A comparison of cardiac (31)P MRS at 1.5 and 3 T.

(31)P MRS was evaluated on normal volunteers at 1.5 and 3 T, and the signal-to-noise ratio (SNR) of the two field strengths was calculated. The in vivo spin-lattice, T(1), relaxation times for PCr and gamma-ATP, which are essential for correcting for the effects of radiofrequency saturation on the PCr/ATP ratio, were determined at 3 T. The T(1) values for six volunteers were 3.8 +/- 0.7 s for PCr (mean +/- SD) and 2.4 +/- 1.1 s for gamma-ATP, which are similar to reported values at 1.5 T, allowing us to use protocols developed at 1.5 T at the new clinical field strength of 3 T. Direct comparison between 1.5 T and 3 T in the same 10 subjects, using coils of identical geometry and identical pulse sequences gave a mean SNR for PCr at 3 T which was 206 +/- 94% of that at 1.5 T. The linewidth for PCr increased from 13 +/- 6 Hz at 1.5 T to 22 +/- 12 Hz at 3 T. The coefficient of variation in the measurement of PCr/ATP, based on the Cramer-Rao lower bounds, was reduced from 32 +/- 25% at 1.5 T to 18 +/- 13% at 3 T. Thus, (31)P MRS at 3 T is greatly improved by the increase in SNR compared with acquisitions at 1.5 T because of the higher field strength.