Levels of glutamatergic neurometabolites in patients with severe treatment-resistant schizophrenia: a proton magnetic resonance spectroscopy study

Abstract

Approximately 30% of patients with schizophrenia do not respond to antipsychotics and are thus considered to have treatment-resistant schizophrenia (TRS). To date, only four studies have examined glutamatergic neurometabolite levels using proton magnetic resonance spectroscopy (¹H-MRS) in patients with TRS, collectively suggesting that glutamatergic dysfunction may be implicated in the pathophysiology of TRS. Notably, the TRS patient population in these studies had mild-to-moderate illness severity, which is not entirely reflective of what is observed in clinical practice. In this present work, we compared glutamate + glutamine (Glx) levels in the dorsal anterior cingulate cortex (dACC) and caudate among patients with TRS, patients with non-TRS, and healthy controls (HCs), using 3T ¹H-MRS (PRESS, TE = 35 ms). TRS criteria were defined by severe positive symptoms (i.e., ≥5 on 2 Positive and Negative Syndrome Scale (PANSS)-positive symptom items or ≥4 on 3 PANSS-positive symptom items), despite standard antipsychotic treatment. A total of 95 participants were included (29 TRS patients [PANSS = 111.2 ± 20.4], 33 non-TRS patients [PANSS = 49.8 ± 13.7], and 33 HCs). dACC Glx levels were higher in the TRS group vs. HCs (group effect: F[2,75] = 4.74, p = 0.011; TRS vs. HCs: p = 0.012). No group differences were identified in the caudate. There were no associations between Glx levels and clinical severity in either patient group. Our results are suggestive of greater heterogeneity in TRS relative to non-TRS with respect to dACC Glx levels, necessitating further research to determine biological subtypes of TRS.

Fig. 1. Voxel locations and placement procedures of the ¹H-MRS voxels.

a The dorsal anterior cingulate cortex (dACC) voxel (voxel size: 9.0 mL [3.0 × 2.0 × 1.5 cm³]) was positioned on an oblique axial image acquired parallel to the AC–PC line and oblique sagittal image acquired to parallel to head midline. The tip of the voxel was placed on top of the most anterior part of genu with paralleling to the cingulate cortex. b The caudate voxel was positioned on an oblique axial image acquired parallel to the AC–PC line; the voxel was 7.5 mL (2.5 × 1.5 × 2.0 cm³) and its center was 14 mm superior to the AC–PC line. AC–PC anterior commissure–posterior commissure, dACC dorsal anterior cingulate cortex.

Fig. 2. Representative spectra.

a dACC ¹H-MRS spectra of non-TRS patient. b Caudate ¹H-MRS spectra of non-TRS patient. dACC dorsal anterior cingulate cortex, TRS treatment-resistant schizophrenia.

Fig. 3. Comparisons of glutamate plus glutamine (Glx) levels between groups.

a Glx levels in the dACC in TRS patients, non-TRS patients, and HCs. Glx levels in the dACC were higher in TRS patients than in HCs (p = 0.012 by post-hoc Tukey’s test). b Glx levels in the caudate in TRS patients, non-TRS patients, and HCs. There were no group differences in Glx levels in the caudate. dACC dorsal anterior cingulate cortex, Glx glutamate plus glutamine, HCs healthy controls, TRS treatment-resistant schizophrenia.