Purpose

Accurate quantification of proton spectra acquired at relatively long echo-time (T_E) requires the knowledge of T₂ relaxation time constants in order to correct for signal losses. Several studies, both in humans¹ and animals², have shown that the apparent T₂ rates of singlet resonances such as tCr, tNAA and tCho are increased under the effect of a Carr-Purcell (CP) pulse train. In addition, we previously showed that the increase in the apparent T₂ for strongly J-coupled metabolites (glutamate, glutamine, myo-inositol and taurine) is much larger (at least 2 times) compared to singlets under T_2p regime². However, so far there have not been any studies comparing the T₂ values of metabolites between LASER and PRESS, sequences commonly used in the research settings. Based on findings from previous studies, we postulate that the apparent T₂ of metabolites measured with LASER will be higher for singlets and J-coupled metabolites compared to PRESS since LASER consists of 3 pairs of refocusing pulses which act as a CP pulse train. The aim of this study was to measure the T₂ of metabolites using LASER and PRESS in the human brain at 3 T.

Methods

5 healthy subjects (21 ± 1 years) were scanned on a Siemens 3 T scanner after giving informed consent approved by the IRB. Body coil was used for excitation while the 32-channel receive-only head-coil was used for signal reception. A VOI of 25×25×25 mm³ was positioned in the prefrontal cortex using T₁-weighted MPRAGE images. B₀ shimming was achieved using system 3D-GRE shim (Brain shim mode). Localized spectra were acquired using both PRESS and LASER sequences. VAPOR scheme interleaved with OVS pulses was used in both sequences. Data were acquired at six different T_Es (35, 140, 230, 290, 330, 400 ms in LASER and 23, 60, 140, 210, 270 and 400 ms in PRESS) to measure the T₂ relaxation of metabolites. Water reference scans were also acquired for eddy-current correction. All spectra were processed in Matlab³: eddy current effect was first corrected followed by single-shot frequency and phase corrections. Spectra were analyzed with LCModel using simulated basis sets and measured macromolecule spectra. T₂ values were obtained by fitting the amplitudes obtained from LCModel with an exponential fit.

Results and Discussion

Proton spectra acquired from one subject at different T_Es with PRESS and LASER sequences are shown in Figure 1. There is a noticeable difference in the peak heights of tNAA, tCr and tCho, mostly visible at long echo times suggesting that the apparent T₂s of these metabolites are different between both sequences. Indeed, the T₂ relaxation times of these non-coupled metabolites were found to be statistically longer with LASER compared to PRESS (Table 1). A previous study reported an increase of about ~75% for tNAA and tCr with CP-LASER compared to PRESS at 4 T¹. However in the current study, the increase was merely ~25% on average with LASER, suggesting that the apparent T₂ will be further lengthened when moving from LASER to CP-LASER as demonstrated in the rat brain². The increase in apparent T₂ under LASER could be explained by the fact that the CP train suppresses the diffusion component efficiently, since these moieties do not have exchangeable protons. The T₂ values for Glu and Ins were found to be comparable between LASER and PRESS (Table 1). These values agree with previously published values at 3 T using PRESS3-5 where T₂ of these metabolites ranged from 180 to 200 ms. Contrary to our hypothesis, no lengthening of the apparent T₂ for strongly coupled spins were observed with LASER in this study. One possible explanation is that the duration of the CP pulse train in LASER is too short such that there is no refocusing of cross-correlation effects between different dipole–dipole interactions compared to when using CP-LASER and T_2p-LASER sequences².

Conclusion

In summary, this study shows that the apparent T₂ relaxation times of singlets are lengthened by ~25% under LASER compared to PRESS while similar apparent T₂ values were observed for strongly J-coupled metabolites with both sequences. In conclusion, the LASER sequence seems to be more efficient in suppressing the diffusion component of non-coupled singlets (having non-exchangeable protons) compared to coupled metabolites.

Acknowledgements

Supported by NIH grants: R21AG045606, P41 EB015894, P30 NS076408