BACKGROUND

The pathophysiology of psychosis is incompletely understood. Disruption in cortical glutamatergic signalling causing aberrant striatal dopamine synthesis capacity is a proposed model for psychosis, but has not been tested in vivo. We therefore aimed to test the relationship between cortical glutamate concentrations and striatal dopamine synthesis capacity, and psychotic symptoms.

METHODS

In this cross-sectional multimodal imaging study, 28 individuals with first-episode psychosis and 28 healthy controls underwent F-DOPA PET (measuring striatal dopamine synthesis capacity), and proton magnetic resonance spectroscopy (measuring anterior cingulate cortex glutamate concentrations). Participants were recruited from first-episode psychosis services in London, UK and were required to be in the first episode of a psychotic illness, with no previous illness or treatment episodes. Exclusion criteria for all participants were: history of substantial head trauma, dependence on illicit substances, medical comorbidity (other than minor illnesses), and contraindications to scanning (such as pregnancy). Symptoms were measured using the Positive and Negative Syndrome Scale. The primary endpoint was the relationship between anterior cingulate cortex glutamate concentrations and striatal dopamine synthesis capacity in individuals with their first episode of psychosis as shown by imaging, examined by linear regression. Linear regression was used to examine relationships between measures.

FINDINGS

Glutamate concentrations showed a significant inverse relationship with striatal dopamine synthesis capacity in patients with psychosis (R=0·16, p=0·03, β -1·71 × 10, SE 0·76 × 10). This relationship remained significant after the addition of age, gender, ethnicity, and medication status to the model (p=0·015). In healthy controls, there was no significant relationship between dopamine and glutamate measures (R=0·04, p=0·39). Positive and Negative Syndrome Scale positive psychotic symptoms were positively associated with striatal dopamine synthesis capacity (R=0·14, p=0·046, β 2546, SE 1217) and showed an inverse relationship with anterior cingulate glutamate concentrations (R=0·16, p=0·03, β -1·71 × 10, SE 7·63 × 10). No relationships were seen with negative symptoms (positive symptoms, mean [SD] -18·4 (6·6) negative symptoms, mean [SD] -15·4 [6·1]).

INTERPRETATION

These observations are consistent with the hypothesis that cortical glutamate dysfunction is related to subcortical dopamine synthesis capacity and psychosis. Although the precise mechanistic relationship between cortical glutamate and dopamine in vivo remains unclear, our findings support further studies to test the effect of modulating cortical glutamate in the treatment of psychosis.

FUNDING

Medical Research Council, Wellcome Trust, Biomedical Research Council, South London and Maudsley NHS Foundation Trust, JMAS Sim Fellowship, Royal College of Physicians (Edinburgh) (SJ).