Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2018 Nov 14;76(2):199–207. doi: 10.1001/jamapsychiatry.2018.3252

Association of Hippocampal Glutamate Levels With Adverse Outcomes in Individuals at Clinical High Risk for Psychosis

Matthijs G Bossong 1,2,✉, Mathilde Antoniades 1, Matilda Azis 1, Carly Samson 1, Beverley Quinn 3, Ilaria Bonoldi 1, Gemma Modinos 1, Jesus Perez 3,4, Oliver D Howes 1, James M Stone 1,5, Paul Allen 1,6, Philip McGuire 1
PMCID: PMC6440239  PMID: 30427993

This cross-sectional study investigates whether increased hippocampal glutamate levels are associated with adverse clinical outcomes in individuals at clinical high risk for psychosis compared with healthy control individuals, between clinical high-risk individuals who develop and do not develop psychosis, and between individuals with good and poor outcomes.

Key Points

Question

What is the association between hippocampal glutamate levels and subsequent clinical outcomes in individuals at clinical high risk for psychosis?

Findings

In this cross-sectional study of 86 individuals from the United Kingdom, baseline hippocampal glutamate levels were significantly higher in clinical high-risk individuals who developed psychosis or had poor functional outcome at a mean clinical follow-up of 18.5 months.

Meaning

The association between adverse clinical outcomes in individuals at clinical high risk for psychosis and increased baseline hippocampal glutamate levels may suggest that these measures could contribute to the stratification of clinical high-risk individuals according to future clinical outcomes.

Abstract

Importance

Preclinical and human data suggest that hippocampal dysfunction plays a critical role in the onset of psychosis. Neural hyperactivity in the hippocampus is thought to drive an increase in subcortical dopamine function through glutamatergic projections to the striatum.

Objective

To examine the association between hippocampal glutamate levels in individuals at clinical high risk for psychosis and their subsequent clinical outcomes.

Design, Setting, and Participants

This cross-sectional study of 86 individuals at clinical high risk for psychosis and 30 healthy control individuals, with a mean follow-up of 18.5 months, was conducted between November 1, 2011, and November 1, 2017, at early detection services in London and Cambridge, United Kingdom.

Main Outcomes and Measures

Concentrations of glutamate and other metabolites were measured in the left hippocampus using 3-T proton magnetic resonance spectroscopy at the first clinical presentation. At follow-up, clinical outcomes were assessed in terms of transition or nontransition to psychosis using the Comprehensive Assessment of the At-Risk Mental State criteria and the level of overall functioning using the Global Assessment of Function scale.

Results

Of 116 total participants, 86 were at clinical high risk for psychosis (50 [58%] male; mean [SD] age, 22.4 [3.5] years) and 30 were healthy controls (14 [47%] male; mean [SD] age, 24.7 [3.8] years). At follow-up, 12 clinical high-risk individuals developed a first episode of psychosis whereas 74 clinical high-risk individuals did not; 19 clinical high-risk individuals showed good overall functioning (Global Assessment of Function ≥65), whereas 38 clinical high-risk individuals had a poor functional outcome (Global Assessment of Function <65). Compared with clinical high-risk individuals who did not become psychotic, clinical high-risk individuals who developed psychosis showed higher hippocampal glutamate levels (mean [SD], 8.33 [1.48] vs 9.16 [1.28] glutamate levels; P = .048). The clinical high-risk individuals who developed psychosis also had higher myo-inositol levels (mean [SD], 7.60 [1.23] vs 6.24 [1.36] myo-inositol levels; P = .002) and higher creatine levels (mean [SD], 8.18 [0.74] vs 7.32 [1.09] creatine levels; P = .01) compared with clinical high-risk individuals who did not become psychotic, and higher myo-inositol levels compared with healthy controls (mean [SD], 7.60 [1.23] vs 6.19 [1.51] myo-inositol levels; P = .005). Higher hippocampal glutamate levels in clinical high-risk individuals were also associated with a poor functional outcome (mean [SD], 8.83 [1.43] vs 7.76 [1.40] glutamate levels; P = .02). In the logistic regression analyses, hippocampal glutamate levels were significantly associated with clinical outcome in terms of transition and nontransition to psychosis (β = 0.48; odds ratio = 1.61; 95% CI, 1.00-2.59; P = .05) and overall functioning (β = 0.53; odds ratio = 1.71; 95% CI, 1.10-2.66; P = .02).

Conclusions and Relevance

The findings indicate that adverse clinical outcomes in individuals at clinical high risk for psychosis may be associated with an increase in baseline hippocampal glutamate levels, as well as an increase in myo-inositol and creatine levels. This conclusion suggests that these measures could contribute to the stratification of clinical high-risk individuals according to future clinical outcomes.

Introduction

Both preclinical and human studies suggest that hippocampal dysfunction plays a critical role in the onset of psychosis. Data from animal models indicate that neural hyperactivity in the hippocampus drives an increase in subcortical dopamine function through glutamatergic projections to the striatum.1,2 Neuroimaging studies in individuals at clinical high risk for psychosis suggest that the subsequent onset of psychosis is associated with changes in several measures of hippocampal integrity, including hypermetabolism,3 increased resting perfusion,4 altered activation in response to cognitive tasks,5 and reduced gray matter volume.3,6,7 The mechanisms underlying these changes are unclear, but experimental work in rodents suggests that these changes may be secondary to increases in hippocampal glutamate levels.3

A large body of independent research suggests that psychosis involves alterations in glutamate neurotransmission.8,9 For example, noncompetitive N-methyl-D-aspartate receptor antagonists such as ketamine hydrochloride and phencyclidine can induce psychotic symptoms in healthy individuals,10,11 and exacerbate psychotic symptoms in individuals with a psychotic disorder.12,13 In addition, autoantibodies to the N-methyl-D-aspartate receptor are present in a proportion of individuals with psychosis,14,15 and several risk genes associated with psychosis code for proteins involved in glutamatergic neurotransmission.16

Brain glutamate levels can be measured in vivo using proton magnetic resonance spectroscopy (1H-MRS). A meta-analysis on levels in levels of glutamatergic metabolites in individuals with psychosis suggests that there are elevations of glutamatergic metabolites in several brain regions, including increased concentrations of Glx (a combined measure of glutamine and glutamate) in the medial temporal lobe.17 The few 1H-MRS studies that examined hippocampal glutamate concentrations in clinical high-risk individuals did not find differences compared with healthy control individuals,18,19,20 but they did not investigate hippocampal glutamate concentrations in association with clinical outcomes. Associations between glutamate levels and adverse outcomes in clinical high-risk individuals have been identified in 1H-MRS studies of other brain regions. de la Fuente Sandoval et al21 found that glutamate levels in the striatum were elevated in clinical high-risk individuals who developed psychosis subsequent to scanning but not in clinical high-risk individuals who did not. Furthermore, in the thalamus, low baseline glutamate levels were associated with poor functioning at clinical follow-up22 and with a failure to achieve symptomatic remission from the clinical high-risk state.23

The primary aim of the present study was to investigate the association between hippocampal glutamate levels in clinical high-risk individuals and subsequent clinical outcomes. We used 1H-MRS to examine a sample of clinical high-risk individuals and a group of healthy volunteers. Clinical high-risk individuals were followed up to determine their clinical outcomes, which were assessed in terms of transition or nontransition to psychosis and level of overall functioning. Our primary hypothesis was that in clinical high-risk individuals, elevated hippocampal glutamate levels at baseline would be associated with the following adverse clinical outcomes: the onset of psychosis and a low level of overall functioning. In view of the evidence of a more general disruption of hippocampal function before the onset of psychosis, we also explored the association between clinical outcomes and levels of other hippocampal metabolites.

Methods

Participants

A total of 116 individuals participated in the study, including 86 individuals at high risk and 30 healthy controls. The study protocol was approved by the National Research Ethics Service Committee of London-Camberwell St Giles, United Kingdom, and all participants gave written informed consent. The study was conducted between November 1, 2011, and November 1, 2017.

The 86 clinical high-risk individuals were recruited in the United Kingdom through the following early detection services for people at clinical high risk for psychosis: Outreach and Support in South London, the West London Early Intervention service, and the Cambridge Early Onset service.24 The diagnosis was made using the Comprehensive Assessment of the At-Risk Mental State.25 Individuals met 1 or more of the following criteria: (1) attenuated psychotic symptoms; (2) brief, limited intermittent psychotic symptoms (a history of 1 or more episodes of frank psychotic symptoms that resolved spontaneously within 1 week in the past year); or (3) a recent decline in function, together with either the presence of schizotypal personality disorder or a family history of psychosis in a first-degree relative.

Thirty healthy controls were recruited from the local community. All were native English speakers and had no history of psychiatric disorder. None of the healthy controls was using prescription medication.

On the day of scanning, symptoms of clinical high-risk patients were assessed using the Comprehensive Assessment of the At-Risk Mental State.26 For all participants, psychosocial functioning was examined using the Global Assessment of Function (GAF) scale,26 measures of anxiety were obtained using the Hamilton Anxiety Rating Scale, and measures of depression were obtained using the Hamilton Depression Rating Scale.27,28 Premorbid IQ was assessed using the National Adult Reading Test,29 and handedness was determined using the Annett Handedness Scale.30 Participants provided information on tobacco use (number of cigarettes per day) and cannabis use (0 indicated no cannabis use; 1, experimental use; 2, occasional use; 3, moderate use; and 4, severe use). Participants were excluded if they reported illicit substance use during the week before scanning or alcohol use during the 24 hours before scanning, met Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) criteria for a substance misuse or dependence disorder, or had a history of neurologic or psychotic disorders.

Clinical Follow-up

The clinical high-risk individual sample was followed up to determine clinical outcomes. Fifty-seven individuals underwent a face-to-face clinical reassessment. The mean (SD) interval between the baseline and follow-up assessments was 18.5 (9.6) months (range, 4-59 months). Clinical outcome was assessed as transition or nontransition to psychosis, defined using the criteria in the Comprehensive Assessment of the At-Risk Mental State,25 and the level of overall functioning determined using the GAF scale. Of the 86 clinical high-risk individuals, 29 (34%) could not be interviewed again because they were too unwell, declined to be seen, or were unable to be contacted. For these individuals, the transition to psychosis was determined from information in their clinical records, but it was not possible to rate their overall functioning.

Proton Magnetic Resonance Spectroscopy

Images were obtained on a 3.0T HDx MR system (General Electric Healthcare). The 1H-MRS spectra (Point RESolved Spectroscopy; 30-millisecond echo time; 3000-millisecond repetition time; 96 averages) were acquired in the left hippocampus (Figure 1).19 We used the standard General Electric probe (proton brain examination) sequence, which uses a standardized, chemically selective suppression, water suppression routine. For each metabolite spectrum, unsuppressed water reference spectra (16 averages) were also part of the standard acquisition. Shimming and water suppression were optimized, with auto-prescan performed twice before each scan. Using standardized protocols, the hippocampal voxel (20 mm [right-left] × 20 mm [anterior-posterior] × 15 mm [superior-inferior]) was prescribed from the structural T1-weighted scan.

Figure 1. Example of Proton Magnetic Resonance Spectroscopy (1H-MRS) Voxel Placement and Spectrum.

Figure 1.

Open in a new tab

A, Example of 1H-MRS voxel placement in the left hippocampus is indicated by the box. B, 1H-MRS spectrum obtained from the voxel in A (black line) and the overlay of the spectral fit (red line). All spectra were analyzed with LCModel version 6.3-0A (S.W. Provencher). Cho indicates choline; Cre, creatine; Glu, glutamate; Glx, combined measure of glutamine and glutamate; mI, myo-inositol; NAA, N-acetylasparate; and ppm, parts per million.

Structural Magnetic Resonance Imaging

Structural images were acquired in the same session using a whole-brain, 3-dimensional, sagittal T1-weighted scan, with parameters based on the Alzheimer Disease Neuroimaging Initiative (2.85-millisecond echo time; 6.98-millisecond repetition time, inversion time of 400 milliseconds, flip angle of 11°, and voxel size of 1.0 × 1.0 × 1.2 mm) (http://adni.loni.usc.edu/methods/mri-tool/mri-acquisition/). Structural T1-weighted images were segmented into gray matter, white matter, and cerebrospinal fluid using Statistical Parametric Mapping software, version SPM8 (Wellcome Trust Centre for Neuroimaging) to allow correction of the 1H-MRS data for partial volume cerebrospinal fluid contamination.

1H-MRS Data Processing

All spectra were analyzed with LCModel, version 6.3-0A (S.W. Provencher)31 using a standard basis set of 16 metabolites (L-alanine, aspartate, creatine, phosphocreatine, -aminobutyric acid, glucose, glutamine, glutamate, glycerophosphocholine [choline], glycine, myo-inositol, L-lactate, N-acetylaspartate, N-acetylaspartylglutamate, phosphocholine, and taurine), acquired with the same field strength (3T), localization sequence (Point RESolved Spectroscopy), and echo time (30 milliseconds). Model metabolites and concentrations used in the basis set are fully detailed in the LCModel manual.32 Poorly fitted metabolite peaks (Cramer-Rao minimum variance bounds >20% as reported by LCModel) were excluded from further analysis, and water-scaled glutamate, Glx, myo-inositol, creatine, choline, and N-acetylaspartate values were corrected for voxel tissue composition (eMethods in the Supplement). The eTables 1-3 in the Supplement give the scan-quality parameters and voxel tissue composition.

Statistical Analysis

Group differences in clinical and demographic variables were assessed using 2-sample t tests or χ2 tests. To examine the association between metabolite levels and clinical outcomes, the clinical high-risk sample was dichotomized according to transition vs nontransition to psychosis25 and good overall functioning (GAF≥65) vs poor overall functioning (GAF<65) at follow-up.22 Because the primary hypothesis involved the association between hippocampal glutamate levels and clinical outcomes, general linear models were used to identify group differences in glutamate levels between the respective clinical high-risk subgroups and healthy controls, as well as between the total clinical high-risk group and healthy controls. Glutamate concentrations were included as the dependent variable with group as the independent variable (2-tailed P < .05 was considered to be statistically significant). Concentrations of other metabolites (Glx, myo-inositol, creatine, choline, and N-acetylaspartate) were also assessed in exploratory general linear models and were corrected for multiple comparisons (thresholded P < .01). Multiple regression analyses were performed to examine how hippocampal glutamate levels were associated with clinical outcomes. Age and tobacco use were included as covariates in all analyses because both can influence neurometabolite levels.33,34 All analyses were performed in SPSS, version 22 (SPSS Inc). Effect sizes are reported as Hedges g.

Results

Demographic, Clinical, and Medication Data

Of 116 total participants, 86 were at clinical high risk for psychosis (50 [58%] male; mean [SD] age, 22.4 [3.5] years) and 30 were healthy controls (14 [47%] male; mean [SD] age, 24.7 [3.8] years). All 86 clinical high-risk participants met the Attenuated Psychotic Symptoms diagnostic criteria, with some participants also fulfilling the brief limited intermittent psychotic symptoms criteria (n = 5) or schizotypy/familial risk criteria (n = 2). At the time of scanning, 72 of 86 clinical high-risk individuals (84%) were naive to antipsychotic medication. Ten clinical high-risk individuals were receiving low doses of antipsychotic medication (<1.5-mg haloperidol equivalents per day).

The clinical high-risk and healthy control groups did not differ significantly in terms of sex, IQ, handedness, or cannabis use. However, the clinical high-risk group was younger (mean [SD] age, 22.4 [3.5] vs 24.7 [3.8] years; P = .005), had fewer years of education (mean [SD], 14.5 [2.2] vs 15.8 [3.3] years; P = .02), and smoked more cigarettes (mean [SD], 5.5 [8.5] vs 1.9 [3.3] cigarettes; P = .02) compared with healthy controls. As expected, they also had higher Hamilton Anxiety Rating Scale scores (mean [SD], 18.4 [11.0] vs 3.6 [4.2] scores; P < .001), Hamilton Depression Rating Scale scores (mean [SD], 17.4 [11.0] vs 1.7 [3.6] scores; P < .001), and lower levels of functioning at baseline (mean [SD], 57.7 [9.4] vs 93.0 [5.1] GAF scores; P < .001) compared with controls (Table 1).

Table 1. Baseline Demographic, Clinical, and Medication Data.

Measure Healthy Controls vs CHR Individuals CHR Transition vs Nontransition Subgroups CHR Good vs Poor Functional Outcome Subgroups
Healthy Controls (n = 30) CHR Individuals (n = 86) P Value Nontransition Group (n = 74) Transision Group (n = 12) P Value Good Outcome (n = 19) Poor Outcome (n = 38) P Value
Age, mean (SD), y 24.7 (3.8) 22.4 (3.5) .005 22.5 (3.7) 22.1 (2.8) .71 22.1 (3.3) 23.3 (3.9) .24
NART IQ, mean (SD) 104.8 (13.6) 103.9 (12.2) .75 104.7 (12.0) 99.3 (12.5) .17 106.3 (9.2) 105.2 (13.3) .76
Education, mean (SD), y 15.8 (3.3) 14.5 (2.2) .02 14.6 (2.1) 14.3 (2.5) .71 14.8 (2.3) 14.3 (2.1) .45
CAARMS score, mean (SD)
Positive NA 10.2 (4.2) NA 10.0 (4.4) 11.3 (3.3) .33 10.2 (4.2) 10.5 (4.3) .75
Negative NA 5.5 (4.2) NA 5.4 (4.1) 6.1 (4.9) .63 6.4 (3.9) 5.5 (4.3) .44
Total NA 43.6 (21.8) NA 42.8 (21.7) 48.9 (22.9) .39 45.5 (19.9) 42.7 (20.8) .63
Baseline GAF score, mean (SD) 93.0 (5.1) 57.7 (9.4) <.001 58.4 (9.5) 53.6 (7.6) .11 56.8 (8.9) 54.8 (9.6) .48
HAM-A score, mean (SD) 3.6 (4.2) 18.4 (11.0) <.001 17.1 (10.3) 27.3 (12.1) .01 18.3 (12.8) 20.6 (11.8) .57
HAM-D score, mean (SD) 1.7 (3.6) 17.4 (11.0) <.001 16.4 (11.1) 24.4 (8.2) .05 15.5 (10.3) 19.6 (11.9) .28
Tobacco use, mean (SD), cigarettes/d 1.9 (3.3) 5.5 (8.5) .02 6.1 (8.9) 1.83 (3.6) .11 6.7 (10.0) 5.7 (8.5) .70
Alcohol use, mean (SD), U/daya 1.6 (2.2) 1.5 (3.1) .82 1.6 (3.4) 0.83 (0.72) .44 1.4 (1.0) 1.5 (4.0) .91
Cannabis use, median (range)b 0 (0-3) 0 (0-4) .71 0 (0-4) 0 (0-4) .81 1 (0-4) 0 (0-3) .18
Antipsychotic medication, No. (%) 0 10 (12) NA 10 (13) 1 (0.08) .63 3 (16) 2 (5) .19
Male sex, No. (%) 14 (47) 50 (58) .28 43 (58) 7 (58) .95 9 (47) 23 (61) .35
Right-handed, No. (%) 27 (90) 70 (81) .13 60 (80) 11 (92) .33 17 (90) 29 (76) .24
Open in a new tab

Abbreviations: CAARMS, Comprehensive Assessment for the At-Risk Mental State; CHR, clinical high-risk; GAF, Global Assessment of Functioning scale; HAM-A, Hamilton Anxiety Rating Scale; HAM-D, Hamilton Depression Rating Scale; NA, not applicable; NART, National Adult Reading Test.

a

Units/d is defined as alcohol units per day where 1 U equals 10 mL or 8g of pure alcohol, which is approximately the amount of alcohol that the average adult can metabolize in an hour.

b

0 indicates never; 1, experimental use; 2, occasional use; 3, moderate use; 4, severe use.

At follow-up of 18.5 months, 12 clinical high-risk individuals (14%) developed a first episode of psychosis (transition group) and 74 (86%) did not (nontransition group). When dichotomized according to their GAF scores, 19 of the 57 clinical high-risk individuals who were reinterviewed (33%) showed good overall functioning (GAF≥65), whereas 38 of the 57 clinical high-risk individuals who were reinterviewed (67%) had a poor functional outcome (GAF<65).

The transition group had higher baseline Hamilton Anxiety Rating Scale scores (mean [SD], 27.3 [12.1] vs 17.1 [10.3] scores; P < .01) and Hamilton Depression Rating Scale scores (mean [SD], 24.4 [8.2] vs 16.4 [11.1] scores; P < .05) than did the nontransition group, but there were no other significant differences in symptom ratings or demographic measures between these subgroups. There were no significant differences at baseline in any clinical or demographic measures between the good outcome and poor outcome subgroups (Table 1) (eTables 4 and 5 in the Supplement).

Hippocampal Metabolite Differences

Results from logistic regression analyses showed that hippocampal glutamate levels were significantly associated with clinical outcome in terms of transition and nontransition to psychosis (β = 0.48; odds ratio = 1.61; 95% CI, 1.00-2.59; P = .05) and overall functioning (β = 0.53; odds ratio = 1.71; 95% CI, 1.10-2.66; P = .02). There were no significant group differences in any of the metabolite concentrations between the total clinical high-risk group (independent of outcomes) and healthy controls (Table 2 and eFigures 1-3 in the Supplement).

Table 2. Hippocampal Metabolite Levels in Healthy Controls and Clinical High-Risk Individuals.

Metabolite Healthy Controls, Mean (SD) (n = 30) Clinical High-Risk Individuals, Mean (SD)
(n = 86)		 Analysis
F P Value
Glutamate 8.31 (1.12) 8.45 (1.48) 0.49 .48
Glx 11.61 (2.23) 11.57 (2.45) 0.01 .92
myo-Inositol 6.19 (1.51) 6.43 (1.42) 2.11 .15
Creatine 7.42 (1.10) 7.43 (1.08) 0.20 .66
Choline 2.30 (0.40) 2.41 (0.42) 1.95 .17
N-acetylaspartate 9.34 (1.43) 9.36 (1.13) 0.002 .97
Open in a new tab

Abbreviation: Glx, combined measure of glutamine and glutamate.

Transition to Psychosis

The transition subgroup had significantly higher hippocampal glutamate levels than did the nontransition subgroup (mean [SD], 9.16 [1.28] vs 8.33 [1.48]; Hedges g, 0.57; F3,81 = 4.03; P = .048) and higher glutamate levels compared with healthy controls (mean [SD], 9.16 [1.28] vs 8.31 [1.12]; Hedges g, 0.73; F3,38 = 3.54; P = .07) although the latter fell short of statistical significance. There was no difference in glutamate levels between the nontransition and healthy control groups (Figure 2 and eFigure 1 and eTable 6 in the Supplement).

Figure 2. Hippocampal Metabolite Concentrations and Transition to Psychosis.

Figure 2.

Open in a new tab

Left hippocampal metabolite concentrations in 30 healthy controls, 74 clinical high-risk patients who did not develop psychosis (nontransition group), and 12 clinical high-risk patients who developed psychosis (transition group). au indicates arbitrary units.

aP < .05.

bP < .01.

Exploratory testing revealed that the transition group also had significantly higher hippocampal myo-inositol levels than did the nontransition group (mean [SD], 7.60 [1.23] vs 6.24 [1.36]; F3,81 = 10.26, P = .002), higher creatine levels than did the nontransition group (mean [SD], 8.18 [0.74] vs 7.32 [1.09]; F3,82 = 7.26; P = .01), and higher myo-inositol levels than did the healthy controls (mean [SD], 7.60 [1.23] vs 6.19 [1.51]; F3,38 = 8.82; P = .005). The differences in myo-inositol levels were large; in the transition group, the concentration was 22% higher than in the nontransition group (Hedges g, 1.01) and 23% higher than in healthy controls (Hedges g, 0.98). In contrast, there were no significant differences between nontransition individuals and healthy controls in the levels of any hippocampal metabolite (Figure 2, eFigures 2 and 3, and eTable 6 in the Supplement).

Functional Outcome

Clinical high-risk individuals with a poor functional outcome had significantly higher glutamate levels than did those with a good outcome (mean [SD], 8.83 [1.43] vs 7.76 [1.40]; Hedges g, 0.75; F3,52 = 6.39; P = .02). There were no other significant differences in metabolite levels. None of the metabolite levels was significantly different between either of the clinical high-risk functional outcome subgroups and the healthy controls (Figure 3, eFigures 1-3, and eTable 7 in the Supplement).

Figure 3. Hippocampal Metabolite Concentrations and Functional Outcome.

Figure 3.

Open in a new tab

Left hippocampal metabolite concentrations in 30 healthy controls, 19 clinical high-risk patients with a good functional outcome, and 38 clinical high-risk patients with a poor functional outcome. au, arbitrary units.

aP < .05.

Discussion

To our knowledge, this was the largest 1H-MRS study of metabolite levels in individuals at clinical high risk of psychosis conducted to date. The overall finding was that adverse clinical outcomes in these individuals were associated with increases in hippocampal glutamate levels and in the levels of some other metabolites. The subsequent onset of psychosis was associated with higher baseline levels of glutamate, myo-inositol, and creatine at first clinical presentation, and a low level of functioning at follow-up was associated with increased glutamate levels. In contrast to the differences within the clinical high-risk group, there were no differences in metabolite levels between the total clinical high-risk sample and healthy controls or between clinical high-risk individuals who did not have adverse clinical outcomes and controls.

In line with our main hypothesis, increased hippocampal glutamate levels at baseline were associated with adverse clinical outcomes at follow-up, including onset of psychosis and a low level of overall functioning. These observations are consistent with preclinical and human data implicating hippocampal dysfunction and glutamate transmission in the development of psychosis. In preclinical models, neural hyperactivity of the hippocampus drives an increase in subcortical dopamine activity through glutamatergic projections to the striatum.1,2 Neuroimaging data from clinical high-risk samples indicate that the subgroup of individuals who subsequently develop psychosis had increased resting hippocampal metabolism3 and perfusion,4 altered hippocampal response to cognitive tasks,5 and smaller hippocampal volumes.3,6,7 As previously suggested by experimental work in rodents,3 one possibility is that these alterations are secondary to increases in hippocampal glutamate levels. Consistent with data from previous 1H-MRS studies,18,19,20 there were no differences in hippocampal glutamate levels between the clinical high-risk nontransition group or the overall clinical high-risk group (independent of clinical outcomes) and controls. This finding is also in line with previous studies using other neuroimaging modalities that showed differences within the clinical high-risk group rather than between the overall clinical high-risk group and controls in terms of hippocampal volume,7 brain activity patterns,5 and dopamine synthesis capacity.35 However, adverse clinical outcomes in clinical high-risk individuals have been linked to altered glutamate metabolite levels in other brain regions. de la Fuente Sandoval et al21 demonstrated increased baseline glutamate levels in the striatum of clinical high-risk individuals who developed a first episode of psychosis. Allen et al22 found that a poor functional outcome in clinical high-risk individuals was linked to lower glutamate concentrations in the thalamus at baseline, whereas Egerton et al23 reported that lower thalamic glutamate levels were associated with a failure to achieve symptomatic remission from the clinical high-risk state.

Our second main finding was that adverse clinical outcomes were also associated with elevations in the levels of myo-inositol and creatine in the hippocampus. For both these metabolites and for glutamate, the pattern of group differences was similar, with higher levels in clinical high-risk individuals who developed psychosis compared with those who did not become psychotic (Figure 2, eFigures 1-3, and eTable 6 in the Supplement). This consistent pattern across different metabolites suggests that the onset of psychosis was associated with a more general increase in hippocampal metabolite levels, as opposed to a change that was specific to glutamate. Such a widespread change in metabolites is consistent with previous evidence that the subsequent onset of psychosis in clinical high-risk individuals is associated with an overall change in hippocampal integrity, as indicated by hypermetabolism,3 increased resting perfusion,4 and reduced gray matter volume.3,6,7 Although previous 1H-MRS studies in clinical high-risk individuals have not reported associations between clinical outcomes and changes across multiple metabolites, higher levels of glutamate, myo-inositol, and choline have been described in the striatum in medication-naive individuals experiencing their first episode compared with controls.36,37

In the present study, the elevation in myo-inositol levels was large, with concentrations around 22% higher (and effect sizes approximately 1.0) in the clinical high-risk individuals who developed psychosis than in both clinical high-risk individuals who did not develop psychosis and healthy controls. Myo-inositol is regarded as a marker for glial activation,38 and independent data from positron emission tomographic studies of glial activity have reported that this is increased in the hippocampus (and in other regions) in clinical high-risk individuals and those with psychosis.39,40,41 The 1H-MRS studies of myo-inositol and creatine levels in the hippocampus in individuals with chronic psychosis have not found a consistent pattern of differences in comparison with controls.42,43,44,45 However, inconsistencies in 1H-MRS findings in individuals with chronic psychosis may be associated with confounding effects of age, duration of illness, and treatment,17 and alterations in metabolite levels may be more marked in the early than the later stages of the disorder.17,46

Limitations

Because the number of clinical high-risk individuals who developed psychosis was modest (12 of 86 patients [14%]), we cannot exclude the possibility that additional findings were undetected because of limited statistical power. This issue could be addressed by studying larger clinical high-risk patient samples, which can be achieved by combining 1H-MRS data from multiple centers.47 Although the mean time of clinical follow-up was 18.5 months, the variance in duration of follow-up intervals was fairly high (range, 4-59 months). The main reason was that follow-up times were not a priori defined. A recent study of transitions in our early intervention service showed that about 60% of the transitions occurred in the first 18 months, with the rate decreasing thereafter.48 Our findings could be confounded by the use of antipsychotic treatment. This is unlikely, however, because most of the clinical high-risk individuals (72 of 86 [84%]) were naive to antipsychotic medications, and if treated, low doses of antipsychotics were prescribed. Moreover, there were no significant differences in any of the hippocampal metabolites between medicated and unmedicated clinical high-risk individuals. Residual effects of illicit substance use cannot be excluded because this was checked by self-report rather than by urine toxicology screening. Given the dimensions and orientation of our 1H-MRS voxel, other medial temporal lobe regions than the hippocampus, such as the parahippocampal gyrus, are also included in the voxel, which may have confounded our results. Although 1H-MRS values were corrected for cerebrospinal fluid volume, we cannot exclude the possibility that increased metabolite concentrations were associated with changes in hippocampal volume. Finally, by using conventional 1H-MRS, it is not possible to determine whether differences in glutamate levels are associated with neurotransmission or metabolism, an issue that may be addressed by using more sophisticated MRS protocols.49

Conclusions

Our study suggests that clinical outcomes in patients at clinical high risk for psychosis may be associated with baseline hippocampal metabolite concentrations. Although the findings require replication, they raise the possibility that measuring hippocampal metabolite levels could contribute to the stratification of clinical high-risk individuals according to future clinical outcomes.

Supplement.

eMethods. 1H-MRS Data Processing

eFigure 1. Hippocampal Glutamate Concentrations

eFigure 2. Hippocampal Myo-inositol Concentrations

eFigure 3. Hippocampal Creatine Concentrations

eTable 1. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls vs Clinical High-Risk Subjects

eTable 2. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T)

eTable 3. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

eTable 4. Participant Demographic, Clinical, and Medication Data at Baseline: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T)

eTable 5. Participant Demographic, Clinical, and Medication Data at Baseline: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

eTable 6. Hippocampal Metabolite Concentrations and Transition to Psychosis: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T).

eTable 7. Hippocampal Metabolite Concentrations and Functional Outcome: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

Click here for additional data file. (668.2KB, pdf)

References

  • 1.Grace AA. Dysregulation of the dopamine system in the pathophysiology of schizophrenia and depression. Nat Rev Neurosci. 2016;17(8):524-532. doi: 10.1038/nrn.2016.57 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lodge DJ, Grace AA. Hippocampal dysregulation of dopamine system function and the pathophysiology of schizophrenia. Trends Pharmacol Sci. 2011;32(9):507-513. doi: 10.1016/j.tips.2011.05.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Schobel SA, Chaudhury NH, Khan UA, et al. Imaging patients with psychosis and a mouse model establishes a spreading pattern of hippocampal dysfunction and implicates glutamate as a driver. Neuron. 2013;78(1):81-93. doi: 10.1016/j.neuron.2013.02.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Allen P, Chaddock CA, Egerton A, et al. Resting hyperperfusion of the hippocampus, midbrain, and basal ganglia in people at high risk for psychosis. Am J Psychiatry. 2016;173(4):392-399. doi: 10.1176/appi.ajp.2015.15040485 [DOI] [PubMed] [Google Scholar]
  • 5.Allen P, Seal ML, Valli I, et al. Altered prefrontal and hippocampal function during verbal encoding and recognition in people with prodromal symptoms of psychosis. Schizophr Bull. 2011;37(4):746-756. doi: 10.1093/schbul/sbp113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Pantelis C, Velakoulis D, McGorry PD, et al. Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet. 2003;361(9354):281-288. doi: 10.1016/S0140-6736(03)12323-9 [DOI] [PubMed] [Google Scholar]
  • 7.Mechelli A, Riecher-Rössler A, Meisenzahl EM, et al. Neuroanatomical abnormalities that predate the onset of psychosis: a multicenter study. Arch Gen Psychiatry. 2011;68(5):489-495. doi: 10.1001/archgenpsychiatry.2011.42 [DOI] [PubMed] [Google Scholar]
  • 8.Konradi C, Heckers S. Molecular aspects of glutamate dysregulation: implications for schizophrenia and its treatment. Pharmacol Ther. 2003;97(2):153-179. doi: 10.1016/S0163-7258(02)00328-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Moghaddam B, Javitt D. From revolution to evolution: the glutamate hypothesis of schizophrenia and its implication for treatment. Neuropsychopharmacology. 2012;37(1):4-15. doi: 10.1038/npp.2011.181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Javitt DC, Zukin SR. Recent advances in the phencyclidine model of schizophrenia. Am J Psychiatry. 1991;148(10):1301-1308. doi: 10.1176/ajp.148.10.1301 [DOI] [PubMed] [Google Scholar]
  • 11.Krystal JH, Karper LP, Seibyl JP, et al. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry. 1994;51(3):199-214. doi: 10.1001/archpsyc.1994.03950030035004 [DOI] [PubMed] [Google Scholar]
  • 12.Lahti AC, Koffel B, LaPorte D, Tamminga CA. Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacology. 1995;13(1):9-19. doi: 10.1016/0893-133X(94)00131-I [DOI] [PubMed] [Google Scholar]
  • 13.Malhotra AK, Pinals DA, Adler CM, et al. Ketamine-induced exacerbation of psychotic symptoms and cognitive impairment in neuroleptic-free schizophrenics. Neuropsychopharmacology. 1997;17(3):141-150. doi: 10.1016/S0893-133X(97)00036-5 [DOI] [PubMed] [Google Scholar]
  • 14.Pollak TA, Beck K, Irani SR, Howes OD, David AS, McGuire PK. Autoantibodies to central nervous system neuronal surface antigens: psychiatric symptoms and psychopharmacological implications. Psychopharmacology (Berl). 2016;233(9):1605-1621. doi: 10.1007/s00213-015-4156-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lennox BR, Palmer-Cooper EC, Pollak T, et al. ; PPiP study team . Prevalence and clinical characteristics of serum neuronal cell surface antibodies in first-episode psychosis: a case-control study. Lancet Psychiatry. 2017;4(1):42-48. doi: 10.1016/S2215-0366(16)30375-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Schizophrenia Working Group of the Psychiatric Genomics Consortium Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421-427. doi: 10.1038/nature13595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Merritt K, Egerton A, Kempton MJ, Taylor MJ, McGuire PK. Nature of glutamate alterations in schizophrenia: a meta-analysis of proton magnetic resonance spectroscopy studies. JAMA Psychiatry. 2016;73(7):665-674. doi: 10.1001/jamapsychiatry.2016.0442 [DOI] [PubMed] [Google Scholar]
  • 18.Nenadic I, Maitra R, Basu S, et al. Associations of hippocampal metabolism and regional brain grey matter in neuroleptic-naive ultra-high-risk subjects and first-episode schizophrenia. Eur Neuropsychopharmacol. 2015;25(10):1661-1668. doi: 10.1016/j.euroneuro.2015.05.005 [DOI] [PubMed] [Google Scholar]
  • 19.Stone JM, Day F, Tsagaraki H, et al. ; OASIS . Glutamate dysfunction in people with prodromal symptoms of psychosis: relationship to gray matter volume. Biol Psychiatry. 2009;66(6):533-539. doi: 10.1016/j.biopsych.2009.05.006 [DOI] [PubMed] [Google Scholar]
  • 20.Wood SJ, Kennedy D, Phillips LJ, et al. Hippocampal pathology in individuals at ultra-high risk for psychosis: a multi-modal magnetic resonance study. Neuroimage. 2010;52(1):62-68. doi: 10.1016/j.neuroimage.2010.04.012 [DOI] [PubMed] [Google Scholar]
  • 21.de la Fuente-Sandoval C, León-Ortiz P, Azcárraga M, Favila R, Stephano S, Graff-Guerrero A. Striatal glutamate and the conversion to psychosis: a prospective 1H-MRS imaging study. Int J Neuropsychopharmacol. 2013;16(2):471-475. doi: 10.1017/S1461145712000314 [DOI] [PubMed] [Google Scholar]
  • 22.Allen P, Chaddock CA, Egerton A, et al. Functional outcome in people at high risk for psychosis predicted by thalamic glutamate levels and prefronto-striatal activation. Schizophr Bull. 2015;41(2):429-439. doi: 10.1093/schbul/sbu115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Egerton A, Stone JM, Chaddock CA, et al. Relationship between brain glutamate levels and clinical outcome in individuals at ultra high risk of psychosis. Neuropsychopharmacology. 2014;39(12):2891-2899. doi: 10.1038/npp.2014.143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Fusar-Poli P, Byrne M, Badger S, Valmaggia LR, McGuire PK. Outreach and support in south London (OASIS), 2001-2011: ten years of early diagnosis and treatment for young individuals at high clinical risk for psychosis. Eur Psychiatry. 2013;28(5):315-326. doi: 10.1016/j.eurpsy.2012.08.002 [DOI] [PubMed] [Google Scholar]
  • 25.Yung AR, Phillips LJ, McGorry PD, et al. Prediction of psychosis. A step towards indicated prevention of schizophrenia. Br J Psychiatry Suppl. 1998;172(33):14-20. doi: 10.1192/S0007125000297602 [DOI] [PubMed] [Google Scholar]
  • 26.Hall RC. Global assessment of functioning: a modified scale. Psychosomatics. 1995;36(3):267-275. doi: 10.1016/S0033-3182(95)71666-8 [DOI] [PubMed] [Google Scholar]
  • 27.Hamilton M. The assessment of anxiety states by rating. Br J Med Psychol. 1959;32(1):50-55. doi: 10.1111/j.2044-8341.1959.tb00467.x [DOI] [PubMed] [Google Scholar]
  • 28.Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56-62. doi: 10.1136/jnnp.23.1.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Nelson HE, O’Connell A. Dementia: the estimation of premorbid intelligence levels using the New Adult Reading Test. Cortex. 1978;14(2):234-244. doi: 10.1016/S0010-9452(78)80049-5 [DOI] [PubMed] [Google Scholar]
  • 30.Annett M. A classification of hand preference by association analysis. Br J Psychol. 1970;61(3):303-321. doi: 10.1111/j.2044-8295.1970.tb01248.x [DOI] [PubMed] [Google Scholar]
  • 31.Provencher SW. Estimation of metabolite concentrations from localized in vivo proton NMR spectra. Magn Reson Med. 1993;30(6):672-679. doi: 10.1002/mrm.1910300604 [DOI] [PubMed] [Google Scholar]
  • 32.LCModel manual. (http://s-provencher.com/pub/LCModel/manual/manual.pdf). Accessed September 23, 2018.
  • 33.Kaiser LG, Schuff N, Cashdollar N, Weiner MW. Age-related glutamate and glutamine concentration changes in normal human brain: 1H MR spectroscopy study at 4 T. Neurobiol Aging. 2005;26(5):665-672. doi: 10.1016/j.neurobiolaging.2004.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Gallinat J, Lang UE, Jacobsen LK, et al. Abnormal hippocampal neurochemistry in smokers: evidence from proton magnetic resonance spectroscopy at 3 T. J Clin Psychopharmacol. 2007;27(1):80-84. doi: 10.1097/JCP.0b013e31802dffde [DOI] [PubMed] [Google Scholar]
  • 35.Howes OD, Bose SK, Turkheimer F, et al. Dopamine synthesis capacity before onset of psychosis: a prospective [18F]-DOPA PET imaging study. Am J Psychiatry. 2011;168(12):1311-1317. doi: 10.1176/appi.ajp.2011.11010160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.de la Fuente-Sandoval C, León-Ortiz P, Azcárraga M, et al. Glutamate levels in the associative striatum before and after 4 weeks of antipsychotic treatment in first-episode psychosis: a longitudinal proton magnetic resonance spectroscopy study. JAMA Psychiatry. 2013;70(10):1057-1066. doi: 10.1001/jamapsychiatry.2013.289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Plitman E, de la Fuente-Sandoval C, Reyes-Madrigal F, et al. Elevated myo-inositol, choline, and glutamate levels in the associative striatum of antipsychotic-naive patients with first-episode psychosis: a proton magnetic resonance spectroscopy study with implications for glial dysfunction. Schizophr Bull. 2016;42(2):415-424. doi: 10.1093/schbul/sbv118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Chang L, Munsaka SM, Kraft-Terry S, Ernst T. Magnetic resonance spectroscopy to assess neuroinflammation and neuropathic pain. J Neuroimmune Pharmacol. 2013;8(3):576-593. doi: 10.1007/s11481-013-9460-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.van Berckel BN, Bossong MG, Boellaard R, et al. Microglia activation in recent-onset schizophrenia: a quantitative (R)-[11C]PK11195 positron emission tomography study. Biol Psychiatry. 2008;64(9):820-822. doi: 10.1016/j.biopsych.2008.04.025 [DOI] [PubMed] [Google Scholar]
  • 40.Doorduin J, de Vries EF, Willemsen AT, de Groot JC, Dierckx RA, Klein HC. Neuroinflammation in schizophrenia-related psychosis: a PET study. J Nucl Med. 2009;50(11):1801-1807. doi: 10.2967/jnumed.109.066647 [DOI] [PubMed] [Google Scholar]
  • 41.Bloomfield PS, Selvaraj S, Veronese M, et al. Microglial activity in people at ultra high risk of psychosis and in schizophrenia: an [(11)C]PBR28 PET brain imaging study. Am J Psychiatry. 2016;173(1):44-52. doi: 10.1176/appi.ajp.2015.14101358 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Maier M, Ron MA, Barker GJ, Tofts PS. Proton magnetic resonance spectroscopy: an in vivo method of estimating hippocampal neuronal depletion in schizophrenia. Psychol Med. 1995;25(6):1201-1209. doi: 10.1017/S0033291700033171 [DOI] [PubMed] [Google Scholar]
  • 43.Lutkenhoff ES, van Erp TG, Thomas MA, et al. Proton MRS in twin pairs discordant for schizophrenia. Mol Psychiatry. 2010;15(3):308-318. doi: 10.1038/mp.2008.87 [DOI] [PubMed] [Google Scholar]
  • 44.Hutcheson NL, Reid MA, White DM, et al. Multimodal analysis of the hippocampus in schizophrenia using proton magnetic resonance spectroscopy and functional magnetic resonance imaging. Schizophr Res. 2012;140(1-3):136-142. doi: 10.1016/j.schres.2012.06.039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Meyer EJ, Kirov II, Tal A, et al. Metabolic abnormalities in the hippocampus of patients with schizophrenia: a 3D multivoxel MR spectroscopic imaging study at 3T. AJNR Am J Neuroradiol. 2016;37(12):2273-2279. doi: 10.3174/ajnr.A4886 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Marsman A, van den Heuvel MP, Klomp DW, Kahn RS, Luijten PR, Hulshoff Pol HE. Glutamate in schizophrenia: a focused review and meta-analysis of 1H-MRS studies. Schizophr Bull. 2013;39(1):120-129. doi: 10.1093/schbul/sbr069 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Egerton A, Broberg BV, Van Haren N, et al. Response to initial antipsychotic treatment in first episode psychosis is related to anterior cingulate glutamate levels: a multicentre 1H-MRS study (OPTiMiSE). Mol Psychiatry. 2018. Epub ahead of print. doi: 10.1038/s41380-018-0082-9 [DOI] [PubMed] [Google Scholar]
  • 48.Fusar-Poli P, Rutigliano G, Stahl D, et al. Long-term validity of the At Risk Mental State (ARMS) for predicting psychotic and non-psychotic mental disorders. Eur Psychiatry. 2017;42:49-54. doi: 10.1016/j.eurpsy.2016.11.010 [DOI] [PubMed] [Google Scholar]
  • 49.Jelen LA, King S, Mullins PG, Stone JM. Beyond static measures: a review of functional magnetic resonance spectroscopy and its potential to investigate dynamic glutamatergic abnormalities in schizophrenia. J Psychopharmacol. 2018;32(5):497-508. doi: 10.1177/0269881117747579 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eMethods. 1H-MRS Data Processing

eFigure 1. Hippocampal Glutamate Concentrations

eFigure 2. Hippocampal Myo-inositol Concentrations

eFigure 3. Hippocampal Creatine Concentrations

eTable 1. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls vs Clinical High-Risk Subjects

eTable 2. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T)

eTable 3. Scan Quality Parameters and Voxel Tissue Composition: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

eTable 4. Participant Demographic, Clinical, and Medication Data at Baseline: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T)

eTable 5. Participant Demographic, Clinical, and Medication Data at Baseline: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

eTable 6. Hippocampal Metabolite Concentrations and Transition to Psychosis: Healthy Controls (HC) vs CHR Non-Transition (CHR-NT) vs CHR Transition (CHR-T).

eTable 7. Hippocampal Metabolite Concentrations and Functional Outcome: Healthy Controls (HC) vs Good Functional Outcome (CHR-GO) vs Poor Functional Outcome (CHR-PO)

Click here for additional data file. (668.2KB, pdf)

Articles from JAMA Psychiatry are provided here courtesy of American Medical Association

RESOURCES