In vivo observation of quadrupolar splitting in (39)K magnetic resonance spectroscopy of human muscle tissue

• Published in: NMR in biomedicine Vol. 29; no. 4; pp. 451 - 457
• Main Authors: Rösler, M B, Nagel, A M, Umathum, R, Bachert, P, Benkhedah, N
• Format: Journal Article
• Language: English
• Published: England 01.04.2016
• Subjects: Adult; Female; Head; Humans; Magnetic Resonance Spectroscopy - methods; Male; Middle Aged; Muscles - metabolism; Phantoms, Imaging; Potassium - metabolism; Thigh; 39K magnetic resonance imaging; muscle tissue anisotropy; multi-quantum filtering; in vivo quadrupolar splitting; potassium 39K NMR; 7T
• ISSN: 1099-1492
• DOI: 10.1002/nbm.3476

The purpose of this work was to explore the origin of oscillations of the T(*)2 decay curve of (39)K observed in studies of (39)K magnetic resonance imaging of the human thigh. In addition to their magnetic dipole moment, spin-3/2 nuclei possess an electric quadrupole moment. Its interaction with non-vanishing electrical field gradients leads to oscillations in the free induction decay and to splitting of the resonance. All measurements were performed on a 7T whole-body MRI scanner (MAGNETOM 7T, Siemens AG, Erlangen, Germany) with customer-built coils. According to the theory of quadrupolar splitting, a model with three Lorentzian-shaped peaks is appropriate for (39)K NMR spectra of the thigh and calf. The frequency shifts of the satellites depend on the angle between the calf and the static magnetic field. When the leg is oriented parallel to the static magnetic field, the satellites are shifted by about 200 Hz. In the thigh, rank-2 double quantum coherences arising from anisotropic quadrupolar interaction are observed by double-quantum filtration with magic-angle excitation. In addition to the spectra, an image of the thigh with a nominal resolution of (16 × 16 × 32) mm(3) was acquired with this filtering technique in 1:17 h. From the line width of the resonances, (39)K transverse relaxation time constants T(*)2, fast  = (0.51 ± 0.01) ms and T(*)2, slow  = (6.21 ± 0.05) ms for the head were determined. In the thigh, the left and right satellite, both corresponding to the short component of the transverse relaxation time constant, take the following values: T(*)2, fast  = (1.56 ± 0.03) ms and T(*)2, fast  = (1.42 ± 0.03) ms. The centre line, which corresponds to the slow component, is T(*)2, slow  = (9.67 ± 0.04) ms. The acquisition time of the spectra was approximately 10 min. Our results agree well with a non-vanishing electrical field gradient interacting with (39)K nuclei in the intracellular space of muscle tissue.