Corellation Between Gaba/Glutamate Ratio Magnetic Resonance Spectroscopy and Electroencephalogram in Pediatric Epilepsy

Abstract

To investigate the association between the gamma-aminobutyric acid (GABA)/glutamate ratio measured by magnetic resonance spectroscopy (MRS) and electroencephalographic (EEG) findings in children with epilepsy. This prospective cross-sectional study included 33 children with clinically diagnosed epilepsy at Dr. Moewardi Regional General Hospital, Surakarta, Indonesia, from July 2025 to January 2026. Participants underwent routine EEG, structural magnetic resonance imaging (MRI), and single-voxel MRS. The GABA/glutamate ratio was categorized using a reference range of 0.63-0.67. Associations among modalities were examined using contingency coefficients, and Cochran Q testing compared proportions of abnormal findings. MRI interobserver agreement was assessed using Cohen kappa. All participants had an abnormal GABA/glutamate ratio; 24 (72.7%) had a ratio <0.63 and 9 (27.3%) had a ratio >0.67. MRI abnormalities were identified in 28 participants (84.8%) and EEG abnormalities in 25 (75.8%). Abnormal finding proportions differed across modalities (p=0.004). The GABA/glutamate ratio was not significantly associated with EEG (p=0.061) or MRI findings (p=0.200). MRI and EEG were significantly associated (coefficient=0.612, p=0.001). MRI interobserver agreement was very high (kappa=0.901, p<0.001). The GABA/glutamate ratio was not significantly associated with EEG or structural MRI categories. The three modalities captured different dimensions of epilepsy and should be interpreted as complementary assessments.