Mapping of sodium and water content ratio in vivo in the human calf

Anja M. Marschar¹, Moritz C. Berger¹, Mark E. Ladd¹, Peter Bachert¹, and Armin M. Nagel^1,2

¹Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany, ²Diagnostic and Interventional Radiology, University Medical Center Ulm, Ulm, Germany

Synopsis

Both lower legs of three male healthy volunteers were examined at 3 T and 7 T by means of ¹H and ²³Na MRI to obtain quantitative maps of the sodium and water content in muscle tissue. Corrections for T₂^* and T₁ relaxation and RF profiles (B₁⁺, B₁^–) were performed. The resulting water content value in calf muscle tissue was in the range of 75–85 mol/l, the level of sodium was 15–21 mmol/l (sodium-to-water content ratio: 0.20–0.27 ‰).

Purpose

To quantify the sodium and water content in human calf muscle tissue in vivo and to derive sodium–to–water content ratio maps.

Methods

Both lower legs of three male healthy volunteers (23–27 years) were examined by means of ¹H and ²³Na MRI (¹H: 3–T MR scanner; Biograph mMR; Siemens, Erlangen, Germany; ²³Na: 7–T whole–body MR scanner; Magnetom 7T; Siemens, Erlangen, Germany). At 3 T spine and body matrix coils were used for signal reception and the body coil for excitation, while a ¹H/²³Na quadrature Tx/Rx birdcage coil (Rapid Biomed, Rimpar, Germany) was employed in studies at 7 T. Sequence parameters are listed in Tab. 1.

For processing, all datasets were interpolated to an isotropic resolution of (1.5 mm)³. All 3 T and 7 T images were co–registered to a ¹H 3D gradient echo (GRE) dataset (Tab. 1A), which was obtained at 7 T for registration purposes. Identical positioning of the calves was ensured by a half–open formed plaster. In addition, MR–visible vitamin–E pills (Tocopherol from vegetable oil) were positioned proximal and distal on the skin as slice–positioning landmarks. See Figs. 2 and 3 for details of the reference tubes.

Proton–density maps were derived from data obtained with 2D multi–echo GRE acquisition (Tab. 1B) by extrapolating the signal intensities towards TE = 0 ms by a 3^rd–order polynomial to correct for T₂^* decay (cf. Neeb et al.¹).

To correct for T₁ weighting, T₁ maps were calculated by pixelwise regression of 2D inversion-recovery single–shot Turbo FLASH data (Tab. 1C). A delay of 10 s was waited after each TI measurement.

The transmission field B₁⁺ was corrected using a flip angle map acquired by a 3D phase–sensitive technique² (Tab. 1D). B₁⁺ values of fat (Fig.1) were excluded by a fat mask. A 2D 4^th–order polynomial was fitted to the muscle data to achieve a smooth B₁⁺ map. The previously neglected subcutaneous fat was taken into account by extrapolation.

The reception field B₁^– of the body coil (Fig. 1) was measured with a low–resolution fat–saturated 2D GRE sequence (Tab. 1E). The method of Wang et al.³ for minimizing tissue contrast was modified to include only voxels with T₁ values in the range of ± 200 ms around the most frequent T₁ value of muscle tissue. This guarantees a T₁ bias free determination of B₁^–. The processing steps for B₁^– determination are illustrated in Fig. 1. Analogous to the B₁⁺ assessment, the resulting B₁^– distribution was fitted and extrapolated to subcutaneous fat.

The receive sensitivity profile of the spine and body matrix coils was obtained by dividing two 3D GRE datasets (Tab. 1F). In the first dataset signal reception was performed with the spine and body matrix coils and in the second dataset with the body coil alone.

Fat and water maps were derived by means of a 3D GRE two-point Dixon⁴ technique (Tab. 1G).

The tissue sodium concentration was measured at 7 T with a 3D radial density–adapted sequence⁵ (Tab. 1H). Effects of B₀ and B₁ inhomogeneities were corrected (Tab. 1 I,J).

Results

Figure 2 displays the maps that contribute to the final water content values in the calf muscles (only the central slices are shown). Figure 3 shows maps of sodium and water content and of their ratio. Table 2 lists values of sodium and water concentration for two different muscle groups (n = 3 volunteers). Evaluated ROIs without vessels are delineated in Fig. 3.

The resulting water content value in calf muscle tissue was in the range of 75–85 mol/l, the level of sodium was 15–21 mmol/l and the sodium–to–water content ratio yields 0.20–0.27 ‰.

Discussion & Conclusion

The quantification of the water content of muscle tissue by MRI has so far not been studied in detail. We obtained 75–85 mol/l (68–75 %) water content, which is slightly lower than the literature values of 77.3–81.7 % obtained with muscle biopsy (volume ratio derived from mass ratio 73.7–77.8 % (paper) with assumed muscle density of 1.05 kg/l) (Graham et al.⁶).

The artifact in the water content map (Figs. 2, 3) at the boundaries between muscle and fat tissue is attributed to an instable 2D polynomial fit. Figure 1 reveals a pseudo–mirror–symmetric behavior of B₁⁺ and B₁^–. Accordingly, the simplification B₁⁺ = B₁^– for small magnetic fields can no longer be assumed to be true at 3 T. This requires the consideration and measurement of B₁^– in quantitative MRI.

It remains to be shown in future studies whether the ratio of sodium and water content can provide additional relevant physiological information, for example in the case of muscular channelopathies⁷.

Acknowledgements

This work was funded by the German Research Foundation (DFG). Grant number: NA736/2-1.

References

1. Neeb H, Zilles K, Shah NJ, A new method for fast quantitative mapping of absolute water content in vivo, NeuroImage, 2006; 31(3): 1156–1168

2. Morrell GR, A phase–sensitive method of flip angle mapping, Magn. Reson. Med., 2008; 60(4): 889–894

3. Wang J, Qiu W, Yang QX, Smith MB, Constable RT, measurement and correction of transmitter and receiver induced nonuniformities in vivo, Magn. Reson. Med., 2005; 53(2): 408–417

4. Dixon WT, Simple proton spectroscopic imaging, Radiology, 1984; 153: 189–194

5. Nagel AM, Laun FB, Weber MA, Matthies C, Semmler W, Schad LR, Sodium MRI using a density–adapted 3D radial acquisition technique, Magn. Reson. Med., 2009; 62(6): 1565–1573

6. Graham JA, Lamb JF, Linton AL, Measurement of body water and intracellular electrolytes by means of muscle biopsy, The Lancet, 1967; 7527(290): 1172–1176

7. Lehmann–Horn F, Jurkat–Rott K, Voltage–gated ion channels and hereditary disease, Physiol. Rev., 1999; 79(4): 1317–72

Figures

Table 1: List of employed measurement parameters. TR = repetition time; TE = echo time; FA = flip angle; TA = acquisition time. All ¹H 3D images were acquired in coronal orientation, field–of–view (FOV) 288 mm × 288 mm. 2D sequences were acquired in transversal orientation, FOV 288 mm × 144 mm.

Figure 1: B₁^– map calculation workflow for the body coil. For time–efficient measurement, the input images are of low spatial resolution because B₁⁺ and B₁^– are considered smooth.

Figure 2: Calculation of the final water content map from measurements of both human calves. Three tubes: (a) water with Magnevist (0.6 ml per 1 liter H₂O), (b) pure water, and (c) silicon oil (Elbesil B20). (a) served as a reference for 100 % (110 mol/l) H₂O in the map.

Figure 3: Sodium and water concentration maps are shown with regions of interest used for Tab. 2. Three reference tubes containing 20, 30, and 40 mmol/l NaCl and 4% agarose gel were placed close to the calves at 7 T. On the right, the ratio of both quantities is shown.

Table 2: Mean values of sodium and water concentration in four regions of interest and their corresponding ratio are listed for volunteers V1–V3, including (mean ± SD) values. The regions of interest are delineated in Fig. 3.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)

3945