Reduced N-acetyl-aspartate (NAA) levels have been reported in the prefrontal cortex (PFC) in patients with schizophrenia using proton magnetic resonance spectroscopy. However, it is unclear whether this NAA reduction predates the illness onset and is reported in subjects at-risk for developing schizophrenia (HRS). The aim of this study was to assess NAA levels in the PFC in HRS. We hypothesized that HRS display lower NAA levels than healthy controls in the PFC. Studies assessing levels of NAA/Creatine (NAA/Cr) in the PFC in HRS were extracted from literature. Meta-analysis tools were used to compute effect sizes of nine selected studies meeting our inclusion criteria (clinical and/or genetic HRS, groups of HRS, and healthy controls matched for age and gender, spectral acquisition in the PFC). We reported that HRS exhibited a significant lower NAA/Cr level (2.15 ± 0.29; n = 208) than healthy controls (2.21 ± 0.32; n = 234) in the PFC with a medium pooled effect size [Hedges's g = -0.42; 95% confidence interval: (-0.61; -0.23); p < 0.0001] corresponding to an average 5.7% of NAA/Cr decrease. Secondary analysis revealed that this reduction was observed in young HRS (<40 years old) who have not reached the peak age of risk for schizophrenia (-11%, g = -0.82, p < 0.00001) but not in old HRS (>40 years old) who have already passed the peak age (g = 0.11, p = 0.56), when they are compared with their matched healthy controls. Our findings suggest that the NAA/Cr reduction in the PFC reported in patients with schizophrenia is observable only in HRS who have not passed the peak age of risk for schizophrenia. NAA/Cr level in the PFC could therefore be considered as a biological vulnerability marker of schizophrenia.
Keywords: MRS; NAA; prefrontal cortex; risk; schizophrenia; vulnerability.