Materials and methods
Ninety-five (95) participants were recruited – 36 with schizophrenia, 27 with methamphetamine-induced psychosis, and 32 healthy controls (Table 1). All participants underwent the Diagnostic and Statistical Manual IV – Text Revision (DSM-IV-TR) Structured Clinical Interview For DSM-IV-TR Axis I Disorders (SCID) to assess eligibility. Additional clinical questionnaires were completed with all participants, which include the Positive and Negative Syndrome Scale (PANSS), Clinical Global Impressions scale (GGI), Global Assessment of Functioning scale (GAF) and a subjective questionnaire on drug abuse, the Kreek-McHugh-Schluger-Kellogg scale (KMSK). Participants were excluded during screening for chronic medical illnesses known to affect metabolic processes (e.g., hyper/hypo thyroidism, diabetes type I or II, etc.), illnesses where the immune system is dysfunctional or compromised (e.g., HIV, lupus), major brain trauma, and intellectual disability. Additionally, female participants were excluded if there was current or recent pregnancy, or if they were breastfeeding. The study was approved by the Human Research Ethics Committee, Faculty of Health Sciences of the University of Cape Town – HREC Reference Number: 062/2017 and was conducted in accordance with the Declaration of Helsinki .
Additional screening for MRI brain imaging for all participants included ensuring that participants were not claustrophobic or had any form of foreign material in their bodies which could interfere with the MRI scanning process.
***Table 1***
All participants underwent magnetic resonance imaging on a Siemens Skyra 3 Tesla scanner with a 32-channel head coil. A high-resolution Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE) structural image was acquired and was used for placement of single-voxel spectroscopy (SVS) voxels located in the anterior cingulate cortex (ACC) (Figure 1) and left thalamus (Figure 1) as well as chemical-shift imaging (CSI) 2-dimensional voxel grid (Figure 2). Single voxel spectroscopy (SVS) of the ACC and left thalamus were acquired for standard metabolites (PRESS, TE = 30 ms, TR = 2000 ms, 128 averages, delta = -2.6 ppm delta frequency, VOI 20 x 20 mm with a thickness of 15mm, scan time 4:40, with unsuppressed water MRS spectra for the same volume, two averages were acquired). An additional SVS sequence was acquired for the ACC, with parameters optimized for glutamate / glutamine separation (Schubert et al., 2004). The parameters used for this sequence were similar to that of the sequence for standard metabolites, except for the echo time, which was increased to 80 milliseconds (PRESS, TE = 80 ms, TR = 2000 ms, 128 averages, delta = -2.6 ppm delta frequency, VOI 20 x 20 mm with a thickness of 15mm, scan time 8:56, with unsuppressed water MRS spectra for the same volume, two averages were acquired). Partial volume correction was applied to SVS voxels to obtain absolute neurometabolite concentrations. This was achieved with combination of LCModel and MRSParVolCo software. The 2D CSI 1H-MRS slice was acquired (PRESS, TE = 30 ms, TR=2000 ms, Hamming filter, 2 averages, delta = -2.7 ppm delta frequency, weighted phase encoding, FOV = 256 × 256 mm, voxel size 10 × 8 mm, thickness 15 mm, automated CHESS water suppression, scan time 10:52). The slice was positioned to include, bilaterally, the dorsolateral prefrontal cortex (DLPFC), anterior cingulate cortex (ACC), frontal white matter (FWM) located between the DLPFC and ACC. Neurometabolites examined were glutamatergic neurometabolites (glutamate (Glu), glutamine (Gln) and glutamate with glutamine (Glx)), neuroinflammatory neurometabolite (myo-inositol (mI)) and neuronal integrity markers (n-acetyl-aspartate (NAA), and n-acetyl aspartate with n-acetyl-aspartyl glutamate (NAA+NAAG)). CSI neurometabolites are reported in relation to creatine with phosphocreatine (Cr+PCr).
***Figure 1***