Comparison Between Respiratory-Triggered and Free-Breathing Signal Acquisition Methods for Single-Voxel Phosphorus...

The optimal protocol for clinical liver ³¹P-magnetic resonance spectroscopy (MRS) remains unclear. Single-voxel ³¹P-MRS using image-selected in vivo spectroscopy (ISIS) employs respiratory-triggering (RT) or free-breathing (FB) acquisition. RT provides robust data but prolongs scan duration; FB allows faster acquisition but may suffer from low signal-to-noise ratio (SNR). Yet, direct comparison using a regulatory-approved coil in a clinical setting has not been reported.

To compare ³¹P-MRS data stability and robustness between RT and FB.

Prospective.

24 volunteers (19 male/5 female).

3 T MRI; single-voxel ³¹P-MRS using ISIS (free induction decay-based) with approved ³¹P coil.

³¹P-MRS was performed using RT and FB techniques (128 and 192 signal averages, respectively; expected scan duration ~13 min each). Spectra were analyzed using jMRUI. SNR and peak areas for PME, Pi, PDE, α-ATP, and NADPH, normalized using γ-ATP, were compared. PME/PDE and NADPH/(PME + PDE) ratios were also compared.

Paired t-tests and Bland-Altman analysis were used. A p value < 0.05 was considered significant.

Scan duration was significantly longer for RT (15 min 44 s) than FB (12 min 56 s). No significant differences were observed for SNR (p = 0.570), NADPH/(PME + PDE) (p = 0.931), PME/γ-ATP (p = 0.556), Pi/γ-ATP (p = 0.931), α-ATP/γ-ATP (p = 0.332), or NADPH/γ-ATP (p = 0.394). Significant differences were noted for PDE/γ-ATP (RT 1.68 vs. FB 1.35, p = 0.003) and PME/PDE (RT 0.434 vs. FB 0.489, p = 0.046). Bland-Altman analysis showed near-zero fixed biases and no proportional bias, with limits of agreement from -0.53 to 0.62 (PME), -0.30 to 0.30 (Pi), -0.41 to 1.07 (PDE), -0.45 to 0.43 (α-ATP), and -0.80 to 0.90 (NADPH).

³¹P-MRS of the liver showed equivalent stability and robustness for RT and FB. FB yielded comparable data within a shorter, predictable scan duration.

2.

1.