INTRODUCTION

Functional information about water and lipid in tissues, e.g., diffusion, perfusion, relaxation time, and lipid fraction (LF) obtained with magnetic resonance imaging (MRI), is useful to assess the physiological conditions, physical properties, and tissue metabolism.^1-3 However, it is difficult to obtain all this information at the same time.^4,5 Therefore, to simultaneously acquire these functional information, we developed a novel method using chemical shift displacement and recovery-based separation of lipid tissue (SPLIT) with different inversion time (TI), echo time (TE), and b-values. Additionally, diffusion, perfusion, T₁, T₂, and LF in the calf were evaluated.

METHODS

On a 3.0 T MRI, single-shot diffusion echo-planar imaging (SSD-EPI) without fat suppression was used, and the scan parameters were optimized to separate water and lipid images using the chemical shift displacement of lipid signals in the phase-encoding direction, ie., echo spacing, 524 μs; receiver bandwidth, 250 kHz; field of view, 430 mm; TE, 87.7 ms; and b-values, 0 to 3000 s/mm² (7 points). Moreover, an inversion pulse (TI of 292 ms) was applied to SSD-EPI to remove olefinic signals, which overlaps water signals. Consecutively, SSD-EPI with a b-value of 0 s/mm² was performed with TI of 0 ms (ie., without inversion pulse) and TE of 31.8 ms for T₁ and T₂ measurements, respectively. Under these conditions, transverse water and lipid images at the maximum diameter of the right calf were obtained in six healthy subjects. The total acquisition time was 3 min 36 s. Then, T₁ and T₂ values were calculated from different TI or TE images, and the apparent diffusion coefficient (ADC) was calculated from multiple b-values with a monoexponential fitting for the tibialis anterior muscle (TA), tibialis bone marrow (TB), and subcutaneous fat (SF). Furthermore, perfusion-related and restricted diffusion coefficients (D_p and D_r) with a biexponential fitting were calculated for the TA. The diffusion parameters of the TA were determined at b-values up to 1000 s/mm² because the signal intensities reached the noise level at a b-value of 3000 s/mm². Finally, the LF of the TA was determined after T₁ and T₂ corrections.

RESULTS AND DISCUSSION

Figure 1 depicts examples of water and lipid diffusion-weighted images using the proposed SPLIT imaging with and without inversion pulse. The SPLIT method allowed sufficient separation of water and lipid images of the calf, and the SPLIT with inversion pulse effectively suppressed olefinic signals. Representative diffusion parameters, T₁, and T₂ maps for one subject are shown in Figure 2. In addition, the mean ADC, T₁, and T₂ in the TA, TB, and SF are presented in Table 1. The ADCs of the TB and SF (lipid diffusion) were 0.02 ± 0.02 × 10^-3 mm²/s and 0.04 ± 0.02 × 10^-3 mm²/s, respectively, which were much lower than that of the TA (water diffusion, 1.67 ± 0.08 × 10^-3 mm²/s). Moreover, the D_p, D_r, and LF values of the TA were 20.1 ± 10.6 × 10^-3 mm²/s, 1.60 ± 0.07 × 10^-3 mm²/s, and 3.3 ± 2.2%, respectively. All these values were in general agreement with previously reported data obtained from water and lipid images scanned separately,^6-9 supporting the validity of our method.

CONCLUSION

With our proposed method, it is possible to simultaneously obtain quantitative values concerning diffusion, perfusion, relaxation time, and LF, thereby increasing the amount of functional information about water and lipid in tissues.

Acknowledgements

No acknowledgement found.