100 Hz cause a significant decrease in both accuracy and precision (P < 0.05). This evaluation of the accuracy and precision of cardiac T2 mapping characterizes the major vulnerabilities of the technique and will help guide protocol definition of studies that include T2 mapping. Magn Reson Med 77:159-169, 2017. © 2016 Wiley Periodicals, Inc