Reproducibility of proton magnetic resonance spectroscopic imaging in patients with schizophrenia

Using proton magnetic resonance spectroscopic imaging (¹H-MRSI) we found in a previous study a specific pattern of neuronal pathology in patients with schizophrenia as determined by relative loss of signal from N-acetyl-containing compounds (NAA). The purpose of the present study was to assess the reproducibility of the results of ¹H-MRSI both in patients with schizophrenia and in normal controls. We studied twice 10 patients and 10 controls on 2 days separated by, on average, 3 months. Reproducibility was assessed with several statistical procedures including ANOVA, coefficients of variation (CVs) and intra-class correlation coefficients (ICC). Patients showed significant reductions of NAA/creatine-phosphocreatine (CRE) and NAA/choline-containing compounds (CHO) selectively in the hippocampal region (HIPPO) and in the dorsolateral prefrontal cortex (DLPFC) on both experimental days. A repeated measures ANOVA showed no effect of time on metabolite ratios in all subjects. CVs were fairly low (especially for NAA/CRE and CHO/CRE) and did not differ significantly between patients and controls. The ICCs of the ROIs reached statistical significance only in a few instances. The present multislice ¹H-MRSI study shows that: (1) patients with schizophrenia, when compared as a group to normal controls, show a consistent ¹H-MRSI pattern of group differences, i.e., bilateral reductions of NAA/CRE and NAA/CHO in HIPPO and DLPFC; (2) ¹H-MRSI data in both patients and controls do not show significant changes over this 90-day period; however, absolute metabolite ratios in individuals show low predictability over this time interval; (3) ¹H-MRSI data show relatively low variability (as measured by the CVs) both in patients and normal controls, especially for NAA/CRE and CHO/CRE.