Glutamate metabolites in treatment resistant schizophrenia: A meta-analysis and systematic review of 1H-MRS studies.

Nearly a third to half of schizophrenia patients are non-responsive to first-line antipsychotics and are labelled treatment resistant schizophrenia (TRS). Neurochemical abnormalities in TRS may not be dopaminergic but possibly glutamate (Glu) related. Studies that have examined glutamatergic abnormalities using proton magnetic resonance spectroscopy (1H-MRS) in TRS, have showed inconsistent results. Hence, we conducted a meta-analysis of 1H-MRS studies comparing levels of Glu-and its metabolites in the brains of TRS and non-treatment resistant schizophrenia (nTRS) patients. Four eligible studies were included in the analysis. Summary effect size for the group difference between TRS (n = 101, including Ultra-TRS) and nTRS (n = 61) in Glu-levels in the anterior cingulate cortex (ACC) as measured with Hedges's g was 0.21 (95% CI: -0.42 to 0.85; p = 0.5) suggesting absence of significant difference. However, on leave one out analysis, one iteration showed significant difference in Glu-levels between the groups (Hedges's g = 0.46; p = 0.02) with higher Glu-levels in TRS implying significant effect of a single study on the effect size. The higher ACC Glu-in TRS was not associated with symptom severity or antipsychotic administration, indicating a possible trait abnormality. The limited number of datasets comparing Glu-metabolites in other brain regions are narratively described. Our analysis is limited by the significant heterogeneity between studies. Further longitudinal, prospective studies are needed to confirm higher Glu-metabolite levels in ACC in TRS and explore this potential trait abnormality.