Datensatz

Zusammenfassung

Background: Glutamate (Glu) and gamma-amino butyric acid (GABA) dysfunction have been reported in patients with mood disorders using several lines of evidence, like abnormalities in the cerebrospinal fluid (CSF) (Berrettini et al, 1986; Frye et al, 2007) and brain tissue (Eastwood and Harrison, 2010;Torrey et al, 2005) of individuals with bipolar disorder (BD). Not surprisingly, Glu and GABA systems have been targeted for the development of novel medications to treat mood disorders (Krystal et al, 2002). Magnetic resonance spectroscopy (MRS) studies have reported increased brain glutamate (Glu) and Glx (Glu + Glutamine) in subjects with BD. However, data on separate measures of GABA and glutamine (Gln) in BD are sparse due to overlapping resonant signals. The development of new sequence methods in the quantification of these metabolites, allowed us a better understanding of the Glu/GABA-Gln cycle in BD, but data on this field of research is very sparse in BD. Studies that investigated in vivo measurements of small signal metabolites such as GABA and glutamine(Gln)withmagneticresonancespectroscopy(1H-MRS) are scarce due to difficulties in extracting data from spectra with overlapping resonant signals and require a special dimensional processing technic. The analysis of isolated measures of Gln is important and it has been suggested that ratio between Glu and Gln may represent a more valuable index of glutamatergic neurotransmission than either metabolite alone. Methods: Eighty-eight subjects (50 euthymic BD and 38 HC) underwent3Tprotonmagneticresonancespectroscopy(1H-MRS) in the anterior cingulate cortex (ACC; 2x2x4.5cm 3) using a two-dimensional JPRESS sequence. GABA, glutamine (Gln) and glutamate (Glu) were quantified with the ProFit program. Using image segmentation and known Cre concentrations for white matter and grey matter we calculated metabolite concentrations for the excited MRS voxel. Results: GABA levels did not differ between groups. Gln level was higher in euthymic BD patients than in healthy controls. The Glu level and Glu/Gln rayio were lower in BD patients than in controls. Glu/GABA ratio was increased in HC compared to BD. Glu/Gln ratio inversely correlated with GABA level, while Gln level demonstrated a positive correlation with GABA level. The use of anticonvulsants influenced Gln and Glu/Gln ratio. Excluding anticonvulsant users from the BD group the difference on Gln level disappeared but the differences on Glu and Glu/Gln weresustained.Neitherlithiumnorantipsychoticusedidinfluence metabolite levels. Conclusion: To the best of our knowledge, this was the most extensive Glu-Gln-GABA MRS study in euthymic BD type I. Using a very sensitive MRS technique, we reported intriguing evidenceofincreasedglutamatergic(Gln/Glu)activityinthepres- ence of reduced Glu in BD patients. The higher levels of Gln and lowerGlu/Glnratiocouldbeanindicationofincreasedglutamate- glutamine cycling in BD. Given the complexity of this cycle and the fact that there are several other pathways open to astrocytic or neuronal Glu not directly involved in neurotransmission, we recommend that additional studies should further investigate the abnormalities of such a complex cycle in BD.