Murray · Brain, behavior, and immunity 2024 · systematic review and meta-analysis · n=21 studies (639 patients, 704 controls)

Measurement of brain glutathione with magnetic Resonance spectroscopy in Schizophrenia-Spectrum disorders - A systematic review and Meta-Analysis.

Cited 24 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of observational case-control neuroimaging studies

PubMed 37769980 · doi:10.1016/j.bbi.2023.09.017 · record verified 2026-08-28

What was done

A systematic review and meta-analysis evaluated 21 in vivo magnetic resonance spectroscopy (MRS) studies assessing brain glutathione levels in individuals with schizophrenia-spectrum disorders compared to healthy controls. The primary analysis examined medial prefrontal cortex glutathione concentrations, subgroup effects across clinical stages (such as stable schizophrenia), and potential confounding from spectroscopy methodology (e.g., voxel size, echo time) and publication year via meta-regression.

What was found

Across all included psychosis patients (N = 639) and controls (N = 704), medial prefrontal glutathione levels did not differ significantly (k = 21, d = -0.09, 95% CI -0.28 to 0.10, p = 0.37). In a subgroup of patients with stable schizophrenia, there was a small but statistically significant reduction in glutathione compared to controls (k = 14, d = -0.20, 95% CI -0.40 to -0.00, p = 0.05). Meta-regression demonstrated that older studies reported larger glutathione reductions, whereas technical parameters including voxel size and echo time had no significant effect.

Why it matters

These results clarify that medial prefrontal glutathione deficits are not a uniform biomarker across all stages of psychosis, but may be subtly present in chronic, stable schizophrenia.

Limits

The included literature consists entirely of observational, cross-sectional case-control imaging studies. The subgroup finding in stable schizophrenia reached only borderline statistical significance (p = 0.05). The abstract does not report data for brain regions outside the medial prefrontal cortex, nor does it quantify statistical heterogeneity (I²), medication exposure, or substance use.

Cited by