Spectrally-selective measurements of reversible and irreversible transverse relaxation rates from single spin-echo PRESS acquisitions in muscle.

The goal of this study was to test a new formalism for extracting reversible and irreversible transverse relaxation rates from resonances within typical proton muscle spectra using only a single spin echo as acquired with routine single-voxel, point-resolved echo spectroscopy (PRESS) acquisitions. Single-voxel, non-water-suppressed PRESS acquisitions within the calf muscles of four healthy subjects were performed at 1.5 T using six echo times ranging from 30 to 576 ms. Novel transverse relaxation analyses of water, choline, creatine, and lipid resonances were performed based upon the disparate relaxation sensitivities of the left versus the right sides of spectroscopically sampled spin echoes. Irreversible and reversible transverse relaxation rates R-2 and R-2 ' were extracted for water, metabolites, and lipids using echo times of 288 ms and longer. The R-2 values so obtained were compared with more conventional "gold standard" Hahn values, R-2Hahn, evaluated from the echo-time dependence of spectral peak areas generated from right-side sampling alone. Water resonances displayed biexponential Hahn signal decays, consistent with observations of decreasing R-2 values with increasing echo time via the new approach. Choline and creatine resonances displayed monoexponential echo-time decays, with R-2Hahn values in reasonable agreement with R-2 values obtained from the single-echo analyses at the longer echo times. Lipid methylene and methyl R-2 values extracted from the new approach were also in reasonable accord with R-2Hahn values. Further validation of the technique was provided through PRESS acquisitions on a water phantom to which various levels of gadolinium were added in order to manipulate transverse relaxation rates, yielding excellent agreement between water-resonance R-2Hahn and single-echo R-2 values. In summary, this work demonstrates the feasibility of measuring reversible and irreversible transverse relaxation rates for individual spectral peaks from single-echo PRESS acquisitions, enabling a reduction in overall scan time relative to the use of multiple acquisitions with varying echo time.