Elsevier

Magnetic Resonance Imaging

Volume 22, Issue 2, February 2004, Pages 171-180
Magnetic Resonance Imaging

Original contribution
Sodium 3-D MRI of the human torso using a volume coil

https://doi.org/10.1016/j.mri.2003.08.007Get rights and content

Abstract

Sodium MR imaging is considered to provide clinically important information about the human body that is not achievable by hydrogen-based approaches. However, due to the low natural abundance in biological tissues, sodium signals usually lead to low spatial resolution, low SNR, and long acquisition times compared to conventional 1H imaging, even using well-adapted surface coils. For our study, a volume coil was designed with nearly homogeneous excitation/receive characteristics and a suitable geometry fitting the human torso. A sufficient penetration throughout the entire thorax, abdomen, or pelvis is provided allowing for sodium imaging of the kidneys, the liver with gall bladder, or the myocardium. All measurements were performed on a 1.5 T whole body scanner using a spoiled 3-D gradient echo sequence. Imaging parameters TE, TR, and readout bandwidth were optimized for sensitive recording of the sodium component with slow transverse relaxation. Nonselective RF excitation pulses with a duration of 2.5 ms and rectangular shape were applied to avoid SAR problems. Narrow receiver bandwidth and excitation near the Ernst angle provided clinically practicable examinations with measuring times of less than 15 min at a spatial resolution of 8 × 8 × 8 mm3. Under these conditions, SNR of 11 for the kidneys and vertebral disks, 9 for the spinal canal, and 6 for the liver was achieved. A special 3-D spin echo sequence was used to determine T2, times which resulted to 15.3 ± 1.1 ms for liver, 27.7 ± 7.2 ms for kidneys, and 24.0 ± 4.7 ms for the content of the spinal canal.

Introduction

Sodium MR imaging and spectroscopy in humans and animals have been performed for more than 15 years [1], [2], but it has not yet found a fixed place in clinical routine. The main reason is relatively long measuring time at only moderate spatial resolution. In addition, the usually applied surface coils did not allow examinations of thoracic or abdominal organs with sufficient homogeneity and sensitivity. High RF energy depositions above the legal SAR limits might have further compromised the application of fast imaging techniques for sodium in vivo.

The motivation for sodium MR imaging results from the fact, that the Na-K-pumps in impaired cells do not work sufficiently and sodium concentrations and/or volume shares of tissue compartments inside and outside the cells change. Since sodium concentration has no influence on proton relaxation, this effect does not manifest in proton images and can only be visualized by sodium NMR.

In comparison to 1H (γH = 42.57 MHz/T, I = 1/2) in 23Na (γNa = 11.26 MHz/T, I = 3/2) experiments MR sensitivity S = γ3 I(I + 1) is about 1 order of magnitude lower due to the lower magnetogyric ratio. Additionally, for in vivo examinations the relatively low spin-density of 23Na compared to 1H has to be taken into account, resulting in an absolute sensitivity more than 4 orders of magnitude lower than for 1H in the mean over all tissues.

The 23Na nucleus has spin 3/2 and possesses an electric quadrupole moment. Relaxation mechanisms are dominated by the interaction of this quadrupole moment with electrical field gradients generated by the surrounding. For this reason, relaxation times T1 and T2 of 23Na are clearly shorter than those of 1H. Due to the short T1 value of approx. 30–50 ms under in vivo conditions [3], [4], potential number of excitations with large flip angles per unit time is relatively high for 23Na and lower absolute sensitivity of 23Na can be partially compensated by shorter repetition times. Unfortunately, application of standard RF pulses would lead to SAR problems under in vivo conditions on whole-body units.

In living tissues, a biexponential decay of transverse magnetization is observed with a slow component T2slow of approx. 20 ms and a fast component with T2fast of ∼2 ms [3]. Different quadrupole interactions in several tissue compartments (cellular space, interstice, and blood) are reported as main reasons. The amplitude ratio between the fast and slow component varies depending on the type of tissue under investigation. Regions with a high amount of intracellular space compared to the interstitial space are dominated by the signal with the fast decay. This can be explained by longer interaction times of the molecules inside the cell due to the slower thermal motion and higher mean molecular weight. Sequences with high sensitivity to the fast intracellular component require short echo time TE, high receiver bandwidth, and fast gradient switching. On the other hand, a high receiver bandwidth leads to noise problems with reduced general sensitivity to sodium signals with slower transverse relaxation.

Inside the cell, the thermal motion of molecules, in part even, is slow enough that quadrupole splitting of the 4 Zeeman terms of the spin I = 3/2 nucleus can be observed. Multiple quantum filter technique can be applied to this fraction of sodium and strong weighting for the intracellular space can be achieved [5], [6], [7]. The feasibility of such measurements has been shown on animals. An application on humans in clinical routine seems to be problematic due to a very long measuring time, even using high B0-field strength.

Most work on 23Na MR imaging in humans has been focused on the assessment of the total sodium content in tissue [3], [8], [9], [10], [11]. The signal difference between infarcted and viable tissue regions has been examined. Much effort was given to design sequences with very short echo times in order to obtain as much signal as possible from the fast decaying component. Radial back projection imaging has been applied because no time consuming phase encoding gradients are necessary for this approach [1].

Complete separation of intra- and extracellular sodium components based on relaxation characteristics seems critical, since prolonged transverse relaxation could be present in some intracellular subcompartments, especially in impaired cells. Special shift reagents only distributing inside the extracellular space can provide distinguishable signal contributions of intra- and extracellular compartments [12], [13], [14], [15], but only animal studies have been performed so far, since none of those shift reagents has reached admission for human studies.

Recent studies applying sodium imaging on humans have been dedicated to examine pathologies concerning the dysfunction of the Na-K-pump in brain or heart and have been performed with surface coils [3], [9], [10], [11]. In contrast, the coil presented here is an adapted volume coil allowing examinations of the human torso for determination of the sodium content of deeper lying tissues as liver and gallbladder, kidneys, and the posterior wall of the myocardium. Requirements for the coil design were homogeneous RF excitation and receive characteristics and a suitable geometrical fit to the human torso. A nearly homogeneous B1-field distribution over the entire abdomen or thorax is very advantageous for carrying out spin-echo studies on relaxivity of sodium signals in the mentioned organs.

A volume coil provides a clearly larger sensitive volume, but leads to a reduced SNR in comparison to smaller surface coils. For 23Na imaging of deeper lying tissues, noise contributions from the whole enlarged volume has to be accounted for.

In our studies, imaging parameters were optimized for measuring the slow component of sodium. An optimization for assessment of the fast sodium component was not really promising, since for a correct spatial encoding of this component a readout bandwidth of at least 500 Hz per pixel would be necessary leading to insufficient SNR by using our large volume coil.

Pabst et al. [10] and Sandstede et al. [11] have shown for myocardial infarctions of humans that pathological changes can be detected by the slow component. Further clinically relevant applications of imaging the slow sodium component could be the assessment of sodium content in the kidneys in hydrated and dehydrated state, or monitoring of cytotoxic therapy of tumors.

The aim of the presented work was to test the feasibility of sodium imaging of the human torso with a large volume coil. Sequence parameters were optimized, achievable spatial resolution and suitable measuring times were analyzed, and human studies were performed.

Section snippets

Theoretical considerations

The SNR in an MR image depends on conditions as coil geometry, density of the nuclei under investigation, T1, and T2. In addition, the type of the imaging sequence and the chosen parameters, especially voxel size and readout bandwidth Δν = N/Ts (Ts is the readout time and N the number of sampling points) play an important role. The measured noise is mainly caused on the one hand by the electrical resistance of the coil and the signal amplifier, and on the other hand by the sample inside the

B1 homogeneity

Coronal, transverse, and sagittal sections through the B1-field distribution with areas of 20 × 20 cm, 20 × 24 cm, and 24 × 20 cm, respectively, were calculated numerically based on the geometry of both loops of the volume coil as shown in Fig. 2. The gray level steps represent 10% difference of the amplitude in the coil center. The positions of the kidneys, the liver, and the heart are indicated. The dimensions of the coil are shown by dotted lines.

The distribution of the RF magnetic field B1

Discussion

A volume coil design is required for MR imaging of sodium distribution in deeper lying organs of the human torso. The Helmholtz condition, namely that the radius and the distance of the two loops are identical, cannot be exactly fulfilled for torso examinations of most subjects inside the bore of whole body MR units, because the radius would be too large. For this reason and in order to obtain a good filling factor, the design of the coil was modified from the Helmholtz configuration. The

References (17)

  • P.A. Turski et al.

    Clinical and experimental sodium magnetic resonance imaging

    Radiol Clin North Am

    (1988)
  • S. Köhler et al.

    Fast three-dimensional sodium imaging of human brain

    MAGMA

    (2001)
  • J.B. Ra et al.

    In vivo magnetic resonance imaging of sodium in the human body

    Magn Reson Med

    (1988)
  • C.D. Constantinides et al.

    Human skeletal musclesodium MR imaging and quantification–potential applications in exercise and disease

    Radiology

    (2000)
  • T.B. Parrish et al.

    Theoretical Basis for Sodium and Potassium MRI of the Human Heart at 1.5 T

    Magn Reson Med

    (1997)
  • N. Bansal et al.

    Three-dimensional triple quantum-filtered 23Na imaging of rabbit kidney with weighted signal averaging

    J Magn Reson Imaging

    (1995)
  • V. Seshan et al.

    Evaluation of triple quantum-filtered 23Na NMR spectroscopy in the in situ rat liver

    Magn Reson Med

    (1997)
  • R. Kalyanapuram et al.

    Three-dimensional triple-quantum-filtered 23Na imaging of the dog head in vivo

    J Magn Reson Imaging

    (1998)
There are more references available in the full text version of this article.

Cited by (0)

View full text