Search

Parahydrogen-induced polarization (PHIP) is a promising new tool for medical applications of MR, including MRI. The PHIP technique can be used to transfer high non-Boltzmann polarization, derived from parahydrogen, to isotopes with a low natural abundance or low gyromagnetic ratio (e.g. (13)C), thus improving the signal-to-noise ratio by several orders of magnitude. A few molecules acting as metabolic sensors have already been hyperpolarized with PHIP, but the direct hyperpolarization of drugs used to treat neurological disorders has not been accomplished until now. Here, we report on the first successful hyperpolarization of valproate (valproic acid, VPA), an important and commonly used antiepileptic drug. Hyperpolarization was confirmed by detecting the corresponding signal patterns in the (1)H NMR spectrum. To identify the optimal experimental conditions for the conversion of an appropriate VPA precursor, structurally related molecules with different side chains were analyzed in different solvents using various catalytic systems. The presented results include hyperpolarized (13)C NMR spectra and proton images of related systems, confirming their applicability for MR studies. PHIP-based polarization enhancement may provide a new MR technique to monitor the spatial distribution of valproate in brain tissue and to analyze metabolic pathways after valproate administration., (Copyright © 2014 John Wiley & Sons, Ltd.)

Previous studies have shown that valproate is detectable in vitro by 1H magnetic resonance spectroscopy (MRS) at 1.5 T, whereas in patients on valproate monotherapy, no significant dose-dependent valproate signal could be seen. To investigate whether an increased signal-to-noise ratio as provided by higher valproate doses and increased magnetic field strength would enable detection of valproate in vivo, six Wistar rats were examined using volume-selective 1H MRS at 2.34 T. The spectra were analyzed by fitting a linear superposition of the basis spectra of valproate, brain metabolites, and simulated lipid signals. The analysis revealed no significant signal contributions after valproate administration of up to 330 mg/kg body weight. To analyze how underlying mechanisms, such as potential drug interactions with macromolecules, may affect the valproate signal, additional in vitro spectra of valproate were measured before and after adding albumin. The spectra exhibited a strong decrease of the valproate signal with increasing albumin concentration. The results support the hypothesis that in vivo valproate is bound to a high degree to macromolecules and will therefore not be detectable by 1H MRS., (Copyright 2004 Springer-Verlag)

We performed volume-selective 1H MR spectroscopy (1H-MRS) on 12 patients on valproate monotherapy to detect valproate in vivo in the brain. We also acquired reference valproate spectra in vitro in subphysiological 15 g/l albumin solution in saline, in which valproate showed two resonance peaks at 0.7 and 1.2 ppm and a minimum detection threshold of 240 mg/l. In vivo 1H-MRS spectra in brain showed peaks between 0.6 and 1.6 ppm. Simultaneous serum valproate concentrations did not correlate with these integrated MRS peaks. On follow-up, changes in these signals also did not correlate with increasing serum valproate levels. The inconsistency of in vivo 1H-MRS signals at varying serum levels and the high detection levels in vitro suggest that valproate signals are missed in vivo because valproate is metabolised or strongly bound, presumably to brain macromolecules.

Previous studies have shown that valproate is detectable in vitro by 1H magnetic resonance spectroscopy (MRS) at 1.5 T, whereas in patients on valproate monotherapy, no significant dose-dependent valproate signal could be seen. To investigate whether an increased signal-to-noise ratio as provided by higher valproate doses and increased magnetic field strength would enable detection of valproate in vivo, six Wistar rats were examined using volume-selective 1H MRS at 2.34 T. The spectra were analyzed by fitting a linear superposition of the basis spectra of valproate, brain metabolites, and simulated lipid signals. The analysis revealed no significant signal contributions after valproate administration of up to 330 mg/kg body weight. To analyze how underlying mechanisms, such as potential drug interactions with macromolecules, may affect the valproate signal, additional in vitro spectra of valproate were measured before and after adding albumin. The spectra exhibited a strong decrease of the valproate signal with increasing albumin concentration. The results support the hypothesis that in vivo valproate is bound to a high degree to macromolecules and will therefore not be detectable by 1H MRS. [ABSTRACT FROM AUTHOR]