SEARCH

SEARCH BY CITATION

Keywords:

  • 1H-MRS;
  • imaging;
  • proton spectroscopy;
  • psychosis;
  • schizophrenia;
  • spectroscopy

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

By reviewing the existing 1H-magnetic resonance spectroscopy literature in schizophrenia, the relationship of different sample characteristics and applied methodologies with metabolite alterations is explored. Furthermore, we emphasize common pitfalls and discrepancies in the methodological framework of the reviewed studies that introduce unwanted variation in findings and complicate the comparison of studies. A total of 92 studies were reviewed. Articles were retrieved by searching the Pubmed database. Care was taken to note down reliability and validity measures of each included study. Despite many methodological differences and shortcomings, progressive NAA reductions could be seen in several brain regions implicated in the pathogenesis of schizophrenia. In terms of treatment effects, cross-sectional evidence implicates a normalizing role for atypical antipsychotic medication; however, longitudinal studies remain inconclusive on this issue. Choline, creatine, and myo-inositol levels remain largely unchanged and a time-dependent role of glutamate finds confirmation in several spectroscopy studies. Other findings are less consistent and need further replication. Most studies lack power and methodological precision. Future studies should aim for standardization and for more distinguished study populations to gain more valid and reliable findings.

Abbreviations used
1H-MRS

proton magnetic resonance spectroscopy

ACG

anterior cingulate gyrus

ARMS

At-risk mental state subjects

AV

absolute metabolite values

CC

cerebellar cortex

Cho

choline

COV

coefficient of variation

Cr

creatine

CRLBV

cramér rao lower bounds

CSP

chronic schizophrenia patients

DLPFC

dorsolateral prefrontal cortex

DO

disease Onset

DUI

duration of untreated illness (months)

DUP

duration of untreated psychosis

FEP

first-episode patients

FWHM

full widths at half maximum

GABA

gamma-aminobutyric acid

Gln

glutamine

Glu

glutamate

Glx

glutamate and glutamine

GM

gray matter

HRS

subjects with a high genetic risk of developing schizophrenia

ID

illness duration in months

mI

myoinositol

mPFC

medial prefrontal cortex

mTL

medial temporal lobes

NAAG

Nacetylaspartylglutamate

NAA

N-acetyl-aspartate

OL

occipital lobes

PANSS

the Positive and Negative Syndrome Scale

PFC

prefrontal cortex

PRESS

point resolved spectroscopy

PVC

partial volume correction

ROI

region of interest

SD

standard deviation

SNR

signal-to-noise ratio

SP

schizophrenia patients

STEAM

stimulated Echo Acquisition Mode

TE

echo time

TL

temporal lobes

T

tesla

WCST

Wisconsin Card Sorting Test

WM

white matter

Schizophrenia is a chronic mental disorder characterized by disturbances in perception, thought, emotion, cognition, and behavior. Over the course of development, interactions between genes and environment may affect neural systems and give rise to several clinical symptoms. The pathology in schizophrenia includes abnormalities in neurotransmission, brain structure, and function. Proton magnetic resonance spectroscopy (1H-MRS) is a non-invasive neuroimaging technique used for measurement of regional brain metabolites that possibly reflect the status of neuronal and glial functions. In schizophrenia an increasing number of 1H-MRS studies have been conducted suggesting abnormal neurometabolism. However, many 1H-MRS studies yielded ambivalent results. 1H-MRS is a relatively new technique and there are many drawbacks with respect to the validity and reliability of 1H-MRS measurements.

The main goal of this literature review was to understand the variation of 1H-MRS findings in schizophrenia. We highlight the differential effects of age, gender, schizophrenia subtype (e.g. paranoid type), stage of disease progression (e.g. first episode patients) and medication status on neurometabolite concentrations measured by 1H-MRS in schizophrenia. Another aim was to trace inconsistent findings of studies back to methodological differences. To allow a better understanding of the reviewed issues and findings, a short summary of the technique and measured metabolites, as a comprised view of major neurophysiologic abnormalities in schizophrenia will be provided in the following.

Magnetic resonance spectroscopy

1H-MRS is a magnetic resonance tool that specifically enables non-invasive in vivo differentiation and quantification of small chemical compounds based on different resonance frequencies (Dager et al. 2008). The proton (1H), phosphorus (31P), and carbon (13C) nuclei are the most often used atomic nuclei in MRS studies given their magnetic properties. 1H-MRS is the most frequently applied measurement as the 1H nucleus forms the most sensitive (after the unstable and radioactive 3H isotope), and most abundant, naturally-occurring nucleus (Gillies 1992; Rothman et al. 1999). The higher sensitivity of 1H-MRS makes it better suited for examining specific small brain regions, as opposed to 31P-MRS which necessitates larger voxel and longer scan times for adequate signal to noise ratio (SNR) in regions of similar sizes (Stanley et al. 2000).

The spectroscopy signal is acquired by using either ‘single voxel spectroscopy’ (SVS) or by using ‘multi voxel spectroscopic imaging’ (MRSI). The former measures resonances exclusively in a predetermined region (voxel) and has the advantages of a more homogenous magnetic field and thus better spectral resolution per voxel despite a decreased spatial resolution. MRSI, on the other hand, is able to measure different spectra in parallel and gives a better SNR ratio for multiple voxels of interest (VOIs) since the signal from each voxel is averaged for the total data collection (Drost et al. 2002). As voxel sizes are generally smaller in MRSI than in SVS, partial volume effects (i.e. combining two or more tissue types within a single voxel) have less impact on the assessment of metabolites (Stanley et al. 2000). Yet, the spatial resolution of MRSI is still quite low as compared to conventional MRI techniques with typical voxel size around 1–1.5 cc, and chemicals must exist in concentrations higher than 100 μM to be measured (Novotny et al. 2003). In sum, SVS is superior for a detailed analysis of specific regions, whereas MRSI is preferentially used when the VOI is of unknown origin and when there is a need for a quick assessment of several regions (Drost et al. 2002; Hammen and Stefan 2004; Morche 2005).

The most commonly applied localization method for sequence acquisition is ‘Point Resolved Spectroscopy’ (PRESS) and ‘Stimulated Echo Acquisition Mode’ (STEAM). With STEAM a shorter echo time (TE) can be achieved and the localization is more precise, yet PRESS has an intrinsic higher signal to noise ratio (Drost et al. 2002). The external magnetic field strength determines the location of different resonance signals, hence differences in tesla (T) result in different resonances. To standardize between different magnetic field strengths, the 1H-MRS signal is transformed to a frequency spectrum. The position of the signal peaks on the x-axis is expressed as ‘chemical shifts’ (shift in resonance frequency that is unique to a given molecule) in units of parts per million (ppm) and therefore called the ppm-axis. Consequently, inter- and intra-individual comparisons can be made (Wobrock et al. 2005; Dager et al. 2008). The measurement units can be expressed as ratios, absolute values (AVs) or arbitrary units.

Higher field strengths and short TEs (< 30 ms) heighten the SNR and narrow peak widths on the ppm-axis, resulting in improved spectral resolution and sensitivity, thereby enabling the detection of overlapping metabolites (Walter 2005). After selection of the VOI the homogeneity of the field is adjusted and optimized by a process called ‘shimming’ [through adjusting direct currents in the gradient coils and 20–25°C shim coils] to maximize the field homogeneity and thus the spectral resolution (Stanley et al. 2000). The much higher intensity of water (because of its abundance) and lipid molecules (because of their hydrogen content) as compared to the measured metabolites disturbs the 1H-MR-signal and must be suppressed (Frahm et al. 1989; Walter 2005; Wobrock et al. 2005). Consequently, water suppression techniques are used that apply frequency specific radio frequency pulses that convert and eliminate disturbing signals (Bertolino and Weinberger 1999). Lipid suppression is only used when there is a need to counteract outer-volume lipid contaminations. Long TEs decrease signals from lipid and macromolecules and thus enhance the validity of metabolite quantification, whereas short TEs are able to differentiate metabolites undetected by long TEs (Drost et al. 2002). Mostly, long TEs are used together with MRSI and short TEs together with SVS, and the former when a higher spatial resolution is required whereas the latter serves best when metabolites with overlapping peaks on the ppm-axis need to be analyzed (Stanley et al. 2000).

Given that metabolites are differentially distributed in cerebrospinal fluid (CSF), white matter (WM) and gray matter (GM), an important factor when assessing metabolites, is to correct for differences in tissue composition of the VOI. Otherwise concentrations might suffer from over and underestimations and thereby lead to so-called partial volume effects.

The general quality of 1H-MRS data depends consequently on the magnetic field strength, the size and composition of the brain region of interest, the acquisition time and the concentration of metabolites (Auer et al. 2001; Drost et al. 2002; Kreis 2004; Wobrock et al. 2005).

Major 1H-MRS metabolites: regional variability and other considerations

Neurometabolites measured by 1H-MRS include N-acetyl-aspartate (NAA), creatine (Cr), choline (Cho), myo-Inositol (mI), lactate (Lac), glutamate (Glu), glutamine (Gln), and Gamma-amino butyric acid (GABA). Glx describes the combined spectra of Glu and Gln, because of their overlapping peaks on the ppm-axis, which appear between 2.1 and 2.5 ppm and between 3.72 and 3.82 ppm (Kreis et al. 1993; Bertolino and Weinberger 1999; Dager et al. 2008). After being released by neurons, Glu is taken up into glial cells and rapidly converted to Gln, which is subsequently transported back to neurons and reconverted to Glu (Watts et al. 2005). The three multiplets resonances (2.31 ppm, 1.9 ppm, and 3.0 ppm) of the main inhibitory neurotransmitter of the brain – GABA – overlap with the Glx, NAA, and Cr spectra. Although careful post-processing may help to differentiate GABA, Glu, and Gln, often high field strengths (> 4.7 T) and short TEs together with sophisticated editing techniques are required to obtain valid results for GABA spectra (Govindaraju et al. 2000; Drost et al. 2002; Novotny et al. 2003; Hammen and Stefan 2004; van Elst et al. 2005; Bogner et al. 2010).

The peaks of total Cr (tCr) at 3.0 and 3.9 ppm comprise phosphocreatine and Cr, which can hardly be separated even when using magnetic resonance devices with magnetic field strengths higher than 1.5 T. tCr is found to be relatively constant in the general population yet higher concentrations were seen in the cerebellum and in regional GM (Pouwels and Frahm 1998; Drost et al. 2002; Yucel et al. 2007; Brief et al. 2009; Ongur et al. 2009). Despite these variations in tCr concentration, and despite the fact that tCr has been found to differ in schizophrenia (Ongur et al. 2009) it is still often taken as internal reference for 1H-MRS metabolite quantification.

Although there are still discussions on the exact role of NAA (Clark 1998; Barker 2001), it is found mainly around mature neuronal cell bodies, axons, and dendrites within the central nervous system, not in glia (Simmons et al. 1991; Urenjak et al. 1993) and with highest concentrations in pyramidal glutamatergic neurons (Moffett and Namboodiri 1995). NAA contributes 75–85% to the total NAA (tNAA) signal with a main peak at 2.01 ppm and a smaller peak around 2.6 ppm, which is only observed with short TE. The other 15–25% of the tNAA signal is accounted for by N-acetylaspartylglutamate, with highest concentrations in parietal and occipital WM (Pouwels and Frahm 1998; Hammen and Stefan 2004).

Cho's peak at 3.22 ppm is comprised of various signals including Cho, phosphocholine, and glycerophosphocholine. The latter two have been ascribed the most important contribution to the signal. The signal is assumed to be predominantly related to phospholipid membrane turnover and metabolism, as seen in myelination or inflammation and related diseases (Bertolino and Weinberger 1999; Hammen and Stefan 2004).

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

Relevant articles were primarily identified via Pubmed search. All studies included only human subjects that underwent 1H-MRS, with resulting units of ratios and arbitrary values. Scientific articles written in English and Spanish that were published between 2000 and 2012 were included. Three studies (Stanley et al. 1996; Callicott et al. 1998; Bartha et al. 1999) were included although they were published slightly before that time, as they comprised a large sample and a big array of assessed structures (Callicott et al. 1998) and because drug-naïve recent-onset psychosis patients were included (Stanley et al. 1996; Bartha et al. 1999).

The search resulted in 92 articles, including three articles that lacked a healthy control group: Braus et al. (2002) compared atypically treated versus typically treated patients with schizophrenia, assessed the effect of treatment with an omega-3 fatty acid in first-episode psychosis and another study compared the effects of anti-psychotic drug treatment on NAA levels (Bertolino et al. 2001). Six articles included psychotic patients other than schizophrenia patients (Berger et al. 2008; Wood et al. 2008, 2009; Tunc-Skarka et al. 200, Block et al. 2000; Fannon et al. 2003), three articles used the same control group (van Elst et al. 2005; Olbrich et al. 2008; Rusch et al. 2008), and some authors included the same patients in distinct studies (Berger et al. 2008; Wood et al. 2008, 2009).

The reliability and validity of the reviewed studies were considered and noted in Tables 1–12. Parameters of interest were amongst others the use of covariates if the sample was not well matched (e.g. age or gender), the execution of partial volume corrections (PVC), and the use of reliability estimates. For the latter, the full width at half maximum (FWHM, showing the distribution spread), SNR, and the Crámer-Rao lower bound of variance (CRLBV, defines a minimum value to the measurement variance) were considered.

Table 1. 1H-MRS studies of the temporal lobes in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, at-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Basoglu et al. (2006)Right ThalamusNAA/Cr[DOWNWARDS ARROW] (FE/CSP vs. control participants; FEP vs. control participants)10 FEP, male, unmedicated, schizophreniform disorder, age: 21.9(2.5), ID: 15.7(4.9) days1.5T, PRESS, long TE,SVSOnly males but covariateIllness days in FEP significantly correlated with TL NAA/Cr & NAA/Cho 
 Right TLNAA/Cr [DOWNWARDS ARROW] (CSP/FEP vs. control participants)15 male CSP, medicated, age: 40.1 (11.2), ID: 196.8(122.7)Cr/Cho ratiosNo reliability estimatesNo differences between CSP & FEP in metabolites 
  Cho10 control participants, age: 30.9 (7.2) No PVC No definite diagnosis of schizophrenia in FE patients
Berger et al. (2008)Bilateral medial TL (anterior hippocampus 50%)Bilateral GSH[UPWARDS ARROW], left Glx[UPWARDS ARROW] (those treated with E-EPA)12 medicated FEP, treated with 2 g/day oral E-EPA for two weeks, age: 19.6 (2.9)3T, short TE, PRESS, SVSUnequal gender distribution, but covariate used Placebo group consisted of only males
  (after treatment)12 medicated FEP, treated with 2 g placebo oil for two weeks, age: 21.4 (4.1)AV & PCr ratios

CRLBV > 30%

FWHM = 0.093 SNR = 11.3(1.9)

Negative symptom improvement correlated with increases in GSH, TMA, & Cr/PCr

Inclusion of affective psychotic disorder

patients, with a higher percentage of them in the placebo group

  NAA, TMA, mI, GLX, GSH  PVCTightly coupled increase in GSH & GLXNo control group
Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFC

NAA/Cr[DOWNWARDS ARROW]

NAA/Cho[DOWNWARDS ARROW]

24 mainly medicated schizophreniform disorder patients, age: 23.7 (6.1)1.5T, MRSI, long TE,Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performance

Schizophrenia

diagnosis in 17 patients afterwards

 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the Positive & Negative Syndrome Scale 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW](while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks on medication), age: 36.9 (8.1), ID: 128.4 (70.8)1.5T, MRSI, long TEUnequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Callicott et al. (2000)DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)1.5T, MRSI, long TE,Unequal gender distributionInverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/post. cingulate, superior temporal gyrus

Cho/Cr

NAA/Cho

73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, Thalamus, OFC   No PVC  
Callicott et al. (1998)Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2(8.8)1.5T, MRSI, long TENo gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WMNAA/Cho60 unaffected siblings, age: 34.6(8.6)Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Chang et al. (2007)Bilateral mTL WM

NAA[DOWNWARDS ARROW]

mI[DOWNWARDS ARROW]

Glx[UPWARDS ARROW]

23 elderly medicated schizophrenia patients, with cognitive & functional decline, age: 66.3 (7.2), ID: 517.2 (64.8)4T, short TE, SVS, short TEUnequal gender distribution, no covariate usedSchizophrenia participants with severe mental decline had the lowest NAA concentrations in FL/TL;Low Glx control participants values
 PF WM, OL WM

Cr

Cho

22 control participants, age: 70.0 (5.3)AVCRLBV FWHMSP had age-related decline in Cr & Cho in the right FL 
     No PVCNo difference between schizophrenia patients treated with atypical versus typical & atypical anti-psychotics 
Fukuzako (2000)Left mTL (hippocampal formation, entorhinal cortex, & amygdala)NAA/Cr[DOWNWARDS ARROW]64 medicated CSP, age: 36.5 (7.2), ID: 178.8 (33.6)2T, STEAM, long TE, SVSEqual gender distribution  
  Cho/Cr51 control participants, age: 35.5 (7.9)Cr ratiosWater line widths<4 HzInverse correlation of NAA/Cr ratio with age & ID 
     No PVC  
Galinska et al. (2009)

Left FL

left TL

NAA

Cho

mI

Glx

30 FEP, medicated (26 paranoid subtype) age: 22.5(3.6), DO: 21.8 (3.3)1.5T, PRESS, SVS, short TEGender covariateNo differences were found between the groups of patients with short & long DUP & control participants 
 Left thalamus 19 control participants, age: 22.5(3.3)Cr & H2O ratiosNo reliability estimates  
     No PVC  
Jessen et al. (2006)Left FL (both vs. control participants)NAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]10 early ARMS, medicated, age: 27.0 (6.8)SVS, 1.5T, PRESS, long TEGender covariateARS that converted to SP had a higher Cho/Cr & lower NAA/Cho ratios in ACG compared with non-convertersAll control values higher than in reference literature
 ACG (both vs. control participants)NAA/Cr[DOWNWARDS ARROW]9 late ARMS, medicated, age: 28.7 (7.0)Cr/Cho ratiosNo reliability estimatesNo difference between unmedicated & medicated ARSP 
 Left superior TLCho/Cr21 medicated SP, age: 33.4 (7.2) No PVC  
   31 control participants, age: 34.8 (13.5)    
Kegeles et al. (2000)Left mTL (hippocampus)NAA/Cho, Glx (Glu/Gln/GABA)/Cho10 CSP (7 unmedicated), age: 28(7), ID: 96 (60)1.5T, SVS, STEAM, short TEOnly maleshippocampus volume reduction not significantLow test-retest reliability for the Glx measurements
 Right mTL (hippocampus)NAA/Cr, Cho/Cr, Glx/Cr10 control participants, age: 29 (5)Cho & Cr ratios5 retests resulting in COV of 36-44%Trend for right-sided Glx/Cho laterality index excess in patients  
     PVCNAA/Cr was lower in the patients although non-significant 
Moore et al. (2002)Left TLCho/Cr[DOWNWARDS ARROW] versus rTL Cho/Cr in both groups20 CSP, age: 33.8(7.1), ID: 168.96 (88.8)1.5T, PRESS, long TE, SVSUnequal gender distribution, but matched Correlation between left GM & WM content & Cho/Cr ratios in schizophrenia subjects  
 Right TL 20 control participants, age: 28 (8.3)Cr ratiosNo reliability estimatesThe Cho/Cr ratio in the lTL of CSP was lower than normal control participants but non-significant 
Shimizu et al. (2007)PCGNAA/Cr[DOWNWARDS ARROW] (PCG)19 CSP, medicated, age: 40.4 (13.1), ID: 195.6 (142.8)1.5T, PRESS, SVS, long TE, Equal gender distribution Age-related decline of NAA/Cr in PCG of control participants 
 Bilateral mTLCho/Cr18 control participants, age: 34.9 (11.4)Cr ratiosNo reliability estimates  
     No PVC  
Szulc et al. (2011)ThalamusGlx/Cr 42 CSP assessed before & after medication; age: 32.2(6.0); ID: 110.4 (68.4)1.5T, long TE, SVS, PRESSUnequal gender distribution but matchedGlobal NAA/Cr lower in patientsNon-random medicational switch
 FLNAA/Cr[DOWNWARDS ARROW] (thalamus & FL)11 control participants, age: 37 (10)Cr RatiosNo reliability estimatesGlx/Cr decreased in the temporal lobe after treatment 
 TL Cr  No PVC  
Szulc et al. (2007)

Left FL

left TL

 58 CSP split based on treatment (typ/atyp), age: 33.1 (6.9), ID: 112.92(70.8) 1.5T, SVS, short TE, PRESSGender covariateNegative correlation between NAA ratios & age in the FL & between ID & hospitalizations in TL  
 Left thalamus NAA/Cr[DOWNWARDS ARROW](CSP typical)control participants, age: 30.2 (5.3)Cr & H2O ratiosNo reliability estimates  
     No PVC  
Szulc et al. (2004)Left thalamus Glx/Cr[UPWARDS ARROW] (FEP vs. control participants)31 FEP, medicated for 8 weeks, age 22.55 (3.50) 1.5T, PRESS, short TE, SVSNo gender covariate Positive correlation of negative symptomatology & Glx levles 
 Left PFC, left TLNAA, Cr, Cho, Glx/H2O17 CSP, medicated, age: 33.59 (7.40)Cr & H2O ratiosNo reliability estimates  
   13 control participants No PVC  
Tang et al. (2007)mTL WM NAA/Cr[DOWNWARDS ARROW]42 medicated CSP, age: 38.69 (11.42), DO: 23.53 (7.09)3T; PRESSNo gender covariateReduced anisotropy in mTL. NAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
 DLPF WM, OL WM  42 control participants, age: 43.3 (20.18)Short TE, MRSI No reliability estimatesNAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
    Cr ratiosNo PVC  
Weber-Fahr et al. (2002)Bilateral hippocampus (mainly GM)NAA[DOWNWARDS ARROW]15 CSP, age: 31.8 (7.1), ID: 131.6 (87)1.5T, 2D-MRSI, long TE, PRESSUnequal gender distribution, no covariateOnly after PVC significantly lower NAA levels were seen in CSP 
 TL WMCr, Cho15 control participants, age: 32.6 (10.7)AVNo reliability estimates  
 CSF from third ventricle   PVC  
Wood et al. (2009)Bilateral medial TL (50% anterior hippocampus)22% [UPWARDS ARROW] in GSH30 FEP: 13 drug-naïve & 17 treated with atypical medication, age: 19.4 (3.4)3T, PRESS, SVS, short TE70% males, no control groupFEP non-responsive to topical niacin showed 35% increased GSH as compared to normal responders 
   18 control participants, age: 20.6 (4.2)AVCRLBV SNR FWHM  Three major depressed FEP with psychotic symptoms
Wood et al. (2008)Bilateral medial TL (50% anterior hippocampus)NAA, Glx, Cho15 drug-naïve FEP (7 rescanned after 12 w of treatment), age: 20.0 (4.0), DO: 19.9 (4.0) 3T, PRESS, SVS, short TE, Unequal gender distribution, only FEP matched with control participants When only considering the schizophreniform spectrum patients, Cr & mI were found elevatedSix major depressed subjects with psychotic symptoms
  

Cr

mI

19 medicated schizophrenia spectrum patients (SSP), age: 19.5 (3.2), DO: 19.4 (3.2) AVNo reliability estimates  
   19 control participants, age: 21.0 (4.4) PVC  
Table 2. 1H-MRS studies of the hippocampus in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, at-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time; WCST, Wisconsin Card Sorting Test.

Berger et al. (2008)Bilateral medial TL (anterior hippocampus 50%)bilateral GSH[UPWARDS ARROW], left Glx[UPWARDS ARROW] (those treated with E-EPA)12 medicated FEP, treated with 2 g/day oral E-EPA for 2 weeks, age: 19.6 (2.9)3T, short TE, PRESS, SVSUnequal gender distribution, but covariate used Placebo group consisted of only males
   12 medicated FEP, treated with placebo oil for 2 weeks, age: 21.4 (4.1)AV & PCr ratios

CRLBV > 30%

FWHM = 0.093 SNR = 11.3(1.9)

Negative symptom improvement correlated with increases in GSH, TMA, & Cr/PCrInclusion of affective psychotic disorder patients, with a higher percentage of them in the placebo group
  NAA, TMA, mI, GLX, GSH  PVCTightly coupled increase in GSH & GLXNo control group
Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFCNAA/Cr[DOWNWARDS ARROW], NAA/Cho[DOWNWARDS ARROW]24 mainly medicated schizophreniform disorder Patients, age: 23.7 (6.1), ID:1.5T, MRSI, long TE, Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients post assessment
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior & posterior cingulate, PF WM, centrum semiovale, putamen, & thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the Positive & Negative Syndrome Scale 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8) 1.5T, MRSI, long TEUnequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Bloemen et al. (2011)Left hippocampusGlu[DOWNWARDS ARROW]11 subjects at ultra high risk of psychosis3T, SVS, PRESS, short TEAge & sex matchedGlu related with striatal pre-synaptic dopaminergic neurotransmission  
   11 control subjectsAVSNR < 20%   
     No PVC  
Callicott et al. (2000) DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)1.5T, MRSI, long TE,Unequal gender distribution Inverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/posterior cingulate, superior temporal gyrusCho/Cr, NAA/Cho73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, thalamus, OFC   No PVC  
Callicott et al. (1998) Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2 (8.8)1.5T, MRSI, long TENo gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WMNAA/Cho60 unaffected siblings, age: 34.6(8.6) Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Fannon et al. (2003)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (in drug-naïve FEP at baseline vs. medicated FEP & control participants)12 drug-naïve FEP, age: 26.1 (5.5) ID: 22.02 (15.41) weeks. Reassessed after treatmentPRESS, short TE, SVSUnequal gender distribution, but covariate usedTreatment eliminated NAA difference3 of the 33 FEP had also a schizo-affective disorder
 Left FL, left BGCho/Cr21 medicated FEP, age: 24 (5.8), ID: 49.45 (33.19) weeksCr ratiosSNR but not specifiedBoth patient groups had reduced left hippocampal volume as compared to control participants 
   18 control participants, 25.3 (6.7) No PVC  
Fukuzako (2000)Left mTL (hippocampal formation, entorhinal cortex, & amygdala)NAA/Cr[DOWNWARDS ARROW]64 medicated CSP, age: 36.5 (7.2), ID: 178.8 (33.6)2T, STEAM, long TE, SVSEqual gender distribution   
  Cho/Cr51 control participants, age: 35.5 (7.9) Cr ratiosWater line widths < 4 HzInverse correlation of NAA/Cr ratio with age & ID  
     No PVC  
He et al. (2012)WM of the FLNAA/Cr, mI/Cr, Cho/Cr63 drug-naïve FEP, age: 23.94 (8.48); ID: 8.38 (12.33)3T, short TEUnequal gender distribution, but matchedNegative correlation between positive symptoms & the NAA/Cho in the WM of the left FL 
 HippocampusNAA/Cho63 control participants, age: 23.97 (8.38)Cr & Cho ratiosFWHM < 10 Hz & SNR > 12 (Hippocampus) or SNR > 40 (FL WM)  
     No PVC  
Hasan et al. (2011)HippocampusNAA/Cr53 FEP mainly medicated, age 28.36 (7.82)1.5T, SVS, short TEUnequal gender & age distribution & unmatchedPositive correlation between reduced left hippocampal volume & poor verbal learning performance 
  Glx/Cr, mI/Cr, Cho/Cr52 control participants, age: 38.83 (12.06)Cr ratiosNo reliability estimates  
     No PVC  
Kegeles et al. (2000)Left mTL (hippocampus)NAA/Cho, Glx (Glu/Gln/GABA)/Cho10 CSP (7 unmedicated), age: 28(7), ID: 96(60)1.5T, SVS, STEAM, short TEOnly maleshippocampus volume reduction not significant Low test-retest reliability for the Glx measurements
 Right mTL (hippocampus)NAA/Cr, Cho/Cr, Glx/Cr10 control participants, age: 29 (5)Cho & Cr ratiosFive retests resulting in COV of 36–44%Glx/Cho laterality index showed a relative right-sided excess in patients but did not reach significance; 
     PVCNAA/Cr was lower in patients although non-significant 
Klar et al. (2010)HippocampusNAA29 medicated CSP, age: 27.6(6.8); ID: 45.6 (4.1)3T, short TE, PRESS, SVSUnequal gender distribution, but matchedNegative correlation between hippocampal NAA concentration & volume Validated delineation protocol for the hippocampus
  Cho, Cr44 control participants, age: 30.9 (7.9)AVNo reliability estimates  
     CSF correction  
Lutkenhoff et al. (2010)mPFC GMGlu[DOWNWARDS ARROW] (both)9 Schizophrenic twins, age: 48.8(11.5), ID: 328.8(133.2)3T, PRESS, short TE, SVSEqual gender distribution Effect of age: greater Glu levels in older subjectsControl group values not in range for mI snf Cho in left hippocampus & mPFC
 Left hippocampusNAA[UPWARDS ARROW], Cr[UPWARDS ARROW], Cho[UPWARDS ARROW]12 Unaffected Co-twins, age: 49.5(10.0) AVSNR > 3, CRLBV > 25%, FWHM > 0.1   
 Left PFC WMGlu, Gln, Cr, Cho, mI21 control participants twin pairs, age: 55.7(3.8) PVC  
Miyaoka et al. (2005)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (both)15 SP with Gilbert's syndrome (GS) & 15 without1.5T, PRESS, SVS, short TEEqual gender distribution Schizophrenia with GS more severely & affected (also in BG & vermis) 
 Left BGCho, mIGS/wGS, all medicated, age: 32.5 (10.7)/34(11), ID: 1.3 (1.8)/1.7 (2.1)Cr RatiosNo reliability estimates  
 Vermis 15 control participants, age: 41.7 (14.7) No PVC  
Olbrich et al. (2008)Left DLPFCGlu[UPWARDS ARROW]9 medicated FEP, age 28.4 (7.3), 2T, PRESS, SVS, short TE Frontal glutamate concentrations significantly correlated with rating scores for schizophreniform symptomsAlready used control group (van Elst et al. 2005) with decr. Glu/gln values, small sample size
 Left hippocampusNAA, Cho, Cr, Gln32 control participants, age 28.2 (5.8) COV < 75%  
 OL WM   No PVC  
Rothermundt et al. (2007)Left amygdalamI 12 mainly male & mainly drug-naïve paranoid schizophrenia patients in an acute episode, age: 25.33(4.75) 1.5T, SVS, short TE, STEAM10/1 but matched control participantsSchizophrenia patients with increased S100β serum concentration had higher mI levels 
 Anterior hippocampal regionSerum levels of S100β12 control participants, age: 25.33 (4.75)AVNo reliability estimatesNo difference between medicated & drug-naïve  
     PVC  
Rusch et al. (2008)Left DLPFC Glx[UPWARDS ARROW]29 recurrent & FEP, medicated, paranoid subtype schizophrenia patients; age: 27.8 (6.1), ID: 52.9 (47.1) 2T, short TE, SVS, PRESSUnequal gender distribution but matched Reduced amygdalar volume associated with impaired executive functioning in schizophrenia patients & control participantsGlu control values very low - see van Elst et al. (2005) and Olbrich et al. (2008)
 Left hippocampusNAA, Cr, Cho31 control participants, age: 28.2 (5.9)AVNo reliability estimates  
     No PVC  
Stone et al. (2009)ACGGln[UPWARDS ARROW]27 ARMS, age: 25 (5)3T, PRESS, short TE, SVSEqual gender distribution Glu in left thalamus was negatively correlated with GM volume in several structures 
 Left thalamusGlu[DOWNWARDS ARROW], NAA[DOWNWARDS ARROW]27 control participants, age: 25 (4)AV

CRLBV

SNR

FWHM

  
 Left hippocampusCr, Cho, mI  PVC   
Weber-Fahr et al. (2002)Bilateral hippocampus (mainly GM)NAA[DOWNWARDS ARROW]15 CSP, age: 31.8 (7.1), ID: 131.6 (87)1.5T, 2D-MRSI, long TE, PRESSUnequal gender distribution, no covariateOnly after PVC significantly lower NAA levels were seen in CSP 
 TL WMCr, Cho15 control participants, age: 32.6 (10.7)AVNo reliability estimates  
 CSF from third ventricle   PVC  
Wood et al. (2009)Bilateral medial TL (50% anterior hippocampus)22% [UPWARDS ARROW] in GSH30 FEP: 13 drug-naïve & 17 treated with atypical medication, age: 19.4 (3.4)3T, PRESS, SVS, short TE70% males, no control groupFEP non-responsive to topical niacin showed 35% increased GSH as compared to normal responders 
   18 control participants, age: 20.6 (4.2)AV

CRLBV

SNR

FWHM

 Three major depressed FEP with psychotic symptoms
Wood et al. (2008)Bilateral medial TL (50% anterior hippocampus)NAA, Glx, Cho, 15 drug-naïve FEP (7 rescanned after 12 weeks of treatment), age: 20.0 (4.0), DO: 19.9 (4.0) 3T, PRESS, SVS, short TE, Unequal gender distribution, only FEP matched with control participants When only considering the schizophreniform spectrum patients, Cr & mI were found elevatedSix major depressed subjects with psychotic symptoms
  Cr, mI19 medicated schizophrenia spectrum patients, age: 19.5 (3.2), DO: 19.4 (3.2) AVNo reliability estimates  
   19 control participants, age: 21.0 (4.4) PVC  
van Elst et al. (2005)Left DLPFC Gln[UPWARDS ARROW], Glu[UPWARDS ARROW]21 medicated CSP, only paranoid subtype, age: 28.5 (1.4), ID: 63.8(9.4) 2T, short TE, SVS, PRESSUnequal gender distribution, no gender covariate Prefrontal & hippocampal Glu concentrations inversely correlated with psychiatric disturbance of the last 2 yearsOverlapping control group with Rusch et al. (2008) and Olbrich et al. (2008)
 Left anterior hippocampusGlu[UPWARDS ARROW]32 control participants, age: 28.2 (1.0)AVCOV < 21% Glu, Gln, mI values of the control group far lower than reported in the literature
  NAA, Cr, Cho, mI  CSF correction Macromolecular contribution to the MRS signal not assessed
Table 3. 1H-MRS studies of the thalamus in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG=anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Aoyama et al. (2011)Left thalamusGln[DOWNWARDS ARROW]17 initially drug-naïve SP, followed up for 80 months after treatment, age: 25 (7)4T, STEAM, SVSEqual gender distributionGlu was reduced after 80 months in SP compared to control participants 
 Anterior cingulateTcr, NAA, mI17 control participants 1, age: 32(10); 17 control participants 2 age: 29 (10)H2O ratiosCOV < 75%  
  Taurine scylloinositol  PVC through water correction  
Auer et al. (2001)Medio-dorsal & lateral thalamic nucleiNAA[DOWNWARDS ARROW]32 acutely-ill, medicated CSP, age: 33.91.5T, PRESS, SVS, short TE20/12, covariateReduced Cr & Cho ratios in the left thalamusControl values of mI, & NAA under range in thalamus; Cr & NAA under range in WM
 Parietal WM

mI[UPWARDS ARROW]

Cr[UPWARDS ARROW]

Cho[UPWARDS ARROW]

17 control participants, age: 31.2Ratios & AVFWHM No distinction between thalamic tissue itself & the surrounding GM & WM
     PVC  
Basoglu et al. (2006)Right thalamusNAA/Cr[DOWNWARDS ARROW] (FEP/CSP vs. control participants; FEP vs. control participants10 FEP, male, unmedicated, schizophreniform disorder, age: 21.9(2.5), ID: 15.7(4.9) days

1.5T PRESS long TE

SVS

Only males but covariateIllness days in FEP significantly correlated with TL NAA/Cr & NAA/Cho 
 Right TLNAA/Cr[DOWNWARDS ARROW] (CSP/FEP vs. control participants)15 male CSP, medicated, age: 40.1(11.2), ID: 196.8 (122.7)Cr/Cho ratiosNo reliability estimatesNo differences between CSP & FEP in metabolites 
  Cho10 control participants, age: 30.9 (7.2) No PVC No definite diagnosis of schizophrenia in FE patients
Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFC

NAA/Cr[DOWNWARDS ARROW]

NAA/Cho[DOWNWARDS ARROW]

24 mainly medicated schizophreniform disorder patients, age: 23.7 (6.1)

1.5T

MRSI

long TE

Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients afterwards
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior & posterior cingulate, PF WM, centrum semiovale, putamen, & thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the PANSS 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Bustillo et al. (2008)Left caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9), ID: 41.4 (69)1.5T, PRESS, SVS, short TEGender covariateGlobal NAA reductions related to global cognitive performance in the whole subject sample 
 Left FL, left OLNAA[DOWNWARDS ARROW] (globally before treatment)21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

No difference after treatment 
 Right cerebellum

Cho

Glx

mI

  PVC  
Callicott et al. (2000)DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)

1.5T

MRSI

long TE

Unequal gender distributionInverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/post. cingulate, superior temporal gyrusCho/Cr, NAA/Cho73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, Thalamus, OFC   No PVC  
Callicott et al. (1998)Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2 (8.8)

1.5T

MRSI

long TE

No gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, prefrontal WMNAA/Cho60 unaffected siblings, age: 34.6 (8.6)Cr & Cho ratiosNo reliability estimates  
 anterior & posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Deicken et al. (2000)Bilateral mediodorsal thalamusNAA[DOWNWARDS ARROW]17 male medicated CSP, age: 35.7 (9.3), ID: 176.4 (127.2)

1.5T

MRSI

long TE

All malesPositive correlation between left & right NAA levels in CSPLow NAA values in control participants & CSP
  Cho, Cr10 male control participants, age: 39.5 (12.8)AVNo reliability estimates  
     No PVC but voxel determination on two occasions with good test-retest reliabilities  
de la Fuente-Sandoval et al. (2011)Dorsal-caudate

Glu[UPWARDS ARROW] (in caudate of both patient groups)

Glu/Gln (caudate FEP)

18 drug-naïve subjects with prodromal symptoms for schizophrenia, age: 19.56 (3.46)

3T

SVS

short TE

Equal gender distribution Two patient groups versus one control participants
 Cerebellum

NAA[UPWARDS ARROW]

(P)Cho[UPWARDS ARROW]

(in both patients & both regions)

18 drug-naïve FEP, age: 23.44 (4.93)AVFWHM  
   (P)Cr, mI40 control participants, age: 21.83(4.47) PVC  
de la Fuente-Sandoval et al. (2009)Right caudate nucleusGlu/Cr[UPWARDS ARROW] NAA/Cr[DOWNWARDS ARROW] (during acute psychosis & after)14 medicated FEP & CSP, age: 24.9 (5.9), DO: 21.2 (18. 6) months

3T

PRESS MRSI

Equal gender distributionMetabolite changes independent of treatment 
  Cho/Cr14 control participants, age: 26.6 (5.9)Cr ratiosNo reliability estimates  
     PVC  
Galinska et al. (2009)

Left FL

left TL

NAA

Cho

mI

Glx

30 FEP, medicated (26 paranoid subtype) age: 22.5(3.6), DO: 21.8 (3.3)

1.5T PRESS

SVS

short TE

gender covariateNo differences were found between the groups of patients with short & long DUP & control participants 
 Left thalamus 19 control participants, age: 22.5(3.3)Cr & H2O ratiosNo reliability estimates  
     No PVC  
Jakary et al. (2005)Mediodorsal & anterior thalamusNAA[DOWNWARDS ARROW]22 CSP, male, medicated, age: 34.5 (9.4), ID: 188.4(140.4)1.5T, long TE, MRSI,Only malesStrong negative correlation between left thalamic NAA & duration of illness 
  

Cr

Cho

22 control participants, age: 36.4 (11.3)AVFWHM  
     PVC  
Martinez-Granados et al. (2008)ThalamusBilateral NAA/Cho[DOWNWARDS ARROW](both)30 male medicated CSP with auditory hallucinations, age: 40 (9)

1.5T

PRESS long TE MRSI

Only malesCSP with auditory hallucinations more severely affected 
  Cho/Cr, NAA/Cr19 male, medicated CSP without auditory hallucinations, age: 41 (14)Cho & Cr ratiosNo reliability estimates  
   37 control participants, age: 33 (11) No PVC  
Omori et al. (2000)Left thalamusNAA/Cr[DOWNWARDS ARROW] Cho/Cr[DOWNWARDS ARROW]20 medicated SP, age: 23–43, ID: 45.8 (15.4)

1.5T

PRESS SVS

long TE

Equal gender distribution  
 FL 16 control participants, age: 24–36Cr ratiosNo reliability estimates  
     No PVC  
O'Neill et al. (2004)Inferior ACG, body caudate nucleus, putamen

Cr

NAA

11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 Frontal WM, parietal WM, PL, OL 20 control participants, age: 11.7 (2.9)2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 Superior ACG

Cr[UPWARDS ARROW]

Cho[UPWARDS ARROW]

  PVC  
 

Caudate head

FL

Cho[UPWARDS ARROW]     
Stone et al. (2009)ACGGln[UPWARDS ARROW]27 ARMS, age: 25 (5)3T, PRESS, short TE, SVSEqual gender distributionGlu in left thalamus was negatively correlated with GM volume in several structures 
 Left thalamus

Glu[DOWNWARDS ARROW]

NAA[DOWNWARDS ARROW]

27 control participants, age: 25 (4)AV

CRLBV

SNR

FWHM

  
 Left hippocampusCr, Cho, mI  PVC  
Szulc et al. (2011)ThalamusGlx/Cr42 CSP assessed before & after medication; age: 32.2(6.0); ID: 110.4 (68.4)

1.5T

long TE SVS

PRESS

Unequal gender distribution but matchedGlobal NAA/Cr lower in patientsNon-random medicational switch
 FLNAA/Cr[DOWNWARDS ARROW] (thalamus & FL)11 control participants, age: 37 (10)Cr ratiosNo reliability estimatesGlx/Cr decreased in the temporal lobe after treatment 
 TLCr  No PVC  
Szulc et al. (2007)Left FL & left TL 58 CSP split based on treatment (typ/atyp), age: 33.1 (6.9), ID: 112.92(70.8)1.5T, SVS, short TE, PRESSGender covariateNegative correlation between NAA ratios & age in the FL & between ID & hospitalizations in TL 
 Left thalamusNAA/Cr[DOWNWARDS ARROW] (CSP typical)control participants, age: 30.2 (5.3) Cr & H2O ratiosNo reliability estimates  
     No PVC  
Szulc et al. (2004)Left thalamusGlx/Cr[UPWARDS ARROW] (FEP vs. control participants)31 FEP, medicated for 8 weeks, age 22.55 (3.50)1.5T, PRESS, short TE, SVSNo gender covariatePositive correlation of negative symptomatology & Glx levels 
 

Left PFC

left TL

NAA,

Cr, Cho,

Glx/H2O

17 CSP, medicated, age: 33.59 (7.40)Cr & H2O ratiosNo reliability estimates  
   13 control participants No PVC  
Théberge et al. (2007a,b)Left ACGln[UPWARDS ARROW]16 drug-naïve FEP, reassessment at 10 & 30 months after treatment with neuroleptics, age: 25 (8)4T, SVS, STEAM, short TEUnequal gender distribution, no gender covariate, no data on control participantsReduced thalamic Gln after 30 months of treatmentThe first MRS assessment of 12 of the FEP & 6 control participants was part of the Théberge et al. (2002) study
 Left medio dorsal thalamusGln[UPWARDS ARROW]16 control participants, age: 29 (12)AVNo reliability estimates but COV < 75%Small GM reductions after 10 months & widespread GM reductions after 30 months 
  Glu, Cho, Cr, NAA  PVCReduced PL & TL GM correlated with thalamic Gln decline 
Théberge et al. (2003)Left AC

Glu[DOWNWARDS ARROW]

Gln[DOWNWARDS ARROW]

21 medicated CSP (1 female); age: 37 (11); ID: 188 (107)4T, SVS, short TE, STEAMUnequal gender distribution, no gender covariat, but matchedAtypically medicated patients showed increased thalamic Cho 
 Left medio dorsal thalamusGln[UPWARDS ARROW]21 control participants; age: 33 (12)AVNo reliability estimates but COV < 75%Correlations between NAA & ID, between Cho & age in patients, & between mI & age in control participants 
  Cho, Cr, NAA  PVC  
Théberge et al. (2002)Left ACGln[UPWARDS ARROW]21 FE, drug-naïve; age: 26 (7)4T, SVS, STEAM, short TEUnequal gender distribution, no covariate, but matched  
 Left medio dorsal thalamusGln[UPWARDS ARROW]21 control participants; age: 26 (7)AVCOV < 75%  
  Glu, Cho, Cr, NAA  PVC  
Yoo et al. (2009)

ACG

left DLPFC

mI, Glx22 HRS, age: 22.64 (5.28)1.5T, PRESS, long TE, SVS, AV, ratiosEqual gender distribution Control participants were recruited from an internet advertisement & via the social networks of hospital staff members
 Left thalamus

NAA[DOWNWARDS ARROW]

Cho[DOWNWARDS ARROW]

Cr[DOWNWARDS ARROW]

22 control participants, age: 23.09 (4.84) 

CRLBV

SNR

FWHM

 Solely reduced metabolites when assessed in AV, not in ratios
     CSF correction  
Yasukawa et al. (2005)

Left insular cortex

left ACG

NAA/Cr[DOWNWARDS ARROW]

Cho/Cr[DOWNWARDS ARROW]

mI/Cr[DOWNWARDS ARROW]

(both groups)

15 SP with Gilbert's syndrome & 15 without, all medicated, age: 33.3 (6.5)/32 (4.9), ID: 15.6 (21.6)/20.4 (25.2)1.5 T, PRESS, SVS, short TE, Cr ratiosEqual gender distributionSchizophrenia patients with Gilbert's syndrome were more severely affected 
 Left thalamusmyI/Cr[DOWNWARDS ARROW] (patients without Gilbert's syndrome)20 control participants, age: 36.1 (6.8) No reliability estimates  
     No PVC  
Table 4. 1H-MRS studies of the basal ganglia in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, at-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Ando et al. (2002)Lenticular nucleusCho/Cr[UPWARDS ARROW] (both CSP groups)7 CSP with tardive dyskinesia, age: 29.7 (13.4), ID: 48 (19.2)

1.5T

SVS PRESS short TE

Equal gender distributionTrend for a higher ratio of Cho/Cr in more severe tardive dyskinesia CSP 
  NAA/Cr7 CSP without tardive dyskinesia, age 26.4 (8.8), ID: 50.4 (50.4)Cr ratiosFWHM < 4 HZ (0.09 ppm)  
   7 control participants, age: 29 (4.3) No PVC  
Block et al. (2000)Left FLNAA/Cho[DOWNWARDS ARROW] (CSP)25 familial medicated CSP, age: 35.6 (8.3), ID: 11.5 (9.1) years1.5T, SVS, short & long TE, PRESSUnequal gender distributionInverse correlation between age & NAA/Cho in BG & FLFamilial schizophrenia patients also included patients with schizoaffective disorder
 Left BGNAA/(P)Cr, Cho/(P)Cr13 family members with mixed psychiatric disorders, age: 45.4 (14.96), ID: 40.8 (81.6)Cr & Cho ratiosNo reliability estimatesInversely correlation between age & NAA/(P)Cr & Glx/(P)Cr in FLNo significant differences between Cr ratios
  mI/(P)Cr & Glx/(P)Cr35 unaffected family members, age: 49.2 (15.41) No PVC  
   19 control participants, age: 40.2 (15.3)    
Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFC

NAA/Cr[DOWNWARDS ARROW]

NAA/Cho[DOWNWARDS ARROW]

24 mainly medicated schizophreniform disorder Patients, age: 23.7 (6.1)1.5T, MRSI, long TE, Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients afterwards
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the Positive & Negative Syndrome Scale 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior & posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Bustillo et al. (2008)Left Caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9), ID: 41.4 (69)1.5T, PRESS, SVS, short TEGender covariateGlobal NAA reductions related to 
 Left FL, left OLNAA[DOWNWARDS ARROW] globally before treatment21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

Global cognitive performance in the whole subject sample 
 Right cerebellumCho, Glx, mI  PVCNo difference after treatment 
Callicott et al. (2000) DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)1.5T, MRSI, long TE,Unequal gender distribution Inverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, Cingulate, superior temporal gyrusCho/Cr, NAA/Cho73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, Thalamus, OFC   No PVC  
Callicott et al. (1998) Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2(8.8)

1.5T

MRSI

long TE

No gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WM NAA/Cho60 unaffected siblings, age: 34.6 (8.6) Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCho/Cr66 control participants; age: 32.9 (8.2) No PVC  
de la Fuente-Sandoval et al. (2011)Dorsal-caudate

Glu[UPWARDS ARROW] (in caudate of both patient groups)

Glu/Gln (caudate FEP)

18 drug-naïve subjects with prodromal symptoms for schizophrenia, age: 19.56 (3.46)

3T

SVS

short TE

Equal gender distribution Trend for mI[DOWNWARDS ARROW] in patients with prodromal symptoms versus control participants and versus FEPTwo patient groups vs. one control group
 Cerebellum

NAA[UPWARDS ARROW]

(P)Cho[UPWARDS ARROW]

(in both patients & both regions)

18 drug-naïve FEP, age: 23.44 (4.93)AVFWHM  
   (P)Cr, mI40 control participants, age: 21.83(4.47) PVC  
de la Fuente-Sandoval et al. (2009)Right caudate nucleusGlu/Cr[UPWARDS ARROW] NAA/Cr[DOWNWARDS ARROW] (during acute psychosis & after 6 weeks of treatment)14 medicated FEP & CSP, age: 24.9 (5.9)3T, PRESS, MRSIEqual gender distribution Metabolite changes independent of treatment 
  Cho/Cr14 control participants, age: 26.6 (5.9)Cr RatiosNo reliability estimates  
     PVC  
Fannon et al. (2003)Left hippocampus

NAA/Cr[DOWNWARDS ARROW]

(drug-naïve FEP at baseline vs. medicated FEP & control participants)

12 drug-naïve FEP, age: 26.1 (5.5) ID: 22.02 (15.41) weeks. Reassessed after treatmentPRESS, short TE, SVSUnequal gender distribution, but covariate usedTreatment eliminated NAA difference3 of the 33 FEP had also a schizo-affective disorder
 Left FL, left BGCho/Cr21 medicated FEP, age: 24 (5.8), ID: 49.45(33.19) weeksCr ratiosSNR but not specifiedBoth patient groups had reduced left hippocampal volume as compared to control participants 
   18 control participants, 25.3 (6.7) No PVC  
Gimenez et al. (2003)Left striatumNAA/Cho11 drug-naïve male FEP, age: 22.18 (4.09), ID: 15.09 (14.88)1.5T, PRESS, SVS, short TEOnly malesTrend for increased NAA/Cho ratio in FEP (= 31, = 0.052) 
   11 male control participants, age: 22.09 (3.78)Cho ratiosNo reliability estimatesInverse correlation between the BG NAA/Cho ratio & procedural learning 
     No PVC  
Goto et al. (2011)BGNAA/Cr[DOWNWARDS ARROW] (BG & parieto-occipital lobe)18 FEP, age: 29 (11), ID: 9.4 (6.8)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution Positive correlation between NAA/Cr of the left BG & plasma MHPG in all subjects 
 Parieto-occipital lobe  18 control participants, age: 30 (11) Cr ratiosNo reliability estimates   
 FL   No PVC  
Goto et al. (2010)FLGABA16 drug-naïve FEP treated with atypical anti-psychotic drugs & reassessed after 6 months, age: 29 (11)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution   
 Parieto-occipital 18 control participants, age: 30 (11) years  No reliability estimates  
 Left BG   No PVC  
Miyaoka et al. (2005)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (both)15 SP with Gilbert's syndrome (GS) & 15 without Gilbert's syndrome (wGS)1.5T, PRESS, SVS, short TEEqual gender distribution Schizophrenia with GS more severely & affected (also in BG & vermis) 
 Left BGCho, mIGS/wGS, all medicated, age: 32.5 (10.7)/34(11), ID: 1.3 (1.8)/1.7 (2.1)Cr ratiosNo reliability estimates  
 Vermis 15 control participants, age: 41.7(14.7) No PVC  
Ohara et al. (2000)Lenticular nucleusNAA/Cho10 mainly medicated simple schizophrenia patients, age: 27.1 (2.73), ID: 98.4 (34.84) 1.5T, SVS, PRESS, short TEUnequal gender distribution, but matched  
  Cho/Cr, NAA/Cr10 control participants, age: 29.1 (4.56)Cr & Cho ratiosFWHM <4 Hz (0.09 ppm)  
     No PVC  
O'Neill et al. (2004)Inferior ACG, body caudate nucleus, putamenCr, NAA11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 Frontal WM, parietal WM, PL, & OL 20 control participants, age: 11.7(2.9) 2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 Superior ACGCr[UPWARDS ARROW], Cho[UPWARDS ARROW]   PVC  
 

Caudate head

FL

Cho[UPWARDS ARROW]     
Tayoshi et al. (2010)Left BG GABA38 CSP, age: 34 (10), ID: 133,2 (112.8)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution The right handed typical anti-psychotically treated schizophrenia patients had significantly higher GABA levels in the BG than those taking atypical antipsychotics  
 ACG 29 control participants, age: 34 (10.2)AVCRLBV < 20%Cingular GABA levels correlated negatively with the dose of anti-psychotics 
     CSF correctionBG GABA levels correlated positively with the dose of anti-cholinergics  
Tayoshi et al. (2009)Left BGGln, NAA, Cr, Cho30 medicated CSP, mainly paranoid subtype, age: 34.9 (10.7), ID: 123.6 (104.4)3T, SVS, short TE, STEAMEqual gender distribution Effect of gender on Gln in the ACG & on Cr & NAA in the ltBG.Control values not in range for ACG: mI [UPWARDS ARROW] (males) Cho[UPWARDS ARROW] (both)
 ACG

Glu[DOWNWARDS ARROW]

mI[DOWNWARDS ARROW]

25 control participants, age: 33.8 (9.5)AVNo reliability estimatesDose-dependent effect of benzodiazepine treatment on mI levels in the ACG & Cr, Gln & Glu levels in the BGBG: Cr[DOWNWARDS ARROW] NAA [DOWNWARDS ARROW] (male vs. female)
Table 5. 1H-MRS studies of the dorso-lateral prefrontal cortex in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; GPCho+PCho, glycerophosphocholine & phosphocholine; PCr+Cr, phosphocreatine & creatine; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; PVC, partial volume correction; PFC, prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; Nacetylaspartylglutamate (NAAG); PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode.; TE, echo time; TL, temporal lobes.

Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFCNAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]24 mainly medicated schizophreniform disorder patients, age: 23.7 (6.1)

1.5T

MRSI

long TE

17 males versus 7 females, no data for control participants givenPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients afterwards
 Inferior frontal gyrus, superior temporal gyrusCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratios

FWHM

(no specified values)

  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the PNSS 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Callicott et al. (2000)

DLPFC

hippocampus

NAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)1.5T, MRSI, long TE,Unequal gender distributionInverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/post. Cingulate, superior temporal gyrus

Cho/Cr

NAA/Cho

73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, Thalamus, OFC   No PVC  
van Elst et al. (2005)Left DLPFC

Gln[UPWARDS ARROW]

Glu[UPWARDS ARROW]

21 medicated CSP, only paranoid subtype, age: 28.5 (1.4), ID: 63.8(9.4)2T, short TE, SVS, PRESSUnequal gender distribution, no gender covariatePrefrontal & hippocampal Glu concentrations inversely correlated with psychiatric disturbance of the last 2 yearsOverlapping control group with Rusch et al. (2008) & Olbrich et al. (2008)
 Left anterior hippocampusGlu[UPWARDS ARROW]32 control participants, age: 28.2 (1.0)AVCOV < 21% Glu, Gln, Myo values of the control group far lower than reported in the literature
  NAA, Cr, Cho, mI  CSF correction Macromolecular contribution to the MRS signal not assessed
Kegeles et al. (2012)DLPFCGlx16 medicated SP, age: 32 (10), ID: 9 (8) & 16 drug-naïve SP, age: 32 (11), ID: 7 (7)

3T

SVS

PRESS

Unequal gender distribution, but matchedNAA levels were lower in unmedicated patients compared with control participants 
 mPFCGABA[UPWARDS ARROW] (mPFC)22 control participants, age: 33 (8)AVNo reliability estimates  
     PVC  
Molina et al. (2005)Bilateral DLPFCCho/Cr16 medicated recent-onset patients, age: 23.2 (2.7), ID: 19.2 (10.8) 1.5T, PRESS, SVS, long TEGender covariateSignificant inverse relationship between left-side NAA/Cr & disease duration 
 Left DLPFCNAA/Cr[DOWNWARDS ARROW] in CSP versus ROP/control participants19 medicated CSP, age: 33.7 (6.9), ID: 117.6 (108)Cr ratiosNo reliability estimates  
   20 control participants, age: 28.4 (7.8)  No PVC  
Ohrmann et al. (2007)Left DLPFC

Glx[DOWNWARDS ARROW]

NAA[DOWNWARDS ARROW]

Cho[DOWNWARDS ARROW] (CSP vs. control participants/FEP)

20 CSP, age: 30.3(7.3) 1.5T, STEAM, SVS, Short TENo gender covariate but matchedReduced NAA significantly correlated with performances in verbal learning & memory & with severity of symptomsNo macromolecules included in spectra
  Cr[DOWNWARDS ARROW] (both)15 unmedicated FEP, age: 27.0(6.9)AVNo reliability estimates   
  mI, GABA, Lac20 control participants, age: 28.1(6.5) PVC  
Ohrmann et al. (2008)Left DLPFCNAA[DOWNWARDS ARROW]43, medicated CSP, age: 27.9 (8.2), ID: 43.7 (67.5)1.5T, PRESS, short TEUnequal gender distribution, not well matched The NAA concentration in the ACG, but not in the DLPFC, correlated negatively with the total PANSS scoreNAA control values in ACG are lower than in reference values stated
 ACG

Cho

Cr

Glx

37 control participants, age: 27.2 (5.9)AVNo reliability estimatesCorrelation between learning potential & NAA levels in the ACG in CSP, whereas in healthy control participants, this variable was related to NAA levels in the DLPFCNo macromolecules included in spectra
     No PVC  
Olbrich et al. (2008)Left DLPFCGlu[UPWARDS ARROW]9 medicated FEP, age 28.4 (7.3), 2T, PRESS, SVS, short TE Frontal glutamate concentrations correlated with rating scores for schizophreniform symptomsAlready used control group (van Elst et al. 2005) with decreased Glu/Gln values, small sample size
 Left hippocampus

NAA, Cho

Cr, Gln

32 control participants, age 28.2 (5.8) COV < 75%  
 OL WM   No PVC  
Rusch et al. (2008)Left DLPFC Glx[UPWARDS ARROW]29 recurrent & FEP, medicated, paranoid subtype schizophrenia patients; age: 27.8 (6.1), ID: 52.9 (47.1) 2T, short TE, SVS, PRESSUnequal gender distribution but matched Reduced amygdalar volume associated with impaired executive functioning in schizophrenia patients & control participants

Glu control values very

low - see van Elst et al. (2005) & Olbrich et al. (2008)

 Left hippocampus

NAA

Cr, Cho

31 control participants, age: 28.2 (5.9)AVNo reliability estimates  
     No PVC  

Sanches et al.

Bilateral cingulate cortexNAA/(Cr+Cho)[DOWNWARDS ARROW]37 medicated CSP, age 29.73 (1.66), ID: 120.36 (16.08)1.5T, PRESS, MRSI, long TEUnequal gender distribution, no control mentioned No difference between electrodermally responsive & non-responsive patients, but between schizophrenia patients split based on low versus normal skin conductance level 
 Right DLPFCNAA/(Cr+Cho)[DOWNWARDS ARROW]37 control participants, age: 29.43 (1.6)Cr & Cho ratiosFWHM < 10  
 Perirolandic areasCho, Cr  No PVC  
Sigmundsson et al. (2003)Bilateral DLPFCNAA, Cr, Cho25 Mainly male, medicated CSP with deficit syndrome, age: 34.9(8.0), ID: 154.8(84)

1.5T PRESS

SVS

long TE

Unequal gender distribution (24/1) but matched (22/4)Negative correlation between NAA & severity of symptoms in SPHigh Cho values in control group
   26 control participants, age: 31.8(6.7)AVSNR mentioned but no data given  
     CSF correction  
Stanley et al. (2007)Left DLPFC NAA[DOWNWARDS ARROW] (both patient groups)8 early-onset, drug-naïve FEP, age: 17.5 (2.1), DO: 15.5 (1.82)

1.5T STEAM SVS

short TE

Gender/age entered in regression modelFEP had a 13% decrease in NAA levels compared to control participants 
  

Glu

mI

(GP)Cho+(P)Cho

10 adult-onset, drug-naïve, FEP, age: 28.0 (4.6), DO: 25.7 (4.0)AVCRLBV SNR FWHMSignificant positive correlation between NAA levels & age in schizophrenia patients 
  (P)Cr+Cr61 control participants, age: 24.0 (6.6)  PVC  
Stanley et al. (1996)Left DLPFC

NAA

Glu

Gln[DOWNWARDS ARROW] (on acute medication vs. off)

Gln[UPWARDS ARROW] (CSP)

13 FEP, age: 26 (7), ID: 2.0 (1.9) years;

12 acute medicated SP, age: 26 (7), ID: 2.5 (2.4) years

1.5T STEAM

SVS

short TE

Only males in control groupPositive correlation between Gln & length of illness in combined schizophrenia groupThree treatment groups versus one control group
  

GABA

aspartate

12 CSP, age: 41 (5), ID: 17 (7) yearsAV

COV: 10–33%

SNR: 30

  
  

NAAG

(P)Cr + (P)Cho

24 control participants, age: 32 (11) No PVC  
Tang et al. (2007)mTL WM NAA/Cr[DOWNWARDS ARROW]42 medicated CSP, age: 38.69 (11.42), DO: 23.53 (7.09)

3T

PRESS

No gender covariateReduced anisotropy in mTL. NAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
 DLPF WM, OL WM  42 control participants, age: 43.3 (20.18)Short TE, MRSI No reliability estimatesNAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
    Cr ratiosNo PVC  
Yoo et al. (2009)

ACG

left DLPFC

mI

Glx

22 HRS, age: 22.64 (5.28) 1.5T, PRESS, long TE, SVS, AV & ratiosEqual gender distribution  Control participants were recruited from an internet advertisement & via the social networks of hospital staff members
 Left thalamus

NAA[DOWNWARDS ARROW]

Cho[DOWNWARDS ARROW]

Cr[DOWNWARDS ARROW]

22 control participants, age: 23.09(4.84) 

CRLBV

SNR

FWHM

 Solely reduced metabolites when assessed in AV, not in ratios
     CSF correction  
Zabala et al. (2007)Left DLPFC NAA/H2O[DOWNWARDS ARROW]8 medicated FEP, age: 15.63 (2.13), ID: 14.11 (11.61) weeks1.5T, PRESS, SVS, long TECovariate  
 Right DLPFC

Cr

Cho

15 medicated non-schizophrenic adolescents with first psychosis, age: 15.80(1.37)H2O ratiosSNR  
   32 control participants, age: 15.42 (1.52) PVC but no according analyses  
Table 6. 1H-MRS studies of the medial prefrontal cortex in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; PVC, partial volume correction; PFC, prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TE, echo time; TL, temporal lobes.

Bartha et al. (1999)Left medial PFCGln[UPWARDS ARROW]10 drug-naïve, ROP, age: 24.4 (5.1)1.5 T, STEAM, SVS, short TEUnequal gender distribution, no covariate, but matched Bad resolution of Glu, Gln, GABA high SD
 (incl. ACG)NAA, Glu, GABA, Cr, Cho10 control participants, age: 26.3 (6.4)AV

SNR

COV < 75%

  
     PVC  
Bustillo et al. (2011)Frontal lateral & medial GM,

NAA+NAAG

Glx

mI

18 young SP, age: 23.5 (3.2); 18 old SP: 49.4 (9.6)4T, STEAM, short TE, SVSUnequal gender distribution & non-matchedPositive correlation between Glx & cognitive performance 
 

Parietal & frontal

WM

NAA+NAAG[DOWNWARDS ARROW] (pure GM of young SP vs old control participants)10 young control participants, age: 22.2(4.4)Cr ratios

FWHM < 0.06

SNR > 5

CRLBV < 20

  
 Parietal lateral & medial GMCr[UPWARDS ARROW] (pure GM), Ins[UPWARDS ARROW] (pure GM & WM of old SP)12 old control participants, age: 49.5 (9.2) PVC  
Delamillieure et al. (2000)

mPFC

(including ACG)

NAA/Cr[DOWNWARDS ARROW] (deficit patients vs control participants/non-deficit patients)5 deficit & 17 non-deficit syndrome SP1.5T, STEAM, SVS, short TEGender distribution not mentionedNo difference in metabolites between the untreated & treated patients. 
  mI/Cr21 control participantsCr ratiosNo reliability estimates  
  Cho/Cr  No PVC  
Do et al. (2000)mPFCNAA[DOWNWARDS ARROW], GSH[DOWNWARDS ARROW]14 mixed male SP, age: 18–431.5T, PRESS, SVS, short TE,Only males, covariateSignificant decreases in CSF concentration of GSH & of its metabolite  
  Cho, Cr10 control participants, age: 35.6 (12.9)AVNo reliability estimates  
     No PVC  
Kegeles et al. (2012)DLPFCGlx16 medicated SP age: 32 (10), ID: 9 (8) 3T, SVS, PRESSUnequal gender distribution, but matchedNAA levels were lower in unmedicated patients compared with control participants 
 mPFC

GABA[UPWARDS ARROW]

Glu[UPWARDS ARROW]

(mPFC)

16 drug-naïve SP, age: 32 (11); ID: 7 (7)AVNo reliability estimates  
   22 control participants, age: 33 (8) PVC  
Lutkenhoff et al. (2010)mPFC GMGlu[DOWNWARDS ARROW] (both)9 twins with schizophrenia, age: 48.8 (11.5), ID: 328.8(133.2)3T, PRESS, short TE, SVSEqual gender distribution Effect of age: greater Glu levels in older subjectsControl group values not in range for mI & Cho in hippocampus & mPFC
 Left hippocampusNAA[UPWARDS ARROW], Cr[UPWARDS ARROW], Cho[UPWARDS ARROW]12 Unaffected Co-twins, age: 49.5 (10.0) AVSNR > 3, CRLBV > 25%/20%, FWHM > 0.1 ppm  
 Left PFC WMGlu, Gln, Cr, Cho, mI21 control participants twin pairs, age: 55.7(3.8) PVC  
Matsuzawa et al. (2008)Posterior medial frontal cortexGSH20 medicated CSP, age: 30.7 (5.8), ID: 87.6 (62.4)3T, MEGA-PRESS, SVS, Short TEUnequal (12/8), no covariate but matched  
   16 control participants, age: 30.0 (7.2)AVSNR, FWHM mentioned no values givenCorrelation between GSH levels & the severity of negative symptoms  
     No PVC  
Purdon et al. (2008)Bilateral mPFCGlu/Cr[UPWARDS ARROW]15 adult siblings, age: 46.32 (6.09)3T, STEAM, SVS, short TEUnequal gender distribution, no covariateAfter median stratification analysis 67% of the siblings versus 29% of the control participants had elevated Glu levelsNo difference between groups in mean metabolite levels, only after median stratification
  mI, Cho, NAA, Gln14 control participantsCr ratiosCRLBV < 15%  
     No PVC  
Reid et al. (2010)ACGNAA/Cr[DOWNWARDS ARROW] (PCG)26 medicated CSP, age: 40.4 (13.1)3T, SVS, long TE, PRESSUnequal gender distribution but matchedNegative correlation between Glx/Cr & negative symptoms  
 mPFCGlx/Cr23 control participants, age: 37.2 (12.4)Cr ratiosFWHM > 25 Hz CRLBV > 30% Trend-level decrease in NAA/Cr levels in CSP 
     No PVC  
        
Tibbo et al. (2004)Right mFCGlx/Cr[UPWARDS ARROW]20 adolescent siblings of SP, age: 16.4 (1.99) 3T, SVS, STEAM, short TEEqual gender distribution GLX/Cr levels correlated with overall functioning in siblings 
  mI, Cho, NAA22 control participants, age: 16.7 (1.70)Cr ratiosNo reliability estimates  
     No PVC  
Table 7. 1H-MRS H-MRS studies of the frontal lobes in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFCNAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]24 mainly medicated schizophreniform disorder Patients, age: 23.7 (6.1), ID:1.5T, MRSI, long TE,Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients afterwards
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior & posterior cingulate, PF WM, centrum semiovale, putamen, & thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the PANSS 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)1.5T, MRSI, long TEUnequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior & posterior cingulate, PF WM, centrum semiovale, putamen, & thalamus   No PVC  
Block et al. (2000)Left FLNAA/Cho[DOWNWARDS ARROW] (CSP)25 familial medicated CSP, age: 35.6 (8.3), ID: 11.5 (9.1) years1.5T, SVS, short & long TE, PRESSUnequal gender distributionInverse correlation between age & NAA/Cho in BG & FLFamilial schizophrenia patients also included patients with schizoaffective disorder
 Left BGNAA/(P)Cr, Cho/(P)Cr13 family members with mixed psychiatric disorders, age: 45.4 (14.96), ID: 40.8 (81.6)Cr & Cho ratiosNo reliability estimatesInversely correlation between age & NAA/(P)Cr & Glx/(P)Cr in FLNo significant differences between Cr ratios
  mI/(P)Cr & Glx/(P)Cr35 unaffected family members, age: 49.2 (15.41) No PVC  
   19 control participants, age: 40.2 (15.3)    
Bustillo et al. (2011)Frontal lateral GM, frontal WM, frontal medial GM, parietal

NAA[DOWNWARDS ARROW](GM of young SP)

Glx/Cr

18 young SP, age: 23.5 (3.2); 18 old SP: 49.4 (9.6)4T, STEAM, short TE, SVSUnequal gender distribution & non-matched Positive correlation between Glx & cognitive performance Pooled ratios from two lobes
 lateral GM, parietal medial GM, parietal WMmI[UPWARDS ARROW]/Cr (GM/WM of old SP)10 young control participants, age: 22.2 (4.4) Cr Ratios

FWHM < 0.06

SNR > 5

CRLBV < 20

  
  Cr[UPWARDS ARROW] (in GM of old SP)12 old control participants, age: 49.5 (9.2) PVC  
Bustillo et al. (2010)ACGGlu 14 minimally treated SP, before & after medicational treatment, age: 27.2 (8.9); ID: 30.8 (43.6)4T, SVS, STEAM, short TEUnequal gender distribution but matchedNo effect of medicational treatment on metabolites 
 Frontal WM

Gln

Glu/Gln[UPWARDS ARROW] (ACG)

10 control participants, age: 28.8 (9.7)AVFWHM < 13 Hz  
 ThalamusNAA[DOWNWARDS ARROW](ACG)  PVC  
Bustillo et al. (2008)Left Caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9),ID: 41.4 (69)1.5T, PRESS, SVS, short TEGender covariateGlobal NAA reductions related to 
 

Left FL

left OL

NAA[DOWNWARDS ARROW]

(globally before treatment)

21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

Global cognitive performance in the whole subject sample 
 Right cerebellum

Cho

Glx

mI

  PVCNo difference after treatment 
Bustillo et al. (2008)Left Caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9),ID: 41.4 (69)

1.5T

PRESS

SVS

short TE

Gender covariateGlobal NAA reductions related to global cognitive performance in the whole subject sample 
 Left FL, left OL

NAA[DOWNWARDS ARROW]

(globally before treatment)

21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

No difference after treatment 
 Right cerebellum

Cho

Glx

mI

  PVC  
Callicott et al. (2000) DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)

1.5T

MRSI

long TE

Unequal gender distribution Inverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/post. cingulate, superior temporal gyrus

Cho/Cr

NAA/Cho

73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, Thalamus, OFC   No PVC  
Callicott et al. (1998) Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2 (8.8)

1.5T

MRSI

long TE

No gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WMNAA/Cho60 unaffected siblings, age: 34.6 (8.6) Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Chang et al. (2007)Bilateral mTL WM

NAA[DOWNWARDS ARROW]

mI[DOWNWARDS ARROW]

Glx[UPWARDS ARROW]

23 Elderly medicated schizophrenia patients, with cognitive & functional decline, age: 66.3 (7.2), ID: 517.2 (64.8)

4T

short TE

SVS

short TE

Unequal gender distribution, no covariate usedSchizophrenia participants with severe mental decline had the lowest NAA concentrations in FL/TLLow Glx control participants values
 

PF WM

OL WM

Cr, Cho22 control participants, age: 70.0 (5.3)AV

CRLBV

FWHM

SP had age-related decline in Cr & Cho in the right FL  
     No PVCNo difference between schizophrenia patients treated with atypical versus typical & atypical anti-psychotics 
Fannon et al. (2003)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (in drug-naïve FEP at baseline vs. medicated FEP & control participants)12 drug-naïve FEP, age: 26.1 (5.5) ID: 22.02 (15.41) weeks, reassessed after treatment

PRESS

short TE

SVS

Unequal gender distribution, but covariate usedTreatment eliminated NAA difference3 of the 33 FEP had also a schizoaffective disorder
 

Left FL

left BG

Cho/Cr21 medicated FEP, age: 24 (5.8), ID: 49.45 (33.19) weeksCr ratiosSNR but not specifiedBoth patient groups had reduced left hippocampal volume as compared to control participants 
   18 control participants, 25.3 (6.7) No PVC  
Galinska et al. (2009)

Left FL

left TL

NAA

Cho

mI

Glx

30 FEP, medicated (26 paranoid subtype) age: 22.5 (3.6), DO: 21.8 (3.3) 1.5T, PRESS, SVS, short TEGender covariateNo differences were found between the groups of patients with short & long DUP & control participants 
 Left thalamus 19 control participants, age: 22.5 (3.3)Cr & H2O RatiosNo reliability estimates  
     No PVC  
Goto et al. (2011)BG

NAA/Cr[DOWNWARDS ARROW]

(BG & parieto-occipital lobe)

18 FEP, age: 29 (11), ID: 9.4 (6.8)3T, MEGA-PRESS, SVS, short TEEqual gender distribution Positive correlation between NAA/Cr of the left BG & plasma MHPG in all subjects 
 Parieto-occipital lobe  18 control participants, age: 30 (11) Cr ratiosNo reliability estimates   
 FL   No PVC  
Goto et al. (2010)FLGABA16 drug-naïve FEP treated with atypical anti-psychotic drugs & reassessed after 6 months, age: 29 (11)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution   
 Parieto-occipital 18 control participants, age: 30 (11) years  No reliability estimates  
 Left BG   No PVC  
He et al. (2012)FL WM

NAA/Cr mI/Cr

Cho/Cr

63 drug-naïve FEP, age: 23.94 (8.48); ID: 8.38 (12.33)3T, short TEUnequal gender distribution, but matchedNegative correlation between positive symptoms & the NAA/Cho in the WM of the left FL 
 HippocampusNAA/Cho63 control participants, age: 23.97 (8.38)Cr & Cho ratios

FWHM < 10 Hz SNR > 12 (hippocampus) SNR > 40

(FL WM)

  
     No PVC  
Jessen et al. (2006)

left FL

(both vs. control participants)

NAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]10 early ARMS, medicated, age: 27.0 (6.8)SVS, 1.5T, PRESS, long TEGender covariateARMS that converted to schizophrenia had a higher Cho/Cr & lower NAA/Cho ratio in ACG compared with non-convertersAll control group values higher than in reference literature
 

ACG

(both vs. control participants)

NAA/Cr[DOWNWARDS ARROW]9 late ARMS, medicated, age: 28.7 (7.0)Cr/Cho ratiosNo reliability estimatesNo difference between unmedicated & medicated ARMS 
 Left superior TLCho/Cr 21 medicated SP, age: 33.4 (7.2) No PVC  
Lutkenhoff et al. (2010)mPFC GMGlu[DOWNWARDS ARROW] (both)9 Schizophrenic twins, age: 48.8 (11.5), ID: 328.8 (133.2)3T, PRESS, short TE, SVSEqual gender distribution Effect of age: greater Glu levels in older subjectsControl group values not in range for mI & Cho in left hippocampus & mPFC
 Left hippocampusNAA[UPWARDS ARROW], Cr[UPWARDS ARROW], Cho[UPWARDS ARROW]12 unaffected co-twins, age: 49.5(10.0) AV

SNR > 3

CRLBV > 25/20% FWHM >0.1

 Threshold SNR in left hippocampus: 5.2(2.1)
 Left PFC WMGlu, Gln, Cr, Cho, mI21 control participants twin pairs, age: 55.7 (3.8) PVC  
Omori et al. (2000)Left thalamusNAA/Cr[DOWNWARDS ARROW] Cho/Cr[DOWNWARDS ARROW]20 medicated SP, age: 23–43, ID: 45.8 (15.4)1.5T, PRESS, SVS, long TEEqual gender distribution   
 FL 16 control participants, age: 24–36Cr ratiosNo reliability estimates  
     No PVC  
O'Neill et al. (2004)Inferior ACG, body caudate nucleus, putamen,

Cr

NAA

11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 frontal WM, parietal WM, PL, OL 20 control participants, age: 11.7 (2.9) 2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 Superior ACGCr[UPWARDS ARROW], Cho[UPWARDS ARROW]   PVC  
Ota et al. (2012)Left FL WMCr, NAA, mI, Cho24 SP with psychotic exacerbation, age: 45.0 (15.4), ID: 20.4 (15.6) 1.5T, SVS, PRESS, short TESex matchedGlx[UPWARDS ARROW] in SP with psychotic exacerbation compared with non-exacerbated patientsExacerbation group included one FEP
 Inferior PL WMGlx[UPWARDS ARROW] (patients with execerbation)22 patients without exacerbation, age: 41.3 (11.2), ID: 14.6 (8.8)AVsSD > 20% Significant different PANSS score & medication treatment between SP subgroups
   27 control participants, age: 42.8 (15.1) No CSF correction  
Rowland et al. (2009)Left middle frontal NAA, Cho, Cr, Glx10 medicated deficit syndrome CSP, age: 43 (6), ID: 288 (96)3T, PRESS, SVS, short TE,Unequal gender distribution, but matched  
 Left inferior parietal  10 medicated non-deficit syndrome CSP, age: 40 (6), ID: 228 (84)AVCRLBV < 20%, FWHM < 0.1Reduced fractional anisotropy in right superior longitudinal fasciculus & middle frontal region in deficit patients vs control group 
   11 control participants, age: 37 (10) No PVC  
Szulc et al. (2011)ThalamusGlx/Cr 42 CSP assessed before & after medication; age: 32.2(6.0); ID: 110.4 (68.4)1.5T, long TE, SVS, PRESSUnequal gender distribution but matchedGlobal NAA/Cr lower in CSPNon-random medicational switch
 FLNAA/Cr[DOWNWARDS ARROW] (thalamus & FL)11 control participants, age: 37 (10)Cr RatiosNo reliability estimatesGlx/Cr decreased in the temporal lobe after treatment 
 TL Cr  No PVC  
Szulc et al. (2007)

Left FL

left TL

 58 CSP split based on treatment, age: 33.1 (6.9), ID: 112.92(70.8) 1.5T, SVS, short TE, PRESSGender covariateNegative correlation between NAA ratios & age in the FL & between ID & hospitalizations in TL  
 Left thalamus

NAA/Cr[DOWNWARDS ARROW]

(CSP typical)

control participants, age: 30.2(5.3) Cr & H2O ratiosNo reliability estimates  
     No PVC  
Steel et al. (2001)Bilateral prefrontal WMNAA10 medicated CSP, age: 34 (14), ID: 180 (144)2T, PRESS, long TE, SVSEqual gender distribution 10–15% reduction of NAA levels, but non-significant, probably due to small sample size 
   10 control participants, age: 35 (7)AVSNROnly CSP showed positive correlation between the right & left FL with NAA in the right FL  
     No PVCCorrelation between left frontal NAA & the volume of the left amygdalo–hippocampal complex  
Tunc-Skarka et al. (2009)Left frontal WMNAA[DOWNWARDS ARROW]23 FEP & CSP, 12 drug-naïve & 11 on atypical medication, age: 31.7 (8.8), ID: 23.5 (50.3)3T, PRESS, SVS, short TEGender covariate, medicational staturs as covariateShortened NAA T2 relaxation time in the patient group3 schizoaffective subjects
  

Cho

Cr

29 control participants, age: 32.5 (9.7)AVCRLBV < 20%No differences between FEP versus CSP 
     PVCAge correlated positively with Cr & a trend correlation with Cho 
Table 8. 1H-MRS studies of the parietal lobes in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; MI, myoinositol; Cr, creatine; Cho, choline; MHPG, 3-Methoxy-4-hydroxyphenylglycol NAA, N-acetyl-aspartate; PANSS, the Positive & Negative Syndrome Scale; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SD, standard deviation; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time; WCST, Wisconsin Card Sorting Test.

Auer et al. (2001)Medio-dorsal & lateral thalamic nucleiNAA[DOWNWARDS ARROW]32 acutely-ill, medicated CSP, age: 33.91.5T, PRESS, SVS, short TEUnequal gender, but covariateReduced Cr & Cho ratios in the left thalamusControl values of mI, & NAA under range in thalamus; Cr & NAA under range in WM
 Parietal white matter

mI[UPWARDS ARROW]

Cr[UPWARDS ARROW]

Cho[UPWARDS ARROW]

17 control participants, age: 31.2

Cr & Cho ratios

AV

FWHM No distinction between thalamic tissue itself & the surrounding GM & WM
     PVC  
Bustillo et al. (2011)Frontal lateral/medial GM, frontal WM

NAA[DOWNWARDS ARROW]

(GM of young SP)

Glx/Cr

18 young SP, age: 23.5 (3.2); 18 old SP: 49.4 (9.6)

4T

STEAM

short TE

SVS

Unequal gender distribution & non-matched Positive correlation between Glx & cognitive performance Pooled ratios from two lobes
 Parietal WM

mI/Cr[UPWARDS ARROW]

(GM/WM of old SP)

10 young control participants, age: 22.2 (4.4) Cr Ratios

FWHM < 0.06

SNR > 5

CRLBV < 20

  
 Parietal lateral/medial GM

Cr[UPWARDS ARROW]

(GM of old SP)

12 old control participants, age: 49.5 (9.2) PVC  
Goto et al. (2011)BG

NAA/Cr[DOWNWARDS ARROW]

(BG & parieto-occipital lobe)

18 FEP, age: 29 (11), ID: 9.4 (6.8)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution Positive correlation between NAA/Cr of the left BG & plasma MHPG in all subjects 
 Parieto-occipital lobe  18 control participants, age: 30 (11) Cr ratiosNo reliability estimates   
 FL   No PVC 
Goto et al. (2010)FLGABA16 drug-naïve FEP treated with atypical anti-psychotic drugs & reassessed after 6 months, age: 29 (11)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution   
 Parieto-occipital 18 control participants, age: 30 (11) years  No reliability estimates  
 Left BG   No PVC  
Goto et al. (2009)FL 18 medicated CSP, age: 29 (11)3T, PRESS & MEGA-PRESS, SVS, short TEEqual gender distribution   
 Left BGGABA/Cr[DOWNWARDS ARROW]18 control participants age: 30 (11) Cr ratiosNo reliability estimates  
 Parieto-occipital lobe   No PVC  
Ongur et al. (2010a, b)ACG

GABA/Cr[UPWARDS ARROW]

(both regions)

21 medicated CSP, age: 39 (10.8), DO: 17.9 (3.5)

4T

MEGAPRESS

long TE

MRSI

Equal gender distribution Positive correlation between GABA/Cr & Gl/CrSchizoaffective disorder patients included
 Parieto-occipitalGlu/Cr19 control participants, age: 36.3 (9.8) Cr ratiosCRLBV: 2.9–15.9%   
  NAA/Cr   PVC  
Ongur et al. (2009)ACGCr[DOWNWARDS ARROW] (SP)

15 medicated, acute-episode CSP, age: 42.9 (9.8)

ID: 223.2 (140.4)

4T, SVS, Equal gender distribution Age correlated negatively with Cr in controls but not in patients with either disorder8 CSP had a history of substance abuse
 Parieto-occipital cortex Cr[DOWNWARDS ARROW] (SP)15 Bipolar patients, age: 36.3 (11.6), ID: 129.6 (147.6)AVCRLBV < 4%  
   22 control participants, age: 34.9 (10.0) PVC  
O'Neill et al. (2004)Inferior ACG, body caudate nucleus, putamen, frontal WM, parietal WM, PL, OL

Cr

NAA

11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 Caudate headCho[UPWARDS ARROW]20 control participants, age: 11.7 (2.9) 2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 Superior ACG

Cr[UPWARDS ARROW]

Cho[UPWARDS ARROW]

  PVC  
 FLCho[UPWARDS ARROW]     
 ThalamusNAA[DOWNWARDS ARROW] (male patients vs. female patients/male controls)     
Ota et al. (2012)Left FL WM

Cr, NAA

mI, Cho

24 SP with psychotic exacerbation, age: 45.0 (15.4), ID: 20.4 (15.6) 1.5T, SVS, PRESS, short TESex matchedGlx[UPWARDS ARROW] in SP with psychotic exacerbation compared with non-exacerbated patientsExacerbation group included one FEP
 Inferior PL WMGlx[UPWARDS ARROW] (patients with exacerbation)22 patients without exacerbation, age: 41.3 (11.2), ID: 14.6 (8.8)AVsSD > 20% Significant different PANSS score & medication treatment between SP subgroups
   27 control participants, age: 42.8 (15.1) No CSF correction  
Rowland et al. (2009)Left middle frontal NAA, Cho, Cr, Glx10 medicated deficit syndrom CSP, age: 43 (6), ID: 288 (96)3T, PRESS, SVS, short TE,Unequal gender distribution, but matched  
 Left inferior parietal lobe 10 medicated non-deficit syndrome CSP, age: 40 (6), ID: 228 (84)AVCRLBV < 20%FWHM < 0.1Reduced fractional anisotropy in right superior longitudinal fasciculus & middle frontal region in deficit patients vs. control group 
   11 control participants, age: 37 (10) No PVC  
Sanches et al. (2008)Bilateral cingulate cortexNAA/(Cr/Cho)[DOWNWARDS ARROW]37 medicated CSP, age 29.73 (1.66), ID: 120.36 (16.08)

1.5T PRESS MRSI

long TE

Unequal gender distribution, no control mentioned No difference between electrodermally responsive & non-responsive patients, but between schizophrenia patients split based on low versus normal skin conductance level 
 Right DLPFCNAA/(Cr/Cho)[DOWNWARDS ARROW]37 control participants, age: 29.43 (1.6)Cr & Cho ratiosFWHM < 10  
 Perirolandic areas

Cho

Cr

  No PVC  
Table 9. 1H-MRS studies of the occipital lobes in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Bertolino et al. (2003)Bilateral hippocampus, bilateral DLPFCNAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]24 mainly medicated schizophreniform disorder Patients, age: 23.7 (6.1)1.5T, MRSI, long TE,Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients post 1H-MRS
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the PANSS 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Bustillo et al. (2008)Left caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9),ID: 41.4 (69)

1.5T PRESS SVS

short TE

Gender covariateGlobal NAA reductions related to 
 

Left FL

left OL

NAA[DOWNWARDS ARROW] (globally before treatment)21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

Global cognitive performance in the whole subject sample 
 Right cerebellumCho, Glx, mI  PVCNo difference after treatment 
Callicott et al. (2000)DLPFC, hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)

1.5T

MRSI

long TE

Unequal gender distributionInverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, anterior/posterior cingulate, superior temporal gyrusCho/Cr, NAA/Cho73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, thalamus, OFC   No PVC  
Callicott et al. (1998)Hippocampal area (amygdala, hippocampus, parahippocampal gyrus)NAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2 (8.8)

1.5T

MRSI

long TE

No gender covariate, only one control groupLow hippocampal NAA/Cr phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WMNAA/Cho60 unaffected siblings, age: 34.6 (8.6)Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Chang et al. (2007)Bilateral mTL WM

NAA[DOWNWARDS ARROW]

mI[DOWNWARDS ARROW]

Glx[UPWARDS ARROW]

23 Elderly medicated schizophrenia patients, with cognitive & functional decline, age: 66.3 (7.2), ID: 517.2 (64.8)4T, short TE, SVS, short TEUnequal gender distribution, no covariate usedSchizophrenia participants with severe mental decline had the lowest NAA concentrations in FL/TLLow Glx control group values
 

PF WM

OL WM

Cr

Cho

22 control participants, age: 70.0 (5.3)AV

CRLBV

FWHM

SP had age-related decline in Cr & Cho in the right FL 
     No PVCNo difference between schizophrenia patients treated with atypical versus typical & atypical anti-psychotics 
Goto et al. (2011)BGNAA/Cr[DOWNWARDS ARROW](BG & parieto-occipital lobe)18 FEP, age: 29 (11), ID: 9.4 (6.8)3T, MEGA-PRESS, SVS, short TEEqual gender distributionPositive correlation between NAA/Cr of the left BG & plasma MHPG in all subjects 
 Parieto-occipital lobe 18 control participants, age: 30 (11)Cr ratiosNo reliability estimates  
 FL   No PVC  
Goto et al. (2010)FLGABA16 drug-naïve FEP treated with atypical anti-psychotic drugs & reassessed after 6 months, age: 29 (11)3T, MEGA-PRESS, SVS, short TEEqual gender distribution  
 Parieto-occipital 18 control participants, age: 30 (11) years No reliability estimates  
 Left BG   No PVC  
Goto et al. (2009)FL 18 medicated CSP, age: 29 (11)3T, PRESS & MEGA-PRESS, SVS, short TEEqual gender distribution  
 Left BGGABA/Cr[DOWNWARDS ARROW]18 control participants, age: 30 (11)Cr ratiosNo reliability estimates  
 Parieto-occipital lobe   No PVC  
Olbrich et al. (2008)Left DLPFCGlu [UPWARDS ARROW]9 medicated FEP, age 28.4 (7.3),2T, PRESS, SVS, short TE Frontal glutamate concentrations significantly correlated with rating scores for schizophreniform symptomsAlready used control group (van Elst et al. 2005) with low Glu/Gln values, small sample size
 Left hippocampusNAA, Cho, Cr, Gln32 control participants, age 28.2 (5.8) COV < 75%  
 OL WM   No PVC  
Ongur et al. (2010a, b)ACGGABA/Cr[UPWARDS ARROW] (in both regions)21 medicated CSP, age: 39 (10.8); DO: 17.9 (3.5)4T, MEGA-PRESS, long TE MRSIEqual gender distributionPositive correlation between GABA/Cr & Gl/CrSchizoaffective disorder patients included
 Parieto-occipitalGlu/Cr19 control participants, age: 36.3 (9.8)Cr ratiosCRLBV: 2.9–15.9%  
  NAA/Cr  PVC  
Ongur et al. (2009)ACGCr[DOWNWARDS ARROW]15 medicated, acute-episode CSP, age: 42.9 (9.8), ID: 223.2 (140.4)

4T

SVS

Equal gender distributionAge negatively correlated with Cr in control participants but not in patients with either disorder 
 Parieto-occipital cortexCr[DOWNWARDS ARROW]15 bipolar patients, age: 36.3 (11.6), ID: 129.6 (147.6)AVCRLBV 8 CSP had a history of substance abuse
   22 control participants, age: 34.9 (10.0) PVC  
O'Neill et al. (2004)ThalamusNAA[DOWNWARDS ARROW] (male patients vs. female patients & male control participants)11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 Superior ACG

Cr[UPWARDS ARROW]

Cho[UPWARDS ARROW]

20 control participants, age: 11.7 (2.9)2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 

Caudate head

FL

Cho[UPWARDS ARROW]  PVC  
 Frontal WM, parietal WM, PL, OL      
 Inferior ACG, body caudate nucleus, putamen

Cr

NAA

     
Tang et al. (2007)mTL WMNAA/Cr[DOWNWARDS ARROW]42 medicated CSP, age: 38.69 (11.42), DO: 23.53 (7.09)

3T

PRESS

No gender covariateReduced anisotropy in mTL. NAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
 DLPF WM, OL WM  42 control participants, age: 43.3 (20.18)

Short TE

MRSI

No reliability estimatesNAA[DOWNWARDS ARROW] correlated with anisotropy[DOWNWARDS ARROW] 
        
Yoon et al. (2010)Calcarine sulci bilaterally GABA/Cr[DOWNWARDS ARROW]13 CSP & ROP (mixed), age: 27.05 (8.8)3T, MEGA-PRESSOnly males but matchedPositive correlation between GABA levels & orientation-specific surround suppression  
  Glx/Cr 13 control participants, age: 28.1(8.2)

SVS

long TE

No reliability estimates  
    Cr ratios   
Table 10. 1H-MRS studies of the cingulate gyrus in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time; WCST, Wisconsin Card Sorting Test.

Aoyama et al. (2011)Left thalamusGln[DOWNWARDS ARROW]17 initially drug-naïve SP, followed up for 80 months after treatment, age: 25 (7)

4T

STEAM

SVS

Equal gender distributionGlu was reduced after 80 months in SP compared to control participants 
 Anterior cingulate

NAA

mI

17 control participants 1, age: 32(10); 17 control participants 2 age: 29(10)H2O ratiosCOV < 75%  
  Taurine scylloinositol  PVC through water correction  
Bartha et al. (1999)Left medial PFC (including ACG)Gln[UPWARDS ARROW]10 drug-naïve, ROP, age: 24.4 (5.1)1.5 T, STEAM, SVS, short TEUnequal gender distribution, no covariate, but matched Bad resolution of Glu, Gln, GABA high SD
  NAA, Glu, GABA, Cr, Cho10 control participants, age: 26.3 (6.4)AV

SNR

COV < 75%

  
     PVC  
Braus et al. (2002)ACGNAA[UPWARDS ARROW] (atypically treated)10 typically treated CSP, age: 40 (6.2), ID: 174 (60)1.5 T, PRESS, MRSI, long TEEqual gender distributionAtypically treated patients had fewer errors on WCST & higher NAA, & better performanceSwitch to atypicals in non-randomized way
  

Cr

Cho

11 atypically treated CSP, age: 37 (13), ID: 145 (110)AVNo reliability estimates No control participants
     No PVC  
Bustillo et al. (2010)ACG

NAA[DOWNWARDS ARROW] & Glu/Gln[UPWARDS ARROW]

(before treatment vs. after)

14 minimally treated SP, before & after medicational treatment, age: 27.2 (8.9); ID: 30.8 (43.6)4T, SVS, STEAM, short TEUnequal gender distribution but matchedNo effect of medicational treatment on metabolites 
 Frontal WMGln, Glu10 control participants, age: 28.8 (9.7)AVFWHM < 13 HzInverse correlation of cingular Gln/Glu with NAA in SP but not in the control group 
 Thalamus   PVC  
Bertolino et al. (2003)

Hippocampi

DLPFC

NAA/Cr[DOWNWARDS ARROW]

NAA/Cho[DOWNWARDS ARROW]

24 mainly medicated schizophreniform disorder patients, age: 23.7 (6.1)

1.5T

MRSI

long TE

Unequal gender distributionPositive correlation between DLPFC NAA levels & working memory performanceSchizophrenia diagnosis in 17 patients afterwards
 Inferior frontal gyrus, superior temporal gyrus, OL, PF WMCho/Cr24 control participants, age: 24.0 (6.1)Cr & Cho ratiosFWHM (no specified values)  
 Anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC but comparisons of signal intensities between control participants & subjectsCorrelation between NAA in the right DLPFC & negative symptom ratings from the PANSS 
Bertolino et al. (2001)Bilateral DLPFCNAA/Cr[UPWARDS ARROW] (medicated CSP)23 CSP assessed twice: off medication for 2 weeks & after 4 weeks of medication, age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Callicott et al. (2000)DLPFC hippocampusNAA/Cr[DOWNWARDS ARROW]36 mainly medicated CSP, age: 34 (8)1.5T, MRSI, long TE,Unequal gender distributionInverse correlation between DLPFC NAA levels & negative symptoms 
 Putamen, cingulate, superior temporal gyrusCho/Cr, NAA/Cho73 control participants; age: 32.2 (8.1)Cr & Cho ratiosNo reliability estimates  
 PF WM, OL, centrum semiovale, thalamus, OFC   No PVC  
Callicott et al. (1998)Hippocampal areaNAA/Cr[DOWNWARDS ARROW] (both)47 mainly medicated CSP, age: 34.2 (8.8)

1.5T

MRSI

long TE

No gender covariate, only one control groupLow hippocampal NAA/CRE phenotypes yielded relative risk estimates, suggesting this characteristic is heritable 
 PFC, superior temporal gyrus, OFC, OL, PF WMNAA/Cho60 unaffected siblings, age: 34.6 (8.6)Cr & Cho ratiosNo reliability estimates  
 Anterior/posterior cingulate, centrum semiovale, thalamus, putamenCr/Cho66 control participants; age: 32.9 (8.2) No PVC  
Delamillieure et al. (2000)mPFC (including ACG)NAA/Cr[DOWNWARDS ARROW] (deficit patients vs. control participants/non-deficit patients)5 deficit & 17 non-deficit syndrome SP1.5T, STEAM, SVS, short TEGender distribution not mentionedNo difference in metabolites between the untreated & treated patients. 
  mI/Cr21 control participantsCr ratiosNo reliability estimates  
  Cho/Cr  No PVC  
Ende et al. (2000)ACG (mainly GM)NAA[DOWNWARDS ARROW] (typical CSP lowest)Nine typical medicated CSP, age: 36.9 (6.4), ID: 130 (55.8)MRSI, PRESS, long TEEqual gender distribution Positive correlation of NAA & ageCho control values very high but in subjects as well
  Cr, Cho10 atypical medicated CSP, age: 32.5 (11), ID: 105 (79.3)AVNo reliability estimatesPositive correlation of age-corrected NAA levels in SP with the ID 
   16 control participants, age: 34.9 (13.6) No PVC  
Hardy et al. (2011)Rostral ACGNAA22 CSP, age: 40.6 (10.7)3T, MRSI, long TE, PRESSEqual gender distribution Significant differences between caudal & rostral NAA concentration only in ACG of CSPPatients with schizoaffective disorder included
 Caudal ACGCr11 control participants, age: 35.5 (10.7)AVNo reliability estimates  
  Cho  CSF correction  
Jessen et al. (2013)ACGNAAG, NAA[DOWNWARDS ARROW]20 CSP, age: 34.5 (10.2); ID: 24 (8.5)

3T

PRESS

long

short TE

Unequal gender distribution but matchedNAAG correlated inversely with negative symptoms 
 Left FLGlx/(P)Cr, Gln/(P)Cr20 control participants, age: 30.7 (9.1) CRLB < 50%, FWHM  
  (P)Cho, Cr, mI  CSF correction  
Jessen et al. (2006)left FL (both vs. control participants)NAA/Cr[DOWNWARDS ARROW] NAA/Cho[DOWNWARDS ARROW]10 early ARMS, medicated, age: 27.0 (6.8)SVS, 1.5T, PRESS, long TEGender covariateARMS that converted to SP had higher cingular Cho/Cr & lower NAA/Cho ratios compared with non-convertersAll control values higher than in reference literature
 ACG (both vs. control participants)NAA/Cr[DOWNWARDS ARROW]9 late ARMS, medicated, age: 28.7 (7.0)Cr/Cho ratiosNo reliability estimatesNo difference between unmedicated & medicated ARSP 
 Left superior TLCho/Cr21 medicated SP, age: 33.4 (7.2) No PVC  
   31 control participants, age: 34.8 (13.5)    
Ongur et al. (2010a, b)ACG

GABA/Cr[UPWARDS ARROW]

(both

regions)

21 medicated CSP, age: 39 (10.8); DO: 17.9 (3.5) 4T, MEGA-PRESS, long TE, MRSIEqual gender distribution Positive correlation between GABA/Cr & Glu/CrSchizoaffective disorder patients included
 Parieto-occipitalGlu/Cr19 control participants, age: 36.3 (9.8) Cr ratiosCRLBV: 2.9–15.9%   
  NAA/Cr   PVC  
Ongur et al. (2009)ACGCr[DOWNWARDS ARROW]15 medicated, acute-episode CSP, age: 42.9 (9.8), ID: 223.2 (140.4)4T, SVS, Equal gender distribution Age negatively correlated with Cr in control participants but not in patients with either disorder 
 Parieto-occipital cortex Cr[DOWNWARDS ARROW]15 Bipolar Patients, age: 36.3 (11.6), ID: 129.6 (147.6)AVCRLBV 8 CSP had a history of substance abuse
   22 control participants, age: 34.9 (10.0) PVC  
O'Neill et al. (2004)Inferior ACG, body caudate nucleus, putamen, frontal WM, parietal WM, PL, & OLCr, NAA11 childhood-onset schizophrenia patients; age: 12.3 (3.8)1.5T, long TE, MRSI, AVGender covariateGender effects on several metabolites 
 Caudate headCho[UPWARDS ARROW]20 control participants, age: 11.7 (2.9) 2D inversion-recovery sequence

SNR

FWHM

Frontal Cho levels 34.5% higher in propofol sedated than in unsedated patients 
 Superior ACGCr[UPWARDS ARROW], Cho[UPWARDS ARROW]   PVC  
 FLCho[UPWARDS ARROW]     
 ThalamusNAA[DOWNWARDS ARROW] (male patients versus female patients & male control participants)     
Reid et al. (2010)ACGNAA/Cr[DOWNWARDS ARROW] 26 medicated CSP, age: 40.4 (13.1)3T, SVS, long TE, PRESSUnequal gender distribution but matchedNegative correlation between Glx/Cr & negative symptoms  
 mPFCGlx/Cr23 control participants, age: 37.2 (12.4)Cr ratiosFWHM > 25 Hz CRLBV > 30% Trend-level decrease in NAA/Cr levels in CSP 
     No PVC  
Sanches et al. (2008)Bilateral cingulate cortexNAA/(Cr+Cho)[DOWNWARDS ARROW]37 medicated CSP, age 29.73 (1.66), ID: 120.36 (16.08)1.5T, PRESS, MRSI, long TEUnequal gender distribution, no control mentioned No difference between electrodermally responsive & non-responsive patients, but between schizophrenia patients split based on low versus normal skin conductance level 
 Right DLPFCNAA/(Cr+Cho) [DOWNWARDS ARROW]37 control participants, age: 29.43 (1.6)Cr & Cho ratiosFWHM < 10  
 Perirol and ic areasCho, Cr  No PVC  
Shimizu et al. (2007)PCGNAA/Cr[DOWNWARDS ARROW] 19 CSP, medicated, age: 40.4 (13.1), ID = 195.6 (142.8)1.5T, PRESS, SVS, long TE, Equal gender distribution Age-related decline of NAA/Cr in PCG of control participants 
 Bilateral mTLCho/Cr18 control participants, age: 34.9 (11.4)Cr ratiosNo reliability estimates  
     No PVC  
Stone et al. (2009)ACGGln[UPWARDS ARROW]27 ARMS, age: 25 (5)3T, PRESS, short TE, SVSEqual gender distribution Glu in left thalamus was negatively correlated with GM volume in several structures 
 Left thalamus

Glu[DOWNWARDS ARROW]

NAA[DOWNWARDS ARROW]

27 control participants, age: 25 (4)AV

CRLBV

SNR

FWHM

  
 Left hippocampusCr, Cho, mI  PVC   
Tayoshi et al. (2009)Left BGGln, NAA, Cr, Cho30 medicated CSP, mainly paranoid subtype, age: 34.9 (10.7), ID: 123.6 (104.4)3T, SVS, short TE, STEAMEqual gender distribution Effect of gender on Gln in the ACG & on Cr & NAA in the ltBG.Control values not in range for ACG: mI[UPWARDS ARROW] (males) Cho[UPWARDS ARROW] (both) & BG: Gln, Glu, mI, NAA[DOWNWARDS ARROW] (both)
 ACGGlu[DOWNWARDS ARROW], mI[DOWNWARDS ARROW]25 control participants, age: 33.8 (9.5)AVNo reliability estimatesDose-dependent effect of benzodiazepine treatment on mI levels in the ACG & Cr, Gln & Glu levels in the BG 
     PVC  
Tayoshi et al. (2010)Left BG GABA38 CSP, age: 34 (10), ID: 133.2 (112.8)3T, MEGA-PRESS, SVS, Short TEEqual gender distribution GABA[UPWARDS ARROW] levels in the BG of right handed typical anti-psychotically treated CSP versus CSP taking atypical anti-psychotics  
 ACG 29 ccontrol participants, age: 34 (10.2)AVCRLBV < 20%Cingular GABA levels correlated negatively with the dose of anti-psychotics 
     CSF correctionBasal ganglia GABA levels correlated positively with the dose of anti-cholinergics  
Théberge et al. (2007a, b)Left ACGGln[UPWARDS ARROW]16 drug-naïve FEP, reassessment at 10 & 30 months after treatment with neuroleptics, age: 25 (8)4T, SVS, STEAM, short TEUnequal gender distribution, no gender covariate, no data on control participantsReduced thalamic Gln after 30 months of treatmentThe first MRS assessment of 12 of the FEP & 6 control participants was part of the Théberge et al. (2002) study
 Left medio dorsal thalamusGln[UPWARDS ARROW]16 control participants, age: 29 (12)AVNo reliability estimates but COV < 75%Small GM reductions after 10 months & widespread GM reductions after 30 months 
  Glu, Cho, Cr, NAA  PVCReduced PL & TL GM correlated with thalamic Gln decline 
Théberge et al. (2003)Left AC

Glu[DOWNWARDS ARROW]

Gln[DOWNWARDS ARROW]

21 medicated CSP (1 female); age: 37 (11), ID: 188 (107)4T, SVS, short TE, STEAMUnequal gender distribution, no gender covariat, but matchedAtypically medicated patients showed increased thalamic Cho  
 Left medio dorsal thalamusGln[UPWARDS ARROW]21 control participants; age: 33 (12)AVNo reliability estimates but COV < 75%Correlations between NAA & ID, between Cho & age in CSP & between mI & age in control participants 
  Cho, Cr, NAA  PVC  
Théberge et al. (2002)Left ACGln[UPWARDS ARROW]21 FE, drug-naïve; age: 26 (7)4T, SVS, STEAM, short TEUnequal gender distribution, no covariate, but matched  
 Left medio dorsal thalamusGln[UPWARDS ARROW]21 control participants; age: 26 (7)AVCOV < 75%  
  Glu, Cho, Cr, NAA  PVC  
Yamasue et al. (2002)Bilateral ACGNAA/Cho[DOWNWARDS ARROW] Cho/Cr[DOWNWARDS ARROW]15 medicated CSP, age: 30.4 (7), ID: 93.6 (58.8)1.5T, PRESS, short TE, SVSUnequal gender distribution Positively correlation of GM with the NAA/Cho ratios in CSP 
  NAA/Cr13 control participants, age: 28.8 (3.6)Cr ratiosFWHM < 6 Hz  
     PVC (with inter-rater reliability of 0.97)  
Yasukawa et al. (2005)

Left insular cortex

left ACG

NAA/Cr[DOWNWARDS ARROW], Cho/Cr[DOWNWARDS ARROW], mI/Cr[DOWNWARDS ARROW] (both groups)15 SP with Gilbert's syndrome & 15 without, all medicated, age: 33.3 (6.5)/32 (4.9), ID: 15.6 (21.6)/20.4 (25.2)1.5 T, PRESS, SVS, short TE, Cr ratiosEqual gender distribution Schizophrenia patients with Gilbert's syndrome were more severely affected 
 Left thalamus mI/Cr[DOWNWARDS ARROW] (patients without Gilbert's syndrome)20 control participants, age: 36.1 (6.8) No reliability estimates  
     No PVC  
Yoo et al. (2009)

ACG

left DLPFC

mI, Glx22 HRS, age: 22.64 (5.28) 1.5T, PRESS, long TE, SVS, AV & ratiosEqual gender distribution  Control participants were recruited from an internet advertisement & via the social networks of hospital staff members
 Left thalamus

NAA[DOWNWARDS ARROW]

Cho[DOWNWARDS ARROW]

Cr[DOWNWARDS ARROW]

22 control participants, age: 23.09 (4.84) CRLBV SNR FWHM Solely reduced metabolites when assessed in AV, not in ratios
     CSF correction  
Table 11. 1H-MRS studies of the cerebellum in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; PVC, partial volume correction; PFC, prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TE, echo time; TL, temporal lobes; WCST, Wisconsin-card sorting test.

Bustillo et al. (2008)Left caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9), ID: 41.4 (69)1.5T, PRESS, SVS, short TEGender covariateGlobal NAA reductions related to 
 Left FL, left OLNAA[DOWNWARDS ARROW] (globally before treatment)21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

Global cognitive performance in the whole subject sample 
 Right cerebellumCho, Glx, mI  PVCNo difference after treatment 
Ende et al. (2000)VermisNAA[DOWNWARDS ARROW] Cho[DOWNWARDS ARROW]14 medicated CSP, age: 38.9 (7.3)

1.5T

PRESS

MRSI

long TE

Unequal gender distribution, but covariateTrend for Cr reductions in vermis (= 0.06) & CC (= 0.07)No data on concentrations
 CCNAA[DOWNWARDS ARROW]14 control participants, age: 35.6 (3.7)AVNo reliability estimates Unequal gender distribution of control group
 Dentate nucleus, ponsCr  PVC  
Deicken et al. (2001)Anterior vermisNAA[DOWNWARDS ARROW] Cr[DOWNWARDS ARROW]20 medicated male CSP, age: 34.47 (10.58), ID: 188.52 (145.68)1.5T, MRSI, long TEOnly malesPositive correlation of NAA levels with the amount of cerebellar GM in control participants, but not in the CSP 
  Cho15 control participants, age: 37.47 (11.62)AVNo reliability estimatesInverse correlation of NAA levels & ageHigh NAA & low Cr levels for control participants & CSP
     Individual PVCAdjusting for age-related declines in NAA, eliminated the correlation of NAA & ID in CSP 
de la Fuente-Sandoval et al. (2011)Dorsal-caudate

Glu[UPWARDS ARROW] (in caudate of both patient groups)

Glu/Gln (caudate FEP)

18 drug-naïve subjects with prodromal symptoms for schizophrenia, age: 19.56 (3.46)

3T

SVS

short TE

Equal gender distribution Two patient groups versus one control group
 Cerebellum

NAA[UPWARDS ARROW]

(P)Cho[UPWARDS ARROW]

(in both patients & both regions)

18 drug-naïve FEP, age: 23.44 (4.93)AVFWHM  
  (P)Cr, mI40 control participants, age: 21.83 (4.47) PVC  
Miyaoka et al. (2005)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (both)15 SP with Gilbert's syndrome (GS) & 15 without1.5T, PRESS, SVS, short TEEqual gender distributionSchizophrenia with GS more severely & affected (also in BG & vermis) 
 Left BG

Cho

mI

GS (wGS), all medicated, age: 32.5 (10.7)/34 (11), ID: 1.3 (1.8)/1.7 (2.1)Cr ratiosNo reliability estimates  
 Vermis 15 control participants, age: 41.7 (14.7) No PVC  
Table 12. 1H-MRS studies of the corpus callosum in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. AV, absolute metabolite values; CSP, chronic schizophrenia patients; FEP, first-episode patients; HRS, subjects at a very high-risk of developing schizophrenia; ID, illness duration in months; PVC, partial volume correction; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; STEAM, Stimulated Echo Acquisition Mode; DO, Disease Onset; FWHM, full widths at half maximum; TE, echo time; T, tesla.

Aydin et al. (2007)Superior/posterior genu of the corpus callosumNAA[DOWNWARDS ARROW], T2B[UPWARDS ARROW] (in CSP & FEP)16 medicated CSP, age: 29.31 (11.41), ID: 83.25 (68.65)1.5T, STEAM, SVS, long & short TEGender covariateDecreased callosal NAA correlated with psychopathology in patients 
  

Cr

Cho

12 drug-naïve FEP, age: 25.5 (5.76)AV

SNR

FWHM < 0.1

 All FEP drug-naïve at admission, but only two FEP drug-naïve until first measurement
   28 control participants, age: 27 (9.46) No PVC  
Aydin et al. (2008)Superior/posterior genu of the corpus callosum

NAA[DOWNWARDS ARROW]

T2B[UPWARDS ARROW]

(HRS & FEP)

17 HRS individuals, age: 19.58 (3.24), DO: 18.23 (3.13)1.5T, STEAM, SVS, long & short TE,Gender covariateNAA concentrations correlated negatively with severity of negative symptoms in HRS subjects & FEP 
  

Cr

Cho

14 mainly medicated FEP, age: 25.00 (5.46), DO: 24.28 (5.13)AV

SNR

FWHM

No difference in T2 relaxation times of NAA between patients & control participants 
   Two control participants groups matched for age to the two treatment group PVCCallosal NAA concentration of HRS subjects significantly higher than in FEP 

1H-MRS findings in schizophrenia

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

Temporal lobes

The temporal lobes were one of the primary areas found to be decreased in volume, and fractional anisotropy by a recent diffusion tensor imaging meta-analysis (Ellison-Wright and Bullmore 2009) and by two voxel-based morphometry meta-analyses (Honea et al. 2005; Chan et al. 2011). Further evidence implicating the temporal lobe as a primary site of pathologic involvement in schizophrenia comes from a twin-study revealing volume decreases in the temporal and frontal lobes in patients with schizophrenia and in their unaffected co-twins. Yet, two large 1H-MRS studies did not find abnormal metabolite concentrations of NAA, Cr and Cho in the superior temporal area in chronic schizophrenia patients (CSP), their siblings and in at-risk syndrome patients that later on converted to schizophrenia (Callicott et al. 1998; Jessen et al. 2006). Unaffected NAA levels were also seen in the superior temporal gyrus of schizophreniform spectrum patients (Bertolino et al. 2003) and CSP (Bertolino et al. 2001), in the temporal lobes of CSP (Kegeles et al. 2000; Weber-Fahr et al. 2002; Szulc et al. 2004, 2007, 2011; Shimizu et al. 2007; Wood et al. 2008), and in first-episode patients (FEP) (Berger et al. 2008; Galinska et al. 2009). Only Berger et al. (2008), Wood et al. (2008), and Weber-Fahr et al. (2002) used AVs and executed PVC. Some studies found reductions in NAA levels: in medicated CSP (Fukuzako 2000; Basoglu et al. 2006; Tang et al. 2007), unmedicated FEP (Basoglu et al. 2006), and CSP (Fukuzako 2000; Chang et al. 2007; Szulc et al. 2011) – however, none of the aforementioned studies assessed tissue differences.

Unchanged temporal Glx levels were found in drug-naïve schizophreniform spectrum patients (Wood et al. 2008), CSP (Kegeles et al. 2000; Szulc et al. 2004, 2011) and FEP (Berger et al. 2008; Wood et al. 2008; Galinska et al. 2009). Schizophreniform spectrum patients (Bertolino et al. 2003), and elderly schizophrenia patients with cognitive decline (Chang et al. 2007) showed increased temporal Glx levels, whereas a decrease of temporal Glx/Cr after anti-psychotic treatment was seen in CSP (Szulc et al. 2011).

Decreased mI levels were found in elderly schizophrenia patients (Chang et al. 2007), whereas no change was found in FEP (Berger et al. 2008; Galinska et al. 2009) or schizophrenia-spectrum patients (Wood et al. 2008), however, the latter did not mention any reliability estimates and included six depressed subjects with psychotic symptoms, as can be inferred from Table 1. No changes were seen in Cr and Cho levels of schizophrenia patients.

In sum, the majority of studies found unaltered temporal metabolite levels in schizophrenia patients, yet, most studies lack PVC and reliability estimates.

Hippocampus

Impaired hippocampal functioning in schizophrenia has been implicated by two recent meta-analyses assessing volume (Vita et al. 2006) and function (Fusar-Poli et al. 2007). Yet, no change was seen in hippocampal metabolite levels of drug-naïve subjects experiencing prodromal symptoms of schizophrenia as compared to control participants (Stone et al. 2009). The CRLBV, SNR, and FHWM were reported, relaxation-time differences were accounted for, and partial volume as well as medication effects was assessed in this study. However, the authors noted that Glu could not be measured reliably in the hippocampus, because of increased FWHM and thus decreased spectral resolution. Another study (Wood et al. 2008) found unchanged temporal metabolite levels in FEP and drug-naïve schizophreniform spectrum patients as compared to control participants. Even after a 12 week reassessment of seven drug-naïve patients that underwent treatment no changes were found. Although CRLBV, SNR and FHWM were specified, it was not controlled for gender (80% males in the drug-naïve patients vs. 60% males in the control group) and partial volume effects. Furthermore, control group values of NAA and mI were much lower than the range suggested using healthy participants (Hurd et al. 2004).

Similarly, NAA was unaffected in medicated and unmedicated CSP (Kegeles et al. 2000; Bertolino et al. 2001; van Elst et al. 2005; Klar et al. 2010), and in medicated and unmedicated FEP (Berger et al. 2008; Olbrich et al. 2008; Rusch et al. 2008; Wood et al. 2008; Klar et al. 2010; Hasan et al. 2011; He et al. 2012). Note that three studies (van Elst et al. 2005; Olbrich et al. 2008; Rusch et al. 2008) used the same participants and included extremely low Glu, Gln, and mI control group values. Contrasting the aforementioned studies, FEP and drug-naïve schizophrenia patients showed decreased NAA/Cr ratios compared to medicated FEP, which normalized upon pharmacotherapy (Fannon et al. 2003). As can be inferred from Table 2, it was not corrected for partial volume differences but SNR were assessed. Furthermore, a large MRSI study found reduced NAA/Cr ratios in the hippocampal area of patients with schizophrenia and their unaffected siblings compared to control participants (Callicott et al. 1998). In agreement, three other MRSI studies (Callicott et al. 2000; Weber-Fahr et al. 2002; Bertolino et al. 2003) and two SVS (Fukuzako 2000; Miyaoka et al. 2005), reported decreased hippocampal NAA levels in medicated CSP. Solely one study conducted a sophisticated correction for the different tissues and applied selective pulses for chemical shifts of metabolites measured (Weber-Fahr et al. 2002). NAA levels were solely reduced after execution of this correction. Callicott et al. (2000) included thirty males and only six females (vs. 28 females and 45 males in the control group) without including an additional gender control in their analysis or matching participants, see also Table 2.

Glutamertergic reductions have only been found by one study assessing subjects at-risk for developing schizophrenia (Bloemen et al. 2011). Most studies found increases in hippocampal Glu and Gln, as in medicated FEP (Olbrich et al. 2008; Rusch et al. 2008), in medicated and ethyl-eicosapentaenoic acid treated FEP (Berger et al. 2008), in at-risk subjects (Stone et al. 2009), and in CSP (van Elst et al. 2005). It should be mentioned that Olbrich et al. (2008) found only a trend for increased Glu in the left hippocampus, used a small sample, and assessed the same participants as other authors before (van Elst et al. 2005; Rusch et al. 2008), lacking a gender control - with very low reference Glu/Gln values and higher Gln than Glu values. See for more details a correspondence on that issue (Théberge et al. 2007a,b) and Table 2. Two studies found unaltered hippocampal Glx levels in medicated (Hasan et al. 2011) and unmedicated FEP (Wood et al. 2008) and in schizophrenia spectrum patients (Wood et al. 2008).

Increased mI levels of the left anterior hippocampal-amygdalar region were solely found in a mainly drug-naïve and male sample with increased S100β serum levels (an astrocytic protein, indicating astroglial function) (Rothermundt et al. 2007). Only one study found increased Cr and Cho in twins with schizophrenia compared with healthy controls and compared with their co-twin (Lutkenhoff et al. 2010). All other studies found unchanged hippocampal Cho (Callicott et al. 2000; Bertolino et al. 2001, 2003; Weber-Fahr et al. 2002; van Elst et al. 2005; Miyaoka et al. 2005; Olbrich et al. 2008; Stone et al. 2009; Hasan et al. 2011; He et al. 2012) and Cr levels (Callicott et al. 1998; Weber-Fahr et al. 2002; van Elst et al. 2005; Olbrich et al. 2008; Rusch et al. 2008; Stone et al. 2009; He et al. 2012). Also mI levels were unaltered in all studies (Fannon et al. 2003; van Elst et al. 2005; Miyaoka et al. 2005; Rusch et al. 2008; Wood et al. 2008; Stone et al. 2009; Lutkenhoff et al. 2010; Hasan et al. 2011).

Summing up, the majority of studies found increased hippocampal Glx, and decreased NAA levels, for the latter solely in CSP.

Thalamus

Alterations in volume and shape of the thalamus have not only been implicated in CSP and FEP (Csernansky et al. 2004; Ellison-Wright and Bullmore 2009) but also in their siblings (Harms et al. 2007), and in familial schizophrenia subjects, which are assumed to experience more serious structural alterations than the sporadic type (Byne et al. 2009; Lui et al. 2009). Consistent reductions in NAA were found by the majority of H-MRS studies, including three MRSI studies in male CSP (Deicken et al. 2000; Jakary et al. 2005; Martinez-Granados et al. 2008), and five SVS (Omori et al. 2000; Auer et al. 2001; Basoglu et al. 2006; Szulc et al. 2007, 2011)assessing FEP and medicated patients with schizophrenia. Furthermore, male childhood onset schizophrenia patients had significantly lower NAA levels compared to female patients and male control participants (O'Neill et al. 2004). Two studies (Auer et al. 2001; Jakary et al. 2005) corrected for partial volume effects, and one study (Jakary et al. 2005) applied a thalamic tissue volume mask to define the thalamic tissue more precisely, see also Table 3. Others (Deicken et al. 2000) did not perform PVC but assessed voxel of the mediodorsal thalamus (MDT) on two occasions to obtain intra-rater reliabilities of 0.99 and hightest-retest reliabilities. Supporting the aforementioned studies, two studies in subjects with a high genetic risk of developing schizophrenia, found significantly reduced thalamic NAA levels (Stone et al. 2009; Yoo et al. 2009), compared to control participants. Subjects treated with typical neuroleptics had significantly reduced NAA/Cr ratios in the left thalamus as compared to atypically treated patients and control participants (Szulc et al. 2007), but it was not controlled for partial volume effects, neither were reliability measures mentioned. The same authors found a trend for increased thalamic NAA/Cr ratios in CSP after treatment with neuroleptic medication (Szulc et al. 2011). However, the medicational switch was non-random and again no reliability estimates were given and no PVC was conducted. Yet a few studies found no change, as in CSP (Callicott et al. 1998, 2000; Bertolino et al. 2001; Théberge et al. 2002, 2003; Yasukawa et al. 2005; Aoyama et al. 2011), schizophreniform disorder patients (Bertolino et al. 2003), and FEP (Szulc et al. 2004; Théberge et al. 2007a,b; Galinska et al. 2009).However, seven of ten unchanged NAA findings are from just three authors (Callicott et al. 1998, 2000; Bertolino et al. 2001, 2003; Théberge et al. 2002, 2003, 2007a, b), which decreases the external reliability of their findings.

Increased levels of Gln were found in the left MDT of drug-naïve CSP (Théberge et al. 2003), and of FEP after anti-psychotic treatment for 30 months (Théberge et al. 2002, 2007a, b). Furthermore, ratios were inversely correlated with poor social functioning and with GM losses in schizophrenia. Similarly, Glx/Cr was increased in FEP and correlated negatively with negative symptoms (Szulc et al. 2004). Yet, another study found decreased thalamic Gln/H2O ratios in schizophrenia patients as compared to control participants, and Glu ratios also decreased after 80 months of pharmacological treatment (Aoyama et al. 2011). Similarly, at-risk subjects for schizophrenia showed decreased Glu levels that correlated negatively with thalamic GM volume (Stone et al. 2009). Some studies found no Glx changes in the thalamus of medicated FEP (Galinska et al. 2009), and in CSP (Szulc et al. 2011), however, only Bustillo et al. (2008) assessed volume differences and mentioned reliability estimates, see Table 3.

The majority of studies found unaltered Cho, Cr, and mI levels in schizophrenia patients; yet one study found decreased mI/Cr ratios in medicated schizophrenia patients (Yasukawa et al. 2005) and high-risk subjects showed also decreased thalamic Cho and Cr (Yoo et al. 2009).

Taken together, the majority of studies indicate a decrease in thalamic NAA and an increase in thalamic Glu and Gln levels, although the latter finding is mainly based on one study group.

Basal ganglia

The basal ganglia show strong connections with thalamic, cerebral, and brainstem areas. Shape abnormalities have been reported in individuals with schizophrenia and to a lesser degree in their unaffected siblings (Mamah et al. 2008). Furthermore, the striatum was found to be one of the key regions of reduced GM in schizophrenia (Fornito et al. 2009). Considering the dopaminergic abnormalities found in schizophrenia, it seems a likely area prone to alterations. Mostly the caudate nucleus has been the focus of research, with dopamine D(2) receptor up-regulations in unaffected monozygotic co-twins as compared to unaffected dizygotic co-twins and healthy control twins (Hirvonen et al. 2005). It should be mentioned that the quality of spectra of the basal ganglia are generally poorer, given iron and copper depositions, leading to more heterogeneity of the magnetic field inside the VOI, and hence to larger line widths.

A 3T study showed left sided reductions in GABA/Cr ratios in CSP, yet no tissue corrections were executed and no reliability estimates were given (Goto et al. 2009), see Table 4. Another study did not find any GABA alterations in CSP (Tayoshi et al. 2010), CRLBV were specified and CSF corrections were executed. Also no changes in GABA concentrations were seen in 16 drug-naïve FEP treated with atypical anti-psychotic drugs and reassessed after 6 months as compared to control participants (Goto et al. 2010).

Increased Glu levels were found in the caudate nucleus of medicated and unmedicated FEP and in CSP, independent of pharmacotherapy (de la Fuente-Sandoval et al. 2009, 2011). Yet, two studies found unaltered Glx levels in the basal ganglia of medicated CSP (Block et al. 2000; Tayoshi et al. 2009).

Decreased NAA levels have been found by three authors (Bustillo et al. 2008; de la Fuente-Sandoval et al. 2009; Goto et al. 2011), assessing never or minimally treated recent-onset FEP before treatment with anti-psychotic medication. Only one study found increased caudal NAA levels in drug-naïve FEP with prodromal symptoms (de la Fuente-Sandoval et al. 2011). This contrasts with no metabolite alterations in childhood-onset patients (O'Neill et al. 2004), simple schizophrenia patients (Ohara et al. 2000), schizophrenia patients with and without Gilbert's Syndrome (Miyaoka et al. 2005), medicated schizophreniform spectrum patients (Bertolino et al. 2003), CSP (Callicott et al. 1998, 2000; Block et al. 2000; Bertolino et al. 2001; Ando et al. 2002; Tayoshi et al. 2009, 2010), schizophrenia patients with tardive dyskinesia (Ando et al. 2002), unmedicated and medicated FEP (Fannon et al. 2003; Gimenez et al. 2003), yet for the latter, there was only a trend (= 31, = 0.052) for increased ratios of NAA/Cho in unmedicated male FEP (Gimenez et al. 2003) and Tayoshi et al. (2010) found solely decreased Cr and NAA levels in male versus female schizophrenia patients.

One study found decreased caudate Cr concentrations in minimally treated recent-onset patients (ROP) that were independent of treatment (Bustillo et al. 2008). Elevated Cho levels were seen in the head of the caudate of 11 childhood-onset schizophrenia subjects (O'Neill et al. 2004), in drug-naïve FEP (de la Fuente-Sandoval et al. 2011) and in the lenticular nucleus of CSP with and without tardive dyskinesia (Ando et al. 2002). This is opposed to several studies finding unaltered Cr (Fannon et al. 2003; Tayoshi et al. 2009, 2010), Cho (Callicott et al. 1998, 2000; Bertolino and Weinberger 1999; Block et al. 2000; Ohara et al. 2000; Bertolino et al. 2003; Fannon et al. 2003; Miyaoka et al. 2005; Bustillo et al. 2008; de la Fuente-Sandoval et al. 2009; Tayoshi et al. 2009), and mI levels (Block et al. 2000; Miyaoka et al. 2005; Bustillo et al. 2008; Tayoshi et al. 2009; de la Fuente-Sandoval et al. 2011) in schizophrenia patients as compared to control participants.

In conclusion, most studies show unaltered NAA, Cr, Cho, and mI levels, Glx and GABA alterations are lacking sufficient studies to reliably state metabolite levels.

Dorsolateral prefrontal cortex

All schizophrenia subtypes have prominent impairments in frontal executive functioning, which seems to precede illness onset (Callicott et al. 2003; Fusar-Poli et al. 2007; Woodward et al. 2009). Yet, no change in metabolites could be found in the dorsolateral prefrontal cortex (DLPFC) of subjects with high genetic risk of schizophrenia (defined as having at least two relatives with schizophrenia) (Yoo et al. 2009), medicated patients with chronic deficit-syndrome schizophrenia (Sigmundsson et al. 2003) and medicated CSP (Tang et al. 2007; Kegeles et al. 2012), as compared to control participants. This contrasts with several studies showing decreased NAA levels in medicated CSP (Callicott et al. 2000; Molina et al. 2005; Ohrmann et al. 2007; Sanches et al. 2008), FEP (Molina et al. 2005; Stanley et al. 2007; Zabala et al. 2007), and in medicated schizophreniform spectrum patients (Bertolino et al. 2003). Correlations between decreased NAA levels and ID (Molina et al. 2005), negative symptoms (Callicott et al. 2000; Bertolino et al. 2003), verbal learning and memory performance, and with the severity of symptoms in schizophrenia patients (Bertolino et al. 2003; Ohrmann et al. 2007) were reported. In contrast, prefrontal NAA/Cr was shown to increase upon pharmacotherapy in CSP (Bertolino et al. 2001), yet no PVC was executed and no control group was included.

Three studies assessing Glu, Gln, and Glx in the DLPFC of schizophrenia patients found elevated levels in medicated FEP and CSP as compared to control participants (van Elst et al. 2005; Olbrich et al. 2008; Rusch et al. 2008), yet two of them used a paranoid-subtype sample and all used the same control group, showing very low Glu and Gln levels (see Table 5, the correspondence named earlier and reference values adapted from Hurd et al. 2004). Lower Glx levels were seen in CSP as compared to control participants and as compared to drug-naïve FEP (Ohrmann et al. 2007). None of the aforementioned studies corrected for volume differences, only van Elst et al. (2005) corrected for CSF. Stanley et al. (1996) compared drug-naïve with acute and chronic medicated patients, and found unaltered Glu/Gln levels in FEP, but increased Gln levels in chronic medicated patients and decreased Gln in acutely medicated patients as compared to off medication. Three additional cohorts of patients found unchanged Glx levels: (i) in drug-naïve and medicated patients compared to controls (Kegeles et al. 2012); (ii) in high risk subjects compared to controls (Yoo et al. 2009); (iii) and in medication-naïve FEP compared to control participants (Ohrmann et al. 2007).

Decreases in Cr and Cho were only found in one study of CSP (Ohrmann et al. 2007), as compared to control participants, whereas no change was found in most other studies (Callicott et al. 2000; Bertolino et al. 2001, 2003; Sigmundsson et al. 2003; van Elst et al. 2005; Molina et al. 2005; Stanley et al. 2007; Zabala et al. 2007; Ohrmann et al. 2008; Olbrich et al. 2008; Rusch et al. 2008; Sanches et al. 2008; Yoo et al. 2009).

In conclusion, decreased NAA levels seem a predominant alteration in DLPFC functioning in schizophrenia, Glu/Gln seems to be unaltered in drug-naïve patients.

Medial prefrontal cotrtex

Adolescent siblings of schizophrenia patients were found to have increased Glx/Cr ratios in the medial prefrontal cortex (mPFC) as compared to control participants (Tibbo et al. 2004). However, no reliability estimates were noted, as shown in Table 6. Also in drug-naïve ROP, increased levels of Gln were found, although the resolution was probably too low to discern Glu/Gln and GABA levels (Bartha et al. 1999). A3T study with J-edited spin-echo difference found elevations in absolute values of GABA and Glx in unmedicated patients compared with control participants, whereas medicated patients showed no change (Kegeles et al. 2012). In contrast, a study of twin pairs discordant for schizophrenia, both twins (unaffected co-twin and twin with schizophrenia) exhibited reduced levels of Glu in the GM of the mPFC (Lutkenhoff et al. 2010). Considering that the former study had a 20-year younger sample (Bartha et al. 1999), a time-dependent neurotoxicity theory of glutamate would be supported, with increased Gln in the onset of the disease and later on reductions in Glu because of neurotoxicity.

From the mPFC studies reviewed here, only two studies found decreased NAA levels in male schizophrenia patients (Do et al. 2000) and in deficit-syndrome schizophrenia subjects (Delamillieure et al. 2000). All others (Bartha et al. 1999; Bustillo et al. 2011, Lutkenhoff et al. 2010; Purdon et al. 2008; Reid et al. 2010; Tibbo et al. 2004) found no change. Choline, Cr, and mI levels were found unchanged by all studies, see Table 6.

Taken together, mPFC Glx levels seem to be increased, whereas the majority of studies show unaltered NAA, Cho, Cr, and mI levels in schizophrenia patients.

Other frontal areas

One sibling study (Block et al. 2000) found no change in left frontal NAA/Cho ratios in unaffected and affected family members, but a significant decrease in NAA/Cho ratios of CSP as compared to healthy control participants. This study was assigned by a previous meta-analysis (Steen et al. 2005) to have the highest power (92%) in detecting differences in NAA concentrations of all studies reviewed, yet no reliability measures were given and it was not controlled for the unequal gender distribution, nor for partial volume effects. Furthermore, patients with schizoaffective disorder were included in the CSP group, see Table 7. Decreased frontal NAA levels were also seen in medicated early and late at-risk syndrome patients, in medicated schizophrenia patients (Jessen et al. 2006), in CSP (Jessen et al. 2006; Tunc-Skarka et al. 2009; Szulc et al. 2011), in FEP (Tunc-Skarka et al. 2009), and in the prefrontal WM of elderly patients with schizophrenia (Chang et al. 2007). This contrasts with a number of studies finding no change in frontal NAA levels of FEP (Fannon et al. 2003; Szulc et al. 2004; Galinska et al. 2009; Goto et al. 2011), CSP (Omori et al. 2000; Bertolino et al. 2001; Steel et al. 2001; Szulc et al. 2004, 2007; Rowland et al. 2009; Ongur et al. 2010a), ROP (Bustillo et al. 2008), young and old schizophrenia patients (Bustillo et al. 2011), siblings (Callicott et al. 1998), childhood-onset schizophrenia patients (O'Neill et al. 2004), schizophreniform disorder patients (Bertolino et al. 2003), and schizophrenic twins and their discordant co-twins (Lutkenhoff et al. 2010).

No frontal Glu alterations have been detected in FEP (Galinska et al. 2009), ROP (Bustillo et al. 2008, 2010), young and old schizophrenia patients (Bustillo et al. 2011), CSP (Block et al. 2000; Rowland et al. 2001; Szulc et al. 2011; Ota et al. 2012), deficit-syndrome schizophrenia patients (Rowland et al. 2001), and patients with an acute exacerbation of symptoms (Ota et al. 2012). An increase in Glx levels was found for elderly patients with schizophrenia compared to control participants (Chang et al. 2007), however, that sample also included three schizophrenia patients with comorbid schizoaffective disorders and all patients suffered from cognitive decline, see Table 7. Decreased Glu was found in twins discordant for schizophrenia, together with an age-related glutamatergic increase (Lutkenhoff et al. 2010). Only one study assessed GABA levels in FEP finding no change, however, GABA was determined at 3T and no tissue differences were accounted for (Goto et al. 2010).

Elderly CSP with cognitive decline showed decreased levels of mI and an age-dependent Cho decrease (Chang et al. 2007), whereas increased Cho levels were detected in patients with childhood-onset schizophrenia (O'Neill et al. 2004). However, none of the other studies found any alterations in Cr, Cho, or mI – see Table 7 for all details.

In summary, the majority of frontal studies indicate unaltered NAA and Glx levels in schizophrenia, evidence for Cho, Cr, and mI is too sparse to draw valid conclusions.

Parietal lobes

There have been only a few parietal spectroscopy studies conducted thus far, including four studies examining Glx levels. One 3T study found no change in any metabolite-including Glx levels of patients with deficit-syndrome and non-deficit syndrome schizophrenia (Rowland et al. 2009). Similarly, unchanged Glx levels were seen in the parieto-occipital corext of CSP (Ongur et al. 2010a). On the contrary, schizophrenia patients with an acute exacerbation of symptoms showed increased inferior parietal Glx/Cr ratios as compared to patients without an exacerbation and to controls (Ota et al. 2012). However, young schizophrenia patients showed decreased Glx/Cr ratios in the lateral and medial parietal lobes that correlated positively with overall cognitive performance (Bustillo et al. 2011). Two control groups and reliability estimates were provided and PVC was executed, however, the gender distribution was unequal and ratios were calculated for the pooled tissues of the parietal and frontal lobes.

Two studies assessed parietal GABA/Cr ratios: Ongur et al. (2010a, b) found increased GABA/Cr in CSP, whereas Goto et al. (2009) found no change in CSP. As can be inferred from Table 8, the former executed PVC and used a 4T scanner, whereas the latter did not execute PVC and used a 3T scanner.

Most studies found unchanged NAA levels in acutely-ill, medicated CSP (Auer et al. 2001), in young and old schizophrenia patients (Bustillo et al. 2011), in medicated CSP (Sanches et al. 2008; Ongur et al. 2010b), in childhood onset schizophrenia (O'Neill et al. 2004), in patients with and without an acute exacerbation of schizophrenia symptoms (Ota et al. 2012), and in deficit and non-deficit syndrome schizophrenia patients (Rowland et al. 2009). Only one study showed decreased NAA/Cr levels in the parieto-occipital cortex of FEP as compared to control participants (Goto et al. 2011).

One study reported increased parietal Cr, Cho, and mI in acutely ill CSP (Auer et al. 2001), whereas another study in found decreased Cr levels in medicated CSP as compared to control participants and bipolar disorder patients (Ongur et al. 2009). Both studies corrected for volume differences and mentioned reliability estimates, yet, only the latter study used a 4T scanner – see Table 8 for more details. Furthermore, an age-related decline in Cr levels was found solely in the control group. All other studies did not find any parietal changes in Cr, Cho, and mI.

In sum, there is some evidence for unaltered parietal NAA, Cr, Cho, and mI levels, yet replications and a pure parietal delineation are needed. Glx findings remain inconclusive, because of only few studies based on different schizophrenia subtypes and different VOIs included.

Occipital lobe

Of the few occipital studies in schizophrenia, most found no metabolite alterations (Callicott et al. 1998, 2000; Bertolino et al. 2001, 2003; O'Neill et al. 2004; Tang et al. 2007; Bustillo et al. 2008; Olbrich et al. 2008; Goto et al. 2009, 2010). Reduced NAA levels were only seen in FEP (Goto et al. 2011), and in the occipital WM of medicated elderly CSP (Chang et al. 2007). Elderly schizophrenia patients with cognitive decline showed also increased Glx and decreased mI levels, but in comparison to low Glx control group values (Chang et al. 2007). Furthermore, both studies did not control for tissue differences – see Table 9.

Creatine was reduced in the parieto-occipital cortex of acute-episode CSP (Ongur et al. 2010a, b), but no change was reported in any other study assessing Cr, Cho, and mI. GABA/Cr ratios were unchanged in CSP (Goto et al. 2010), but reduced by an approximately 10% in the visual cortex of 13 male subjects with schizophrenia as compared to 13 healthy control subjects (Yoon et al. 2010).

The occipital lobe seems to be quite unaffected by schizophrenia, yet, because of the few studies conducted thus far it is difficult to drawn reliable conclusions.

Cingulate gyrus

No metabolic differences were seen in the anterior and posterior cingulate gyrus (ACG/PCG) of medicated CSP, their siblings and control participants (Callicott et al. 1998). In line with this, unaltered metabolite levels were also found in drug-naïve FEP (Bartha et al. 1999; Théberge et al. 2002, 2007a, b), in minimally treated patients (Bustillo et al. 2010; Aoyama et al. 2011), medicated CSP (Callicott et al. 2000; Delamillieure et al. 2000; Théberge et al. 2003; Tayoshi et al. 2010; Hardy et al. 2011), subjects in an at-risk mental state for schizophrenia (Stone et al. 2009), subjects with a high genetic risk of developing schizophrenia (Yoo et al. 2009), in schizophreniform spectrum patients (Bertolino et al. 2003), in schizophrenia patients with and without an acute exacerbation (Ota et al. 2012) and in CSP (Ongur et al. 2010a, b). However, many studies found decreased NAA levels as in CSP (Sanches et al. 2004; Jessen et al. 2013), in early and late at-risk syndrome patients that converted to schizophrenia (Jessen et al. 2006), in medicated CSP (Shimizu et al. 2007), in medicated ROP (Yasukawa et al. 2005), in minimally treated schizophrenia subjects with no changes following anti-psychotic treatment (Bustillo et al. 2010), and in deficit-syndrome schizophrenia patients as compared to non-deficit syndrome patients and control participants (Delamillieure et al. 2000). However, the deficit syndrome schizophrenia patient group was quite small (see Table 10), and age, disease status, reliability measures and PVC were not mentioned. Furthermore, a trend towards decreased NAA/Cr ratios in CSP was found, which correlated positively with the blood oxygen level-dependent signal (Reid et al. 2010). Also decreased NAA levels were seen in atypically versus typically medicated patients (Braus et al. 2002), whereas Ende et al. (2000) found the lowest NAA levels in typically treated patients, with both groups (typically and atypically medicated CSP) showing significant lower NAA levels and an age-dependent correlation, that might explain the lower NAA values in the typically treated group – see Table 10.

Sanches et al. (2008) found that patients with schizophrenia and low skin conductance showed lower NAA/(Cr + Cho) ratios in the left ACG than schizophrenia patients with normal skin conductance (Sanches et al. 2008). Yamasue et al. (2002) found solely decreased NAA/Cho ratios but no change in NAA/Cr ratios in medicated CSP as compared to control participants. This effect was only significant after PVC; all of the former studies did not apply PVC (Braus et al. 2002; Sanches et al. 2004; Yasukawa et al. 2005; Jessen et al. 2006; Shimizu et al. 2007). With exception of Braus et al. (2002) no other study found elevated NAA levels. As can be inferred from Table 10, all studies conducting PVC and reporting AVs (Bartha et al. 1999; Théberge et al. 2002, 2003, 2007a, b; Stone et al. 2009; Yoo et al. 2009; Tayoshi et al. 2010), have not found any alterations in cingular NAA concentrations schizophrenia patients.

Two 4T studies found significant increases in Gln in the left ACG of FEP (Théberge et al. 2002), and significantly decreased Glu and Gln in mainly male medicated CSP (Théberge et al. 2003) as compared to control participants. No change was seen after ten and 30 months reassessment of the FEP. Minimally treated schizophrenia subjects showed higher Gln/Glu in the ACG before anti-psychotic treatment and after treatment (Bustillo et al. 2010). Similarly, two authors (Bartha et al. 1999; Stone et al. 2009) reported significantly increased Gln in the ACG of drug-naïve, ROP and at-risk mental state subjects, respectively. Significantly decreased Glu levels were seen in paranoid CSP as compared to control participants (Tayoshi et al. 2009). Furthermore, male patients had significantly decreased Cr, Glu, mI, and NAA levels as compared to control subjects, underlining the importance of an adequate gender control. There was no evidence of decreased cingulate Glx/Cr levels in CSP, but a significant negative correlation between Glx/Cr levels and negative symptoms was noted (Reid et al. 2010). Some studies found no change in cingular Glx levels, as in initially drug-naïve patients (Aoyama et al. 2011), CSP (Ongur et al. 2010b; Reid et al. 2010; Jessen et al. 2013), and in at-risk individuals (Yoo et al. 2009).

Two studies assessed cingular GABA levels in CSP, one found significantly elevated GABA/Cr ratios that correlated positively with Glu/Cr (Ongur et al. 2010a, b), whereas the other found no change in GABA levels (Tayoshi et al. 2010). Both assessed tissue differences and mentioned CRLBV, yet Tayoshi et al. (2010) used AV, whereas Ongur et al. (2010a, b) did not.

Cingular mI levels were reduced in th ACG of paranoid CSP (Tayoshi et al. 2010), and in medicated ROP (Yasukawa et al. 2005). This contrasts with several other studies reporting no change in mI levels (Delamillieure et al. 2000; Théberge et al. 2003, 2004, 2007a, b; Stone et al. 2009; Yoo et al. 2009; Aoyama et al. 2011; Jessen et al. 2013).

Decreased Cho/Cr ratios were found in the ACG of medicated ROP (Yasukawa et al. 2005), and in medicated CSP (Yamasue et al. 2002). This conflicts with most other studies reporting no change (Bartha et al. 1999; Delamillieure et al. 2000; Ende et al. 2000; Auer et al. 2001; Braus et al. 2002; Jessen et al. 2006, 2013; Shimizu et al. 2007; Ohrmann et al. 2008; Sanches et al. 2008; Stone et al. 2009; Tayoshi et al. 2009, 2010; Yoo et al. 2009; Ongur et al. 2010a, b; Hardy et al. 2011), and two studies reporting increased Cho levels in at-risk syndrome patients that converted to schizophrenia (Jessen et al. 2006), and in childhood-onset schizophrenia patients (O'Neill et al. 2004).

Medicated, acute-episode CSP showed reduced Cr levels as compared to control subjects (Ongur et al. 2010a, b). This finding contradicts most other aforementioned studies that found no changes in cingular Cr levels (Bartha et al. 1999; Ende et al. 2000; Braus et al. 2002; Théberge et al. 2002, 2003, 2007a, b; O'Neill et al. 2004; Shimizu et al. 2007; Sanches et al. 2008; Stone et al. 2009; Tayoshi et al. 2009; Yoo et al. 2009; Hardy et al. 2011; Jessen et al. 2013), and one study that found increased Cr levels (O'Neill et al. 2004). Similarly, mI levels were found consistently unaltered, yet one author found decreased mI/Cr ratios in medicated schizophrenia patients with and without Gilbert's syndrome (Yasukawa et al. 2005).

In sum, the majority of studies found increased cingular Glx levels in schizophrenia patients, and when looking at the methodologically rigorous studies, no NAA, Cho, Cr, and mI alterations could be found.

Cerebellum

A limited number of cerebellar studies have been conducted thus far; one was assessing the effects of treatment over a 2-year period in minimally treated ROP (Bustillo et al. 2008), showing no change after treatment, but global reductions in NAA before undergoing medication treatment. Two other studies found reduced NAA levels in the cerebellar cortex and reduced NAA in the vermis of medicated CSP (Ende et al. 2000; Deicken et al. 2001). Deicken et al. (2001) also found reduced Cr levels, whereas Ende et al. (2000) found reduced Cho. As shown in Table 11, both studies included mainly male participants and applied PVC, and one corrected for the mean metabolite values by averaging the tissue-volume-corrected metabolite estimates of all three cerebellar voxel individually (Deicken et al. 2001). Another study, found increased AVs of NAA, mI, Cr, and Cho in both first-episode and ultra-high risk patients (de la Fuente-Sandoval et al. 2011). Unchanged metabolite ratios were found in the vermis of drug-naїve adult-onset schizophrenia patients when compared to control participants (Miyaoka et al. 2005), no PVC and reliability estimates were included.

Taken together, there is some evidence showing cerebellar NAA reductions in CSP, yet further studies are needed for valid conclusions.

Corpus callosum

Thus far, there have been only two recent studies investigating metabolite abnormalities in the superior and posterior genu of the corpus callosum, both by the same author. The main finding was a reduction in NAA levels in CSP and in minimally treated FEP as compared to control participants (Aydin et al. 2007 – Table 12). The other study compared mainly medicated FEP and individuals at ultra-high-risk of developing schizophrenia, and found also decreased NAA concentrations in both patient groups compared to two control groups (Aydin et al. 2008). NAA levels were negatively correlated with psychopathology in the at-risk individuals as well as in the FEP and CSP in both studies. Furthermore, NAA levels were significantly higher in the at-risk individuals as compared to FEP, implicating a progressive decline with continuation of the disease. No change in Cr and Cho was observed, see Table 12.

Pharmacological effects

There have been only four longitudinal studies assessing the effects of medication on NAA levels in schizophrenia (see Table 13 all studies with an asterisk). One author found higher NAA/Cr ratios in the DLPFC of CSP when under treatment as compared to off treatment (Bertolino et al. 2001), yet typical and atypical neuroleptics were analyzed together and no PVC was executed. Others (Bustillo et al. 2008) found global NAA reductions in minimally treated ROP and no change in NAA levels, irrespective of medication type (haloperidol and quetiapine). However, the dropout in Bustillo et al. (2008) was so high at follow-up, that the medication type could not be compared. Another study found that 3 months of treatment with mainly typical neuroleptics eliminated the baseline hippocampal NAA difference of drug-naïve FEP as compared to medicated FEP and control participants (Fannon et al. 2003). These conflicting findings might be attributable to the differences in methodology applied: Bustillo et al. (2008) studied a larger, mainly unmedicated, and 10 years younger sample that was followed up for 2 years. Furthermore, metabolites were measured in AVs and assessed for partial volume effects, whereas the other two studies (Bertolino et al. 2001; Fannon et al. 2003) used ratios and no PVC and assessed their sample after 4 weeks and 3 months of pharmacological treatment, respectively. Hence, no straightforward longitudinal results are yet confirmed to verify NAA alterations associated with pharmacological treatment. Similarly, only one longitudinal study assessed Gln and Glu levels in initially drug-naïve FEP after 10 and 30 months of neuroleptic treatment (Théberge et al. 2007a, b). At the start of the treatment Gln in the left ACG and left mediodorsal thalamus was increased but reduced in the thalamus after 30 months of treatment together with widespread GM losses.

Table 13. 1H-MRS studies assessing pharmacotherapy in schizophrenia patients
StudyROIMetabolitesSubjectsMethodsReliability/validityAdditional findingsAdditional limitations
  1. a

    Longitudinal studies.

  2. ARMS, At-risk mental state subjects; AV, absolute metabolite values; COV, Coefficient of variation; CRLBV, cramér rao lower bounds; FEP, first-episode patients; CSP, chronic schizophrenia patients; DO, Disease Onset; ID, illness duration in months; DUI, duration of untreated illness (months); FWHM, full widths at half maximum; HRS, subjects with high genetic risk of developing schizophrenia; 1H-MRS, proton magnetic resonance spectroscopy; WM, white matter; GM, gray matter; OL, occipital lobes; DLPFC, dorsolateral prefrontal cortex; mTL, medial temporal lobes; mPFC, medial prefrontal cortex; CC, cerebellar cortex; ACG, anterior cingulate gyrus; GABA, gamma-aminobutyric acid; Glx, glutamate & glutamine; Glu, glutamate; Gln, glutamine; mI, myoinositol; Cr, creatine; Cho, choline; NAA, N-acetyl-aspartate; PVC, partial volume correction; PFC, prefrontal cortex; PRESS, Point Resolved Spectroscopy; SNR, signal-to-noise ratio; SP, schizophrenia patients; STEAM, Stimulated Echo Acquisition Mode; TL, temporal lobes; TE, echo time.

Bertolino et al. (2001)aBilateral DLPFCNAA/Cr[UPWARDS ARROW] (while medicated)23 CSP assessed twice (off medication for 2 weeks & after 4 weeks medication), age: 36.9 (8.1), ID: 128.4 (70.8)

1.5T

MRSI

long TE

Unequal gender distribution, no control participants  
 Inferior frontal gyrus, superior temporal gyrus, bilateral hippocampusCho/Cr Cr ratiosFWHM but no specified values  
 OL, anterior/posterior cingulate, PF WM, centrum semiovale, putamen, thalamus   No PVC  
Braus et al. (2002)ACGNAA[UPWARDS ARROW] in atyp. Treated10 typically treated CSP, age: 40 (6.2), ID: 174 (60)

1.5 T

PRESS MRSI

long TE

Equal gender distributionAtypically treated CSP had fewer errors on WCST and higher NAA thsn typically treatedSwitch to atypical medication in non-randomized way
  Cr, Cho11 atypically treated CSP, age: 37 (13), ID: 145 (110)AVNo reliability estimatesCorrelation between time treated with atypicals & with higher NAA & better test performanceNo control participants
     No PVC  
Bustillo et al. (2008)aLeft caudate nucleusCr[DOWNWARDS ARROW]32 minimally treated ROP, treated & followed up for 2 years, age: 24.7 (6.9),ID: 41.4 (69)

1.5T PRESS

SVS

short TE

Gender covariateGlobal NAA reductions related to 
 

Left FL

left OL

NAA[DOWNWARDS ARROW] globally before treatment21 control participants, age: 24.7 (5.3)AV

FWHM

SNR

Global cognitive performance in the whole subject sample 
 Right cerebellum

Cho

Glx

mI

  PVCNo difference after treatment 
Ende et al. (2000)VermisNAA[DOWNWARDS ARROW] Cho[DOWNWARDS ARROW]14 medicated CSP, age: 38.9 (7.3)

1.5 T

PRESS MRSI

long TE

Unequal gender distribution, but covariateTrend for Cr reductions in vermis (= 0.06) & CC (= 0.07)No data on concentrations
 CCNAA[DOWNWARDS ARROW]14 control participants, age: 35.6 (3.7)AVNo reliability estimates Unequal gender distribution of control group
 Dentate nucleus, ponsCr  PVC  
Fannon et al. (2003)aLeft hippocampusNAA/Cr[DOWNWARDS ARROW] (in drug-naïve FEP at baseline vs. medicated FEP & control participants)12 drug-naïve FEP, age: 26.1 (5.5) ID: 22.02 (15.41) weeks - reassessed after treatmentPRESS short TE SVSUnequal gender distribution, but covariate usedTreatment eliminated NAA difference3 of the 33 FEP had also schizoaffective disorder
 

Left FL

left BG

Cho/Cr21 medicated FEP, age: 24 (5.8), ID: 49.45 (33.19) weeksCr ratiosSNR but not specifiedBoth patient groups had reduced left hippocampal volume as compared to control participants 
   18 control participants, 25.3 (6.7) No PVC  
de la Fuente-Sandoval et al. (2011)Dorsal-caudateGlu[UPWARDS ARROW] (in caudate of both patient groups)18 drug-naïve subjects with prodromal symptoms for schizophrenia, age: 19.56 (3.46)

3T

SVS

short TE

Equal gender distribution Two patient groups versus one control group
 CerebellumNAA [UPWARDS ARROW] (in both patients & both regions)18 drug-naïve FEP, age: 23.44 (4.93)AVFWHM  
  mI, pCr, pCho[UPWARDS ARROW] (both regions & both patients)40 control participants, age: 21.83 (4.47) PVC  
Miyaoka et al. (2005)Left hippocampusNAA/Cr[DOWNWARDS ARROW] (both)15 SP with Gilbert's syndrome (GS) & 15 without1.5T, PRESS, SVS, short TEEqual gender distribution Schizophrenia with GS more severely & affected (in BG & vermis) 
 Left BG

Cho

mI

GS (wGS), all medicated, age: 32.5 (10.7)/34(11), ID: 1.3 (1.8)/1.7 (2.1)Cr ratiosNo reliability estimates  
 Vermis 15 control participants, age: 41.7(14.7) No PVC  
Szulc et al. (2007)Left FL & left TL 58 CSP split based on treatment (typ/atyp), age: 33.1 (6.9), ID: 112.92(70.8) 1.5T, SVS, short TE, PRESSGender covariateNegative correlation between NAA ratios & age in the FL & between ID & hospitalizations in TL  
 Left thalamus NAA/Cr[DOWNWARDS ARROW] (CSP typical)Control participants, age: 30.2 (5.3)Cr & H2O ratiosNo reliability estimates  
     No PVC  
Théberge et al. (2007a, b)aLeft ACGln[UPWARDS ARROW]16 drug-naïve FEP, reassessment at 10 months & 30 months after treatment with neuroleptics, age: 25 (8)4T, SVS, STEAM, short TEUnequal gender distribution, no gender covariate, no data on control participantsReduced thalamic Gln after 30 months of treatmentThe first MRS assessment of 12 of the FEP & 6 control participants was part of the Théberge et al. (2002) study
 Left medio dorsal ThalamusGln[UPWARDS ARROW]16 control participants, age: 29 (12)AVNo reliability estimates but COV < 75%Small GM reductions after 10 months & widespread GM reductions after 30 months 
  Glu, Cho, Cr, NAA  PVCReduced PL & TL GM correlated with thalamic Gln decline 

Most cross-sectional studies (see Table 13) in this review reported no difference between unmedicated and medicated schizophrenic subjects. Yet, in terms of atypical versus typical medication, the broad consensus found a favorable effect when treated with atypical medication (i.e. an increase in NAA levels with atypical medication) (Ende et al. 2000; Szulc et al. 2007). Furthermore, atypically treated patients showed a positive correlation between treatment time, NAA levels in the ACG, and performance as measured by the Wisconsin-card Sorting Test (Braus et al. 2002). Furthermore, Stanley et al. (1996) found only decreased Gln in patients on medication as compared to off medication whereas CSP showed increased Gln. Similarly, Kegeles et al. (2012) found only decreased Glx and GABA levels in the mPFC of unmedicated patients – implying an effect of pharmacological treatment and disease status on Gln levels.

There were no alterations seen in Cr or Cho levels after treatment with either atypical or typical medication as compared to control participants and untreated patients (Ende et al. 2000; Bertolino et al. 2001; Braus et al. 2002; Fannon et al. 2003; Théberge et al. 2007a, b; Bustillo et al. 2008). Only one study found increased Cho levels in the left thalamus of 21 atypically medicated CSP, however, only one female subject was included (Théberge et al. 2003).

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

One aim of this review was to highlight the underlying methodological problems of 1H-MRS, in an attempt to draw objective conclusions from the often, contradictory findings in the spectroscopy literature on schizophrenia. Given the various limiting factors of cross-sectional and longitudinal studies – such as heterogeneous patient populations, diverse implementation styles, and differences in rigor applied in determining voxel and tissues of interest – it becomes very difficult to compare and evaluate findings based on universal parameters. Therefore, our second aim was to disentangle effects of sample and methodological variations on spectroscopy findings in schizophrenia. In the following a summary of the reviewed studies and proposed practice parameters will be given, followed by a discussion of the main methodological drawbacks in 1H-MRS research.

In agreement with other reviews (Rowland et al. 2001; Steen et al. 2005) a decrease of NAA was found by the broad consensus of studies in the DLPFC, thalamus, and in the hippocampus of primarily CSP but also FEP. For the DLPFC, the majority of studies reported decreased NAA in FEP and CSP, with more pronounced results in CSP - implicating a decline in NAA levels with illness progression (Table 5). In addition to thalamic NAA reductions, increases in Glu and Gln levels were found, although the latter finding is based mainly on one group of authors (Théberge et al., 2002, 2003, 2004, 2007; see Table 3). Declined hippocampal NAA levels were predominantly seen in CSP, whereas an increase in Glx levels was found in FEP and CSP (Table 2), although often more pronounced in FEP – supporting a neurotoxicity role of Glu in schizophrenia.

Examination of the ACG has yielded many contradictory findings in terms of NAA levels, yet when considering only the methodologically meticulous studies, no significant NAA reductions were found. There were several positive results for cingular glutamatergic changes (i.e. mostly increased Gln and decreased Glu). Only two studies found no change in Glu or Gln (Ohrmann et al. 2008; Yoo et al. 2009 – see Table 10). Four of the six studies found increased Gln in drug-naïve patients and in individuals with an at-risk mental state for psychosis. The two studies examining the ACG of CSP (Théberge et al. 2003; Tayoshi et al. 2010 – Table 10) found decreased Glu. The aforementioned increase in Gln in unmedicated and early-stage patients, as compared to a decrease in Gln in CSP, further supports a role of glutamatergic neurodegeneration in schizophrenia. Similarly, initially increased and later on decreased Glx levels in the mPFC support a role for glutamatergic neurotoxicity in schizophrenia.

The temporal lobes, basal ganglia, and the occipital lobe were the few cerebral areas that were found predominantly unaltered in schizophrenia patients. Another consistently replicated finding is the absence of Cho changes in almost all areas reviewed here.

Some brain areas are clearly underrepresented by most articles, namely the cerebellum, the corpus callosum, and the parietal lobes. The only two studies examining the corpus callosum have been positive for NAA reductions in FEP and CSP. However, these studies were conducted by the same author.

When comparing FEP and individuals with CSP, similar metabolite alterations were seen. Yet, the overall tendency shows a progressive worsening, in particular for decreases in NAA, that are far more pronounced in all regions for individuals affected by chronic schizophrenia. The contrary holds for Glx, with most studies showing higher Glx levels in the thalamus, hippocampus, ACG, and frontal areas of FEP than in CSP, which would be again in line with a time-dependent role of glutamatergic neurotoxicity (Farber 2003).

Pharmacological effects on metabolites could not be assessed objectively as the few longitudinal studies remain largely inconclusive. However, cross-sectional evidence implicates atypical neuroleptic treatment in normalizing NAA levels.

In the following seven main methodological limitations will be discussed, including issues concerning (i) Sample variability, (ii) Gender, (iii) PVC and Morphology, (iv) Cr and Cho ratio approach, (v) Relaxation Times, and (vi) Spectral and Spatial Resolution.

Sample variability

One of the reasons for the lack of consistency of spectroscopy findings in schizophrenia is the small number of participants in most of the studies. Many studies only included as few as ten subjects and only nine of seventy studies had a sample size of > 39 which was recommended by a recent meta-analysis (Steen et al. 2005) as a minimum to have 80% power to detect a 10% change in metabolites. As most single centers will not be able to recruit adequate sample sizes, it would be ideal if centers would collaborate to produce powerful studies.

Additional heterogeneity is introduced by using samples that differ in psychotic states (acute vs. remission) as it has been shown that dopaminergic changes can occur acutely with the exacerbation of psychotic symptoms (Kapur et al. 2005; Fusar-Poli et al. 2009; Tost et al. 2010). Given the tight coupling of the dopaminergic system with the glutamatergic system and the neuromodulatory role of dopamine (Carlsson et al. 1999; Seamans and Yang 2004; Tost et al. 2010), it is hypothesized that Glu and Gln levels can change depending on the state of psychosis. This was verified as Gln levels were shown to be directly affected by medication and chronic course of the disease (Stanley et al. 1996). Future research should differentiate between acute and non-acute psychotic schizophrenia patients.

Diagnosis is another key issue that increases heterogeneity in the sample population. Some samples, for example, included patients that had not yet been diagnosed with schizophrenia (Basoglu et al. 2006; Bertolino et al. 2003; – Table 3), or included patients with schizoaffective disorders (Block et al. 2000; Fannon et al. 2003; Berger et al. 2008; Wood et al. 2008; Tunc-Skarka et al. 2009; Woodward et al. 2009). Given that it has been shown that there are differences between metabolite levels in FEP depending on their exact diagnosis later on (Zabala et al. 2007 – Table 5), samples should be stratified a posteriori. Similarly, in people with schizophrenia, assessment of metabolite differences is further complicated by the heterogeneity of the disorder as expressed in the various subtypes (e.g. deficit syndrome schizophrenia, paranoid subtype). Several of the reviewed studies here found metabolite alterations specific to or more severe in certain subtypes, e.g. deficit-syndrome schizophrenia patients (Delamillieure et al. 2000; Yasukawa et al. 2005; Martinez-Granados et al. 2008; Rowland et al. 2009). Given that many studies included mixed samples this might disguise true findings and should be interpreted with caution. Others included two or more treatment groups of differing ages or gender ratios, whereas only including one control group (Block et al. 2000 – Table 4; Szulc et al. 2004 – Table 3; Molina et al. 2005 – Table 5; Basoglu et al. 2006 – Table 3; Ongur et al. 2010a, b – Table 7), making a direct comparison impossible as age and gender are major influences on metabolite distribution (Block et al. 2000; Maudsley et al. 2009; Tayoshi et al. 2009; Tunc-Skarka et al. 2009).

Gender

Another aspect, which warrants further discussion, is gender differences in the pathogenesis of schizophrenia (Narr et al. 2001; Canuso and Pandina 2007; Fujiwara et al. 2008). Male and female individuals with schizophrenia differ in CSF volume and frontal GM composition (Gur et al. 1999; Moreno et al. 2005; Tamagaki et al. 2005; Chen et al. 2007; Reig et al. 2009), neurotransmitter levels and function (Cosgrove et al. 2007), brain volume (Blatter et al. 1995; Reiss et al. 1996; Pfefferbaum et al. 1999; Chen et al. 2007; Cosgrove et al. 2007) and in asymmetry and fissurization of the ACG (Yucel et al. 2001) from healthy participants.

Within this review several studies showed obvious gender differences in metabolite concentrations within the subject group (Fannon et al. 2003; O'Neill et al. 2004; Tayoshi et al. 2010), within the control group (Tayoshi et al. 2010), and between control participants and patients (O'Neill et al. 2004; Tayoshi et al. 2010). Furthermore, in one study, both - drug-naïve and medicated - female FEP showed a reduction in hippocampal NAA/Cr ratios over a 3-month treatment period, whereas male patients exhibited increased NAA/Cr ratios (Fannon et al. 2003). These sex differences are an important finding, considering that (i) most studies did not split their sample by gender (ii) had unequally distributed sex ratios, and (iii) often had unmatched control participants and did not control for that in their analyses.

Partial volume correction and morphology

As the brains of schizophrenic patients differ in morphology and tissue composition it is imperative to assess tissue composition when assessing metabolites. A previous review (Auer et al. 2001) considered the lack of PVC as one of the main factors contributing to the ambiguous 1H-MRS findings. In this review only 36 of 92 studies controlled for tissue differences (often only CSF correction). The need for PVC has been repeatedly demonstrated by various studies showing differences in metabolites levels in gray and WM of healthy human subjects (Michaelis et al. 1993; Stanley et al. 1996; Pouwels and Frahm 1998; Brief et al. 2009). Yet, a recent meta-analysis (Steen et al. 2005) found no apparent gray and WM differences in NAA reductions of schizophrenic patients. However, as patients tend to be compared to healthy control participants; this should not eliminate the need for PVC given the brain tissue and structural variations between patients and healthy control participants. This is supported by one study who found a positive correlation of NAA levels with the amount of cerebellar GM in the control group, but not within CSP (Deicken et al. 2001; –Table 11). Similarly, a correlation was found between the amount of gray and WM and Cho/Cr ratios in schizophrenia patients but not in control participants (Moore et al. 2002; Deicken et al. 2000 – Tables 1 and 3). Two studies could only find differences in NAA (Weber-Fahr et al. 2002; Yamasue et al. 2002), and Cho (Yamasue et al. 2002 – Table 10) values in schizophrenia subjects after correcting for GM, WM, and CSF. Another important correction was performed in one study to counteract imperfect radio frequency pulses by using slice and VOI selective radio frequency pulses (Weber-Fahr et al. 2002 – Table 2). This is particularly important for hemispheric comparisons, small samples, higher field strengths, and when examining larger and varying voxel sizes (Kreis 2004). Through application of all these corrections the variance would substantially decrease and sensitivity would increase.

Cho and Cr ratios

One of the most frequently cited controversies in this area of study is the use of ratios of metabolites to Cr or Cho, with many authors doubting its informative value (Pfefferbaum et al. 1999; Deicken et al. 2001; Théberge et al. 2002; Li et al. 2003; van Elst et al. 2005; Dager et al. 2008). While ratios might have a good intra-subject validity, several studies have clearly shown a differential expression of Cho and Cr in pathological conditions as compared to healthy participants (Pouwels and Frahm 1998; Dager et al. 2008; Purdon et al. 2008; Ongur et al. 2010a, b); e.g. Cr and Cho were shown to decrease with age in schizophrenia patients (Chang et al. 2007 – Tables 1, 7, and 10), whereas no change in Cho and an increase with age in Cr was seen in healthy control participants (Chang et al. 2007; Maudsley et al. 2009).

T2 relaxation times

A correction of T2 values has been found of great importance for unbiased quantification of metabolite levels (Jansen et al. 2006; Zaaraoui et al. 2007). Many studies have shown divergent T2 values in several cerebral regions between healthy control participants and schizophrenia patients (Williamson et al. 1992; Supprian et al. 1997; Tunc-Skarka et al. 2009). With short TEs a broader signal is conceived because of minimized signal loss caused by T2 relaxation, however, more unsystematic error variance is included (i.e. macromolecule signals) resulting in less reproducibility than with long TEs (Inglese et al. 2006). Long TEs on the other hand include more systematic error variance, if metabolites remain uncorrected for T2 relaxation times (Zaaraoui et al. 2007). One author found increased NAA levels in the schizophrenia sample only because of shortened T2 relaxation times with increasing TE (Tunc-Skarka et al. 2009 – Table 7). In addition, T2 relaxation times vary depending on the region and tissue at hand (Supprian et al. 1997; Whittall et al. 1997; Zaaraoui et al. 2007; Brief et al. 2009). Thus, to precisely determine metabolites, their relaxation corrections should be also weighted by their tissue content. Unfortunately, this correction is often not executed, because of time and resource constraints (Kreis et al. 2005).

Spectral and spatial resolution

By varying voxel size, localization procedures, field strengths and acquisition times, the observed concentrations of the detected metabolites might differ accordingly. The spatial resolution in 1H-MRS remains a problem, especially in studies using larger voxel (Sanches et al. 2004). Furthermore, control participants and patient groups assessed serially often differ in the exact amount of each tissue in the specified voxel and the localization of the VOI lacks sufficient accuracy to assure a direct comparison (Marshall et al. 1996). Studies deviate enormously in the way they localize and report the region of interest, with only some applying sophisticated methods and rigor to trace certain regions. For instance, one study (Jakary et al. 2005 – Table 3) constructed a thalamic tissue volume mask for each individual subject to not include any extra thalamic gray and WM. Others (Szulc et al. 2004, 2007 – Table 3) only mention the lobe and not the exact voxel location of the assessed region. This is very broad considering the size of their voxel (2 × 2 × 2 cm) and not sufficient, given the structural and functional heterogeneity within lobes and given the differing metabolite concentrations in gray and WM (Michaelis et al. 1993; Stanley et al. 1996; Pouwels and Frahm 1998).

Another issues that raises concerns are that most studies measuring Glu/Gln and GABA concentrations, do so at low field strengths and with inadequate quantification techniques. Some authors (Bartha et al. 1999) stressed that quantification of these metabolites has a very low precision because of their broad and overlapping resonances. Therefore, it is necessary to measure these metabolites with short TEs, sophisticated editing method when appropriate and preferably high field strengths (Pfeuffer et al. 1999; Tkac et al. 2001; Hurd et al. 2004). In addition, studies often fail to report standard deviations and reliability estimates of measured metabolite concentrations, making it difficult to judge the quality of their spectra.

To enhance the validity and reliability of upcoming 1H-MRS studies in schizophrenia, we propose the following minimal acceptable practice parameters for optimal data acquisition:

  • Consider to make the sample as homogenous as possible (e.g. gender, age, schizophrenia subtype, medication) and if not possible match accordingly;
  • When including two treatment groups one ought to include two matching control groups;
  • Studies including gray and white matter in the VOI should aim to assess for tissue differences and CSF;
  • 1H-MRS studies comparing two different sample populations (e.g. healthy participants vs. CSP) should avoid the use of rations given the differential expression in pathologies;
  • T2 relaxation time corrections would be of great value, especially in studies using long TEs. However, one should realize that possible individual differences are not taken into account when using the same T2 relaxation time per resonance for each measurement;
  • VOIs should be accurately defined and preferably not include overlapping regions;
  • When analyzing overlapping metabolites (e.g. Glu, Gln, GABA), studies should use high field strengths;
  • All studies should include mean, standard deviations, and reliability estimates.

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

Overall, a broad consensus of studies agrees on changes in metabolites that are narrowly associated with regions found to be abnormal in schizophrenia. Furthermore, declined NAA concentrations have been found in several regions and there is evidence to support a time-dependent role of glutamatergic neurotoxicity in schizophrenia.

When assessing metabolites, samples should be as homogeneous as possible, consisting of equal gender ratios in the control participants and subjects, be best assessed with methodological thoroughness (e.g. including PVC and relaxation time corrections) and measured with AVs. Good reliability and validity of 1H-MRS will hopefully be achieved by future studies adhering to our proposed ‘minimal acceptable practice parameters’, thus incorporating more rigor and standardization of acquisition parameters, leading to more accurate and comparable findings.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References

We are grateful to the authors who provided additional information about their investigations facilitating the development of this review.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. 1H-MRS findings in schizophrenia
  5. Discussion
  6. Conclusions
  7. Acknowledgments
  8. Conflict of interests
  9. References
  • Ando K., Takei N., Matsumoto H., Iyo M., Isoda H. and Mori N. (2002) Neural damage in the lenticular nucleus linked with tardive dyskinesia in schizophrenia: a preliminary study using proton magnetic resonance spectroscopy. Schizophr. Res. 57, 273279.
  • Aoyama N., Théberge J., Drost D. J. et al. (2011) Grey matter and social functioning correlates of glutamatergic metabolite loss in schizophrenia. Br. J. Psychiatry 198, 448456.
  • Auer D. P., Wilke M., Grabner A., Heidenreich J. O., Bronisch T. and Wetter T. C. (2001) Reduced NAA in the thalamus and altered membrane and glial metabolism in schizophrenic patients detected by 1H-MRS and tissue segmentation. Schizophr. Res. 52, 8799.
  • Aydin K., Ucok A. and Cakir S. (2007) Quantitative proton MR spectroscopy findings in the corpus callosum of patients with schizophrenia suggest callosal disconnection. AJNR Am. J. Neuroradiol. 28, 19681974.
  • Aydin K., Ucok A. and Guler J. (2008) Altered metabolic integrity of corpus callosum among individuals at ultra high risk of schizophrenia and first-episode patients. Biol. Psychiatry 64, 750757.
  • Barker P. B. (2001) N-acetyl aspartate - a neuronal marker? Ann. Neurol. 49, 423424.
  • Bartha R., Drost D. J. and Williamson P. C. (1999) Factors affecting the quantification of short echo in-vivo 1H MR spectra: prior knowledge, peak elimination, and filtering. NMR Biomed. 12, 205216.
  • Basoglu C., Cetin M., Oner O., Ebrinc S., Semiz U. B., Kandilcioglu H., Silit E. and Kizilkaya E. (2006) Comparison of right thalamus and temporal cortex metabolite levels of drug-naive first-episode psychotic and chronic schizophrenia in patients. Turk Psikiyatri Derg. 17, 8591.
  • Berger G. E., Wood S. J., Wellard R. M. et al. (2008) Ethyl-eicosapentaenoic acid in first-episode psychosis. A 1H-MRS study. Neuropsychopharmacology 33, 24672473.
  • Bertolino A. and Weinberger D. R. (1999) Proton magnetic resonance spectroscopy in schizophrenia. Eur. J. Radiol. 30, 132141.
  • Bertolino A., Callicott J. H., Mattay V. S., Weidenhammer K. M., Rakow R., Egan M. F. and Weinberger D. R. (2001) The effect of treatment with antipsychotic drugs on brain N-acetylaspartate measures in patients with schizophrenia. Biol. Psychiatry 49, 3946.
  • Bertolino A., Sciota D., Brudaglio F. et al. (2003) Working memory deficits and levels of N-acetylaspartate in patients with schizophreniform disorder. Am. J. Psychiatry 160, 483489.
  • Blatter D. D., Bigler E. D., Gale S. D., Johnson S. C., Anderson C. V., Burnett B. M., Parker N., Kurth S. and Horn S. D. (1995) Quantitative volumetric analysis of brain MR: normative database spanning 5 decades of life. AJNR Am. J. Neuroradiol. 16, 241251.
  • Block W., Bayer T. A., Tepest R. et al. (2000) Decreased frontal lobe ratio of N-acetyl aspartate to choline in familial schizophrenia: a proton magnetic resonance spectroscopy study. Neurosci. Lett. 289, 147151.
  • Bloemen O. J., Gleich T., de Koning M. B., da Silva Alvis F., de Haan L., Linszen D. H., Booij J. and van Amelsvoort T. A. (2011) Hippocampal glutamate levels and striatal dopamine D(2/3) receptor occupancy in subjects at ultra high risk of psychosis. Biol. Psychiatry 70, e1e2; author reply e3.
  • Bogner W., Gruber S., Doelken M. et al. (2010) In vivo quantification of intracerebral GABA by single-voxel (1)H-MRS-How reproducible are the results? Eur. J. Radiol. 73, 526531.
  • Braus D. F., Ende G., Weber-Fahr W., Demirakca T., Tost H. and Henn F. A. (2002) Functioning and neuronal viability of the anterior cingulate neurons following antipsychotic treatment: MR-spectroscopic imaging in chronic schizophrenia. Eur. Neuropsychopharmacol. 12, 145152.
  • Brief E. E., Moll R., Li D. K. and Mackay A. L. (2009) Absolute metabolite concentrations calibrated using the total water signal in brain (1)H MRS. NMR Biomed. 22, 349354.
  • Bustillo J. R., Rowland L. M., Jung R., Brooks W. M., Qualls C., Hammond R., Hart B. and Lauriello J. (2008) Proton magnetic resonance spectroscopy during initial treatment with antipsychotic medication in schizophrenia.Neuropsychopharmacology: official publication of the American College of. Neuropsychopharmacology 33, 24562466.
  • Bustillo J. R., Rowland L. M., Mullins P., Jung R., Chen H., Qualls C., Hammond R., Brooks W. M. and Lauriello J. (2010) 1H-MRS at 4 tesla in minimally treated early schizophrenia. Mol. Psychiatry 15, 629636.
  • Bustillo J. R., Chen H., Gasparovic C., Mullins P., Caprihan A., Qualls C., Apfeldorf W., Lauriello J. and Posse S. (2010) Glutamate as a marker of cognitive function in schizophrenia: a proton spectroscopic imaging study at 4 Tesla. Biol. Psychiatry 69, 1927.
  • Byne W., Hazlett E. A., Buchsbaum M. S. and Kemether E. (2009) The thalamus and schizophrenia: current status of research. Acta Neuropathol. 117, 347368.
  • Callicott J. H., Egan M. F., Bertolino A., Mattay V. S., Langheim F. J., Frank J. A. and Weinberger D. R. (1998) Hippocampal N-acetyl aspartate in unaffected siblings of patients with schizophrenia: a possible intermediate neurobiological phenotype. Biol. Psychiatry 44, 941950.
  • Callicott J. H., Bertolino A., Egan M. F., Mattay V. S., Langheim F. J. and Weinberger D. R. (2000) Selective relationship between prefrontal N-acetylaspartate measures and negative symptoms in schizophrenia. Am. J. Psychiatry 157, 16461651.
  • Callicott J. H., Egan M. F., Mattay V. S., Bertolino A., Bone A. D., Verchinksi B. and Weinberger D. R. (2003) Abnormal fMRI response of the dorsolateral prefrontal cortex in cognitively intact siblings of patients with schizophrenia. Am. J. Psychiatry 160, 709719.
  • Canuso C. M. and Pandina G. (2007) Gender and schizophrenia. Psychopharmacol. Bull. 40, 178190.
  • Carlsson A., Waters N. and Carlsson M. L. (1999) Neurotransmitter interactions in schizophrenia-therapeutic implications. Eur. Arch. Psychiatry Clin. Neurosci. 249 (Suppl 4), 3743.
  • Chan R. C., Di X., McAlonan G. M. and Gong Q. Y. (2011) Brain anatomical abnormalities in high-risk individuals, first-episode, and chronic schizophrenia: an activation likelihood estimation meta-analysis of illness progression. Schizophr. Bull. 37, 177188.
  • Chang L., Friedman J., Ernst T., Zhong K., Tsopelas N. D. and Davis K. (2007) Brain metabolite abnormalities in the WM of elderly schizophrenic subjects: implication for glial dysfunction. Biol. Psychiatry 62, 13961404.
  • Chen X., Sachdev P. S., Wen W. and Anstey K. J. (2007) Sex differences in regional gray matter in healthy individuals aged 44-48 years: a voxel-based morphometric study. NeuroImage 36, 691699.
  • Clark J. B. (1998) N-acetyl aspartate: a marker for neuronal loss or mitochondrial dysfunction. Dev. Neurosci. 20, 271276.
  • Cosgrove K. P., Mazure C. M. and Staley J. K. (2007) Evolving knowledge of sex differences in brain structure, function, and chemistry. Biol. Psychiatry 62, 847855.
  • Csernansky J. G., Schindler M. K., Splinter N. R. et al. (2004) Abnormalities of thalamic volume and shape in schizophrenia. Am. J. Psychiatry 161, 896902.
  • Dager S. R., Corrigan N. M., Richards T. L. and Posse S. (2008) Research applications of magnetic resonance spectroscopy to investigate psychiatric disorders. Top. Magn. Reson. Imaging 19, 8196.
  • Deicken R. F., Johnson C., Eliaz Y. and Schuff N. (2000) Reduced concentrations of thalamic N-acetylaspartate in male patients with schizophrenia. Am. J. Psychiatry 157, 644647.
  • Deicken R. F., Feiwell R., Schuff N. and Soher B. (2001) Evidence for altered cerebellar vermis neuronal integrity in schizophrenia. Psychiatry Res. 107, 125134.
  • Delamillieure P., Fernandez J., Constans J. M. et al. (2000) Proton magnetic resonance spectroscopy of the medial prefrontal cortex in patients with deficit schizophrenia: preliminary report. Am. J. Psychiatry 157, 641643.
  • Do K. Q., Trabesinger A. H., Kirsten-Kruger M., Lauer C. J., Dydak U., Hell D., Holsboer F., Boesiger P. and Cuenod M. (2000) Schizophrenia: glutathione deficit in cerebrospinal fluid and prefrontal cortex in vivo. Eur. J. Neurosci. 12, 37213728.
  • Drost D. J., Riddle W. R. and Clarke G. D. (2002) Proton magnetic resonance spectroscopy in the brain: report of AAPM MR Task Group #9. Med. Phys. 29, 21772197.
  • Ellison-Wright I. and Bullmore E. (2009) Meta-analysis of diffusion tensor imaging studies in schizophrenia. Schizophr. Res. 108, 310.
  • van Elst L. T., Valerius G., Büchert M., Thiel T., Rüsch N., Bubl E., Hennig J., Ebert D. and Olbrich H. M. (2005) Increased prefrontal and hippocampal glutamate concentration in schizophrenia: evidence from a magnetic resonance spectroscopy study. Biol. Psychiatry 1, 724730.
  • Ende G., Braus D. F., Walter S., Weber-Fahr W., Soher B., Maudsley A. A. and Henn F. A. (2000) Effects of age, medication, and illness duration on the N-acetyl aspartate signal of the anterior cingulate region in schizophrenia. Schizophr. Res. 41, 389395.
  • Fannon D., Simmons A., Tennakoon L., O'Ceallaigh S., Sumich A., Doku V., Shew C. and Sharma T. (2003) Selective deficit of hippocampal N-acetylaspartate in antipsychotic-naive patients with schizophrenia. Biol. Psychiatry 54, 587598.
  • Farber N. B. (2003) The NMDA receptor hypofunction model of psychosis. Ann. N. Y. Acad. Sci. 1003, 119130.
  • Fornito A., Yucel M. and Pantelis C. (2009) Reconciling neuroimaging and neuropathological findings in schizophrenia and bipolar disorder. Curr. Opin. Psychiatry 22, 312319.
  • Frahm J., Bruhn H., Gyngell M. L., Merboldt K. D., Hänicke W. and Sauter R. (1989) Localized high-resolution proton NMR spectroscopy using stimulated echoes: initial applications to human brain in vivo. Magn. Reson. Med. 9, 7993.
  • de la Fuente-Sandoval C., Favila R., Alvarado P., Leon-Ortiz P., Diaz-Galvis L., Amezcua C., Garcia-Munoz E. and Graff-Guerrero A. (2009) Glutamate increase in the associative striatum in schizophrenia: a longitudinal magnetic resonance spectroscopy preliminary study. Gac. Med. Mex. 145, 109113.
  • de la Fuente-Sandoval C., Leon-Ortiz P., Favila R., Stephano S., Mamo D., Ramirez-Bermudez J. and Graff-Guerrero A. (2011) Higher levels of glutamate in the associative-striatum of subjects with prodromal symptoms of schizophrenia and patients with first-episode psychosis.. Neuropsychopharmacology 36, 17811791.
  • Fujiwara H., Shimizu M., Hirao K., Miyata J., Namiki C., Sawamoto N., Fukuyama H., Hayashi T. and Murai T. (2008) Female specific anterior cingulate abnormality and its association with empathic disability in schizophrenia. Prog. Neuropsychopharmacol. Biol. Psychiatry 32, 17281734.
  • Fukuzako H. (2000) Heritability heightens brain metabolite differences in schizophrenia. J. Neuropsychiatry Clin. Neurosci. 12, 9597.
  • Fusar-Poli P., Perez J., Broome M. et al. (2007) Neurofunctional correlates of vulnerability to psychosis: a systematic review and meta-analysis. Neurosci. Biobehav. Rev. 31, 465484.
  • Fusar-Poli P., Howes O., Valli I., Allen P., Broome M., Grasby P. and McGuire P. (2009) Multimodal functional imaging investigation before and after the onset of psychosis. Int. J. Neuropsychopharmacol. 12, 579581.
  • Galinska B., Szulc A., Tarasow E., Kubas B., Dzienis W., Czernikiewicz A. and Walecki J. (2009) Duration of untreated psychosis and proton magnetic resonance spectroscopy (1H-MRS) findings in first-episode schizophrenia. Med. Sci. Monit. 15, CR82CR88.
  • Gillies R. J. (1992) Nuclear magnetic resonance and its applications to physiological problems. Annu. Rev. Physiol. 54, 733748.
  • Gimenez M., Junque C., Perez M., Vendrell P., Baeza I., Salamero M., Mercader J. M. and Bernardo M. (2003) Basal ganglia N-acetylaspartate correlates with the performance in the procedural task ‘Tower of Hanoi’ of neuroleptic-naive schizophrenic patients. Neurosci. Lett. 347, 97100.
  • Goto N., Yoshimura R., Moriya J. et al. (2009) Reduction of brain gamma-aminobutyric acid (GABA) concentrations in early-stage schizophrenia patients: 3T Proton MRS study. Schizophr. Res. 112, 192193.
  • Goto N., Yoshimura R., Kakeda S. et al. (2010) No alterations of brain GABA after 6 months of treatment with atypical antipsychotic drugs in early-stage first-episode schizophrenia. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 14801483.
  • Goto N., Yoshimura R., Kakeda S. et al. (2011) Comparison of brain N-acetylaspartate levels and serum brain-derived neurotrophic factor (BDNF) levels between patients with first-episode schizophrenia psychosis and healthy controls. Eur. Psychiatry 26, 5763.
  • Govindaraju V., Young K. and Maudsley A. A. (2000) Proton NMR chemical shifts and coupling constants for brain metabolites. NMR Biomed. 13, 129153.
  • Gur R. C., Turetsky B. I., Matsui M., Yan M., Bilker W., Hughett P. and Gur R. E. (1999) Sex differences in brain gray and white matter in healthy young adults: correlations with cognitive performance. J. Neurosci. 19, 40654072.
  • Hammen T. and Stefan H. (2004) (1)H MR spectroscopy. Methods and applications in diagnosis and assessment of surgical and conservative treatment strategies in epilepsies. Nervenarzt 75, 970983.
  • Hardy C. J., Tal A., Babb J. S., Perry N. N., Messinger J. W., Antonius D., Malaspina D. and Gonen O. (2011) Multivoxel proton MR spectroscopy used to distinguish anterior cingulate metabolic abnormalities in patients with schizophrenia. Radiology 261, 542550.
  • Harms M. P., Wang L., Mamah D., Barch D. M., Thompson P. A. and Csernansky J. G. (2007) Thalamic shape abnormalities in individuals with schizophrenia and their nonpsychotic siblings. J. Neurosci. 27, 1383513842.
  • Hasan A., Wobrock T., Falkai P. et al. (2011) Hippocampal integrity and neurocognition in first-episode schizophrenia: a multidimensional study. World J. Biol. Psychiatry [Epub ahead of print].
  • He Z. L., Deng W., Li M. L., Chen Z. F., Collier D. A., Ma X. and Li T. (2012) Detection of metabolites in the white matter of frontal lobes and hippocampus with proton in first-episode treatment-naive schizophrenia patients. Early Interv. Psychiatry 6, 166175.
  • Hirvonen J., van Erp T. G., Huttunen J. et al. (2005) Increased caudate dopamine D2 receptor availability as a genetic marker for schizophrenia. Arch. Gen. Psychiatry 62, 371378.
  • Honea R., Crow T. J., Passingham D. and Mackay C. E. (2005) Regional deficits in brain volume in schizophrenia: a meta-analysis of voxel-based morphometry studies. Am. J. Psychiatry 162, 22332245.
  • Hurd R., Sailasuta N., Srinivasan R., Vigneron D. B., Pelletier D. and Nelson S. J. (2004) Measurement of brain glutamate using TE-averaged PRESS at 3T. Magn. Reson. Med. 51, 435440.
  • Inglese M., Spindler M., Babb J. S., Sunenshine P., Law M. and Gonen O. (2006) Field, coil, and echo-time influence on sensitivity and reproducibility of brain proton MR spectroscopy. AJNR Am. J. Neuroradiol. 27, 684688.
  • Jakary A., Vinogradov S., Feiwell R. and Deicken R. F. (2005) N-acetylaspartate reductions in the mediodorsal and anterior thalamus in men with schizophrenia verified by tissue volume corrected proton MRSI. Schizophr. Res. 76, 173185.
  • Jansen J. F., Backes W. H., Nicolay K. and Kooi M. E. (2006) 1H MR spectroscopy of the brain: absolute quantification of metabolites. Radiology 240, 318332.
  • Jessen F., Scherk H., Traber F. et al. (2006) Proton magnetic resonance spectroscopy in subjects at risk for schizophrenia. Schizophr. Res. 87, 8188.
  • Jessen F., Fingerhut N., Sprinkart A. M. et al. (2013) N-acetylaspartylglutamate (NAAG) and N-acetylaspartate (NAA) in patients with schizophrenia. Schizophr. Bull. 39, 197205.
  • Kapur S., Mizrahi R. and Li M. (2005) From dopamine to salience to psychosis–linking biology, pharmacology and phenomenology of psychosis. Schizophr. Res. 79, 5968.
  • Kegeles L. S., Shungu D. C., Anjilvel S. et al. (2000) Hippocampal pathology in schizophrenia: magnetic resonance imaging and spectroscopy studies. Psychiatry Res. 98, 163175.
  • Kegeles L. S., Mao X., Stanford A. D. et al. (2012) Elevated prefrontal cortex gamma-aminobutyric acid and glutamate-glutamine levels in schizophrenia measured in vivo with proton magnetic resonance spectroscopy. Arch. Gen. Psychiatry 69, 449459.
  • Klar A. A., Ballmaier M., Leopold K., Hake I., Schaefer M., Bruhl R., Schubert F. and Gallinat J. (2010) Interaction of hippocampal volume and N-acetylaspartate concentration deficits in schizophrenia: a combined MRI and 1H-MRS study. NeuroImage 53, 5157.
  • Kreis R. (2004) Issues of spectral quality in clinical 1H-magnetic resonance spectroscopy and a gallery of artifacts. NMR Biomed. 17, 361381.
  • Kreis R., Ernst T. and Ross B. D. (1993) Development of the human brain: in vivo quantification of metabolite and water content with proton magnetic resonance spectroscopy. Magn. Reson. Med. 30, 424437.
  • Kreis R., Slotboom J., Hofmann L. and Boesch C. (2005) Integrated data acquisition and processing to determine metabolite contents, relaxation times, and macromolecule baseline in single examinations of individual subjects. Magn. Reson. Med. 54, 761768.
  • Li B. S., Wang H. and Gonen O. (2003) Metabolite ratios to assumed stable creatine level may confound the quantification of proton brain MR spectroscopy. Magn. Reson. Imaging 21, 923928.
  • Lui S., Deng W., Huang X. et al. (2009) Neuroanatomical differences between familial and sporadic schizophrenia and their parents: an optimized voxel-based morphometry study. Psychiatry Res. 171, 7181.
  • Lutkenhoff E. S., van Erp T. G., Thomas M. A. et al. (2010) Proton MRS in twin pairs discordant for schizophrenia. Mol. Psychiatry 15, 308318.
  • Mamah D., Harms M. P., Wang L., Barch D., Thompson P., Kim J., Miller M. I. and Csernansky J. G. (2008) Basal ganglia shape abnormalities in the unaffected siblings of schizophrenia patients. Biol. Psychiatry 64, 111120.
  • Marshall I., Wardlaw J., Cannon J., Slattery J. and Sellar R. J. (1996) Reproducibility of metabolite peak areas in 1H MRS of brain. Magn. Reson. Imaging 14, 281292.
  • Martinez-Granados B., Brotons O., Martinez-Bisbal M. C., Celda B., Marti-Bonmati L., Aguilar E. J., Gonzalez J. C. and Sanjuan J. (2008) Spectroscopic metabolomic abnormalities in the thalamus related to auditory hallucinations in patients with schizophrenia. Schizophr. Res. 104, 1322.
  • Matsuzawa D., Obata T., Shirayama Y. et al. (2008) Negative correlation between brain glutathione level and negative symptoms in schizophrenia: a 3T 1H-MRS study. PLoS One 3, e1944.
  • Maudsley A. A., Domenig C., Govind V., Darkazanli A., Studholme C., Arheart K. and Bloomer C. (2009) Mapping of brain metabolite distributions by volumetric proton MR spectroscopic imaging (MRSI). Magn. Reson. Med. 61, 548559.
  • Michaelis T., Merboldt K. D., Bruhn H., Hanicke W. and Frahm J. (1993) Absolute concentrations of metabolites in the adult human brain in vivo: quantification of localized proton MR spectra. Radiology 187, 219227.
  • Miyaoka T., Yasukawa R., Mizuno S., Sukegawa T., Inagaki T., Horiguchi J., Seno H., Oda K. and Kitagaki H. (2005) Proton magnetic resonance spectroscopy (1H-MRS) of hippocampus, basal ganglia, and vermis of cerebellum in schizophrenia associated with idiopathic unconjugated hyperbilirubinemia (Gilbert's syndrome). J. Psychiatr. Res. 39, 2934.
  • Moffett J. R. and Namboodiri M. A. (1995) Differential distribution of N-acetylaspartylglutamate and N-acetylaspartate immunoreactivities in rat forebrain. J. Neurocytol. 24, 409433.
  • Molina V., Sanchez J., Reig S. et al. (2005) N-acetyl-aspartate levels in the dorsolateral prefrontal cortex in the early years of schizophrenia are inversely related to disease duration. Schizophr. Res. 73, 209219.
  • Moore C. M., Bonello C. M., Sherwood A. R., Cohen B. M., Renshaw P. F. and Yurgulen-Todd D. A. (2002) Mesial temporal lobe Cho to Cr(PCr) ratio asymmetry in chronic schizophrenics. Schizophr. Res. 57, 3542.
  • Morche S. (2005) Single-Voxel- versus Multi-Voxel Protonen-MR-Spektroskopie bei verschiedenen intrakraniellen Läsionen. University Würzburg, Würzburg.
  • Moreno D., Burdalo M., Reig S., Parellada M., Zabala A., Desco M., Baca-Baldomero E. and Arango C. (2005) Structural neuroimaging in adolescents with a first psychotic episode. J. Am. Acad. Child Adolesc. Psychiatry 44, 11511157.
  • Narr K. L., Thompson P. M., Sharma T. et al. (2001) Three-dimensional mapping of temporo-limbic regions and the lateral ventricles in schizophrenia: gender effects. Biol. Psychiatry 50, 8497.
  • Novotny E. J. Jr, Fulbright R. K., Pearl P. L., Gibson K. M. and Rothman D. L. (2003) Magnetic resonance spectroscopy of neurotransmitters in human brain. Ann. Neurol. 54 (Suppl 6), S25S31.
  • Ohara K., Isoda H., Suzuki Y., Takehara Y., Ochiai M., Takeda H., Hattori K., Igarashi Y. and Ohara K. (2000) Proton magnetic resonance spectroscopy of lenticular nuclei in simple schizophrenia. Prog. Neuropsychopharmacol. Biol. Psychiatry 24, 507519.
  • Ohrmann P., Siegmund A., Suslow T. et al. (2007) Cognitive impairment and in vivo metabolites in first-episode neuroleptic-naive and chronic medicated schizophrenic patients: a proton magnetic resonance spectroscopy study. J. Psychiatr. Res. 41, 625634.
  • Ohrmann P., Kugel H., Bauer J. et al. (2008) Learning potential on the WCST in schizophrenia is related to the neuronal integrity of the anterior cingulate cortex as measured by proton magnetic resonance spectroscopy. Schizophr. Res. 106, 156163.
  • Olbrich H. M., Valerius G., Rusch N., Buchert M., Thiel T., Hennig J., Ebert D. and Van Elst L. T. (2008) Frontolimbic glutamate alterations in first episode schizophrenia: evidence from a magnetic resonance spectroscopy study. World J. Biol. Psychiatry 9, 5963.
  • Omori M., Murata T., Kimura H., Koshimoto Y., Kado H., Ishimori Y., Ito H. and Wada Y. (2000) Thalamic abnormalities in patients with schizophrenia revealed by proton magnetic resonance spectroscopy. Psychiatry Res. 98, 155162.
  • O'Neill J., Levitt J., Caplan R., Asarnow R., McCracken J. T., Toga A. W. and Alger J. R. (2004) 1H MRSI evidence of metabolic abnormalities in childhood-onset schizophrenia. NeuroImage 21, 17811789.
  • Ongur D., Prescot A. P., Jensen J. E., Cohen B. M. and Renshaw P. F. (2009) Creatine abnormalities in schizophrenia and bipolar disorder. Psychiatry Res. 172, 4448.
  • Ongur D., Prescot A. P., Jensen J. E., Rouse E. D., Cohen B. M., Renshaw P. F. and Olson D. P. (2010a) T2 relaxation time abnormalities in bipolar disorder and schizophrenia. Magn. Reson. Med. 63, 18.
  • Ongur D., Prescot A. P., McCarthy J., Cohen B. M. and Renshaw P. F. (2010b) Elevated gamma-aminobutyric acid levels in chronic schizophrenia. Biol. Psychiatry 68, 667670.
  • Ota M, Ishikawa M, Sato N, Hori H, Sasayama D, Hattori K, Teraishi T, Nakata Y and Kunugi H. (2012) Glutamatergic changes in the cerebral white matter associated with schizophrenic exacerbation. Acta Psychiatr. Scand. 126, 7278.
  • Pfefferbaum A., Adalsteinsson E., Spielman D., Sullivan E. V. and Lim K. O. (1999) In vivo spectroscopic quantification of the N-acetyl moiety, creatine, and choline from large volumes of brain gray and white matter: effects of normal aging. Magn. Reson. Med. 41, 276284.
  • Pfeuffer J., Tkac I., Provencher S. W. and Gruetter R. (1999) Toward an in vivo neurochemical profile: quantification of 18 metabolites in short-echo-time (1)H NMR spectra of the rat brain. J. Magn. Reson. 141, 104120.
  • Pouwels P. J. and Frahm J. (1998) Regional metabolite concentrations in human brain as determined by quantitative localized proton MRS. Magn. Reson. Med. 39, 5360.
  • Purdon S. E., Valiakalayil A., Hanstock C. C., Seres P. and Tibbo P. (2008) Elevated 3T proton MRS glutamate levels associated with poor Continuous Performance Test (CPT-0X) scores and genetic risk for schizophrenia. Schizophr. Res. 99, 218224.
  • Reid M. A., Stoeckel L. E., White D. M., Avsar K. B., Bolding M. S., Akella N. S., Knowlton R. C., den Hollander J. A. and Lahti A. C. (2010) Assessments of function and biochemistry of the anterior cingulate cortex in schizophrenia. Biol. Psychiatry 68, 625633.
  • Reig S., Moreno C., Moreno D., Burdalo M., Janssen J., Parellada M., Zabala A., Desco M. and Arango C. (2009) Progression of brain volume changes in adolescent-onset psychosis. Schizophr. Bull. 35, 233243.
  • Reiss A. L., Abrams M. T., Singer H. S., Ross J. L. and Denckla M. B. (1996) Brain development, gender and IQ in children. A volumetric imaging study. Brain 119 (Pt 5), 17631774.
  • Rothermundt M., Ohrmann P., Abel S. et al. (2007) Glial cell activation in a subgroup of patients with schizophrenia indicated by increased S100B serum concentrations and elevated myo-inositol. Prog. Neuropsychopharmacol. Biol. Psychiatry 31, 361364.
  • Rothman D. L., Sibson N. R., Hyder F., Shen J., Behar K. L. and Shulman R. G. (1999) In vivo nuclear magnetic resonance spectroscopy studies of the relationship between the glutamate-glutamine neurotransmitter cycle and functional neuroenergetics. Philos. Trans. R. Soc. Lond. B Biol. Sci. 354, 11651177.
  • Rowland L., Bustillo J. R. and Lauriello J. (2001) Proton magnetic resonance spectroscopy (H-MRS) studies of schizophrenia. Semin. Clin. Neuropsychiatry 6, 121130.
  • Rowland L. M., Spieker E. A., Francis A., Barker P. B., Carpenter W. T. and Buchanan R. W. (2009) White matter alterations in deficit schizophrenia. Neuropsychopharmacology: official publication of the American College of. Neuropsychopharmacology 34, 15141522.
  • Rusch N., Tebartz van Elst L., Valerius G., Buchert M., Thiel T., Ebert D., Hennig J. and Olbrich H. M. (2008) Neurochemical and structural correlates of executive dysfunction in schizophrenia. Schizophr. Res. 99, 155163.
  • Sanches R. F., Crippa J. A., Hallak J. E., Araujo D. and Zuardi A. W. (2004) Proton magnetic resonance spectroscopy of the frontal lobe in schizophrenics: a critical review of the methodology. Rev. Hosp. Clin. Fac. Med. Sao Paulo 59, 145152.
  • Sanches R. F., Crippa J. A., Hallak J. E., de Sousa J. P., Araujo D., Santos A. C. and Zuardi A. W. (2008) Proton magnetic resonance spectroscopy of the frontal, cingulate and perirolandic cortices and its relationship to skin conductance in patients with schizophrenia. Braz. J. Med. Biol. Res. 41, 11321141.
  • Seamans J. K. and Yang C. R. (2004) The principal features and mechanisms of dopamine modulation in the prefrontal cortex. Prog. Neurobiol. 74, 158.
  • Shimizu E., Hashimoto K., Ochi S. et al. (2007) Posterior cingulate gyrus metabolic changes in chronic schizophrenia with generalized cognitive deficits. J. Psychiatr. Res. 41, 4956.
  • Sigmundsson T., Maier M., Toone B. K., Williams S. C., Simmons A., Greenwood K. and Ron M. A. (2003) Frontal lobe N-acetylaspartate correlates with psychopathology in schizophrenia: a proton magnetic resonance spectroscopy study. Schizophr. Res. 64, 6371.
  • Simmons M. L., Frondoza C. G. and Coyle J. T. (1991) Immunocytochemical localization of N-acetyl-aspartate with monoclonal antibodies. Neuroscience 45, 3745.
  • Stanley J. A., Williamson P. C., Drost D. J., Rylett R. J., Carr T. J., Malla A. and Thompson R. T. (1996) An in vivo proton magnetic resonance spectroscopy study of schizophrenia patients. Schizophr. Bull. 22, 597609.
  • Stanley J. A., Pettegrew J. W. and Keshavan M. S. (2000) Magnetic resonance spectroscopy in schizophrenia: methodological issues and findings–part I. Biol. Psychiatry 48, 357368.
  • Stanley J. A., Vemulapalli M., Nutche J., Montrose D. M., Sweeney J. A., Pettegrew J. W., MacMaster F. P. and Keshavan M. S. (2007) Reduced N-acetyl-aspartate levels in schizophrenia patients with a younger onset age: a single-voxel 1H spectroscopy study. Schizophr. Res. 93, 2332.
  • Steel R. M., Bastin M. E., McConnell S., Marshall I., Cunningham-Owens D. G., Lawrie S. M., Johnstone E. C. and Best J. J. (2001) Diffusion tensor imaging (DTI) and proton magnetic resonance spectroscopy (1H MRS) in schizophrenic subjects and normal controls. Psychiatry Res. 106, 161170.
  • Steen R. G., Hamer R. M. and Lieberman J. A. (2005) Measurement of brain metabolites by 1H magnetic resonance spectroscopy in patients with schizophrenia: a systematic review and meta-analysis. Neuropsychopharmacology: official publication of the American College of. Neuropsychopharmacology 30, 19491962.
  • Stone J. M., Day F., Tsagaraki H., Valli I., McLean M. A., Lythgoe D. J., O'Gorman R. L., Barker G. J. and McGuire P. K. (2009) Glutamate dysfunction in people with prodromal symptoms of psychosis: relationship to gray matter volume. Biol. Psychiatry 66, 533539.
  • Supprian T., Hofmann E., Warmuth-Metz M., Franzek E. and Becker T. (1997) MRI T2 relaxation times of brain regions in schizophrenic patients and control subjects. Psychiatry Res. 75, 173182.
  • Szulc A. G., Tarasow B., Kubas E., Dzienis B., Walecki W., Czernikiewicz J. and A. (2004) Glutamatergic system dysfunction in schizophrenia. A proton magnetic resonance spectroscopy (1H MRS) Study. Pol. J. Radiol. 69, 3336.
  • Szulc A., Galinska B., Tarasow E. et al. (2007) N-acetylaspartate (NAA) levels in selected areas of the brain in patients with chronic schizophrenia treated with typical and atypical neuroleptics: a proton magnetic resonance spectroscopy (1H MRS) study. Med. Sci. Monit. 13 (Suppl 1), 1722.
  • Szulc A., Galinska B., Tarasow E., Waszkiewicz N., Konarzewska B., Poplawska R., Bibulowicz D., Simonienko K. and Walecki J. (2011) Proton magnetic resonance spectroscopy study of brain metabolite changes after antipsychotic treatment. Pharmacopsychiatry 44, 148157.
  • Tamagaki C., Sedvall G. C., Jonsson E. G., Okugawa G., Hall H., Pauli S. and Agartz I. (2005) Altered white matter/gray matter proportions in the striatum of patients with schizophrenia: a volumetric MRI study. Am. J. Psychiatry 162, 23152321.
  • Tang C. Y., Friedman J., Shungu D. et al. (2007) Correlations between Diffusion Tensor Imaging (DTI) and Magnetic Resonance Spectroscopy (1H MRS) in schizophrenic patients and normal controls. BMC Psychiatry 7, 25.
  • Tayoshi S., Sumitani S., Taniguchi K. et al. (2009) Metabolite changes and gender differences in schizophrenia using 3-Tesla proton magnetic resonance spectroscopy (1H-MRS). Schizophr. Res. 108, 6977.
  • Tayoshi S., Nakataki M., Sumitani S. et al. (2010) GABA concentration in schizophrenia patients and the effects of antipsychotic medication: a proton magnetic resonance spectroscopy study. Schizophr. Res. 117, 8391.
  • Théberge J., Bartha R., Drost D. J. et al. (2002) Glutamate and glutamine measured with 4.0 T proton MRS in never-treated patients with schizophrenia and healthy volunteers. Am. J. Psychiatry 159, 19441946.
  • Théberge J., Al-Semaan Y., Williamson P. C., Menon R. S., Neufeld R. W., Rajakumar N., Schaefer B., Densmore M. and Drost D. J. (2003) Glutamate and glutamine in the anterior cingulate and thalamus of medicated patients with chronic schizophrenia and healthy comparison subjects measured with 4.0-T proton MRS. Am. J. Psychiatry 160, 22312233.
  • Théberge J., Williamson P. C., Manchanda R., Menon R. S., Neufeld R. W., Densmore M., Williamson K. E. and Drost D. J. (2004) Longitudinal Study of First-Episode Schizophrenia using 4.0T 1H MRS and Voxel-Based Morphometry. Proc. Intl. Soc. Mag. Reson. Med. 11, 523.
  • Théberge J., Jensen J. E. and Rowland L. M. (2007a) Regarding “Increased prefrontal and hippocampal glutamate concentration in schizophrenia: evidence from a magnetic resonance spectroscopy study”. Biol. Psychiatry 61, 12181219; author reply 1219–1220.
  • Théberge J., Williamson K. E., Aoyama N. et al. (2007b) Longitudinal grey-matter and glutamatergic losses in first-episode schizophrenia. Br. J. Psychiatry 191, 325334.
  • Tibbo P., Hanstock C., Valiakalayil A. and Allen P. (2004) 3-T proton MRS investigation of glutamate and glutamine in adolescents at high genetic risk for schizophrenia. Am. J. Psychiatry 161, 11161118.
  • Tkac I., Andersen P., Adriany G., Merkle H., Ugurbil K. and Gruetter R. (2001) In vivo 1H NMR spectroscopy of the human brain at 7 T. Magn. Reson. Med. 46, 451456.
  • Tost H., Alam T. and Meyer-Lindenberg A. (2010) Dopamine and psychosis: theory, pathomechanisms and intermediate phenotypes. Neurosci. Biobehav. Rev. 34, 689700.
  • Tunc-Skarka N., Weber-Fahr W., Hoerst M., Meyer-Lindenberg A., Zink M. and Ende G. (2009) MR spectroscopic evaluation of N-acetylaspartate's T2 relaxation time and concentration corroborates white matter abnormalities in schizophrenia. NeuroImage 48, 525531.
  • Urenjak J., Williams S. R., Gadian D. G. and Noble M. (1993) Proton nuclear magnetic resonance spectroscopy unambiguously identifies different neural cell types. J. Neurosci. 13, 981989.
  • Vita A., De Peri L., Silenzi C. and Dieci M. (2006) Brain morphology in first-episode schizophrenia: a meta-analysis of quantitative magnetic resonance imaging studies. Schizophr. Res. 82, 7588.
  • Walter H. (2005) Funktionelle Bildgebung in Psychiatrie und Psychotherapie. Methodische Grundlagen und klinische Anwendungen. Schattauer GmbH, Stuttgart, New York.
  • Watts J., Fowler L., Whitton P. S. and Pearce B. (2005) Release of arginine, glutamate and glutamine in the hippocampus of freely moving rats: Involvement of nitric oxide. Brain Res. Bull. 65, 521528.
  • Weber-Fahr W., Ende G., Braus D. F., Bachert P., Soher B. J., Henn F. A. and Buchel C. (2002) A fully automated method for tissue segmentation and CSF-correction of proton MRSI metabolites corroborates abnormal hippocampal NAA in schizophrenia. NeuroImage 16, 4960.
  • Whittall K. P., MacKay A. L., Graeb D. A., Nugent R. A., Li D. K. and Paty D. W. (1997) In vivo measurement of T2 distributions and water contents in normal human brain. Magn. Reson. Med. 37, 3443.
  • Williamson P., Pelz D., Merskey H., Morrison S., Karlik S., Drost D., Carr T. and Conlon P. (1992) Frontal, temporal, and striatal proton relaxation times in schizophrenic patients and normal comparison subjects. Am. J. Psychiatry 149, 549551.
  • Wobrock T., Scherk H. and Falkai P. (2005) Magnetic resonance spectroscopy in schizophrenia. Possibilities and limitations. Radiologe 45: 124136.
  • Wood S. J., Berger G. E., Wellard R. M., Proffitt T., McConchie M., Velakoulis D., McGorry P. D. and Pantelis C. (2008) A 1H-MRS investigation of the medial temporal lobe in antipsychotic-naive and early-treated first episode psychosis. Schizophr. Res. 102, 163170.
  • Wood S. J., Berger G. E., Wellard R. M., Proffitt T. M., McConchie M., Berk M., McGorry P. D. and Pantelis C. (2009) Medial temporal lobe glutathione concentration in first episode psychosis: a 1H-MRS investigation. Neurobiol. Dis. 33, 354357.
  • Woodward N. D., Waldie B., Rogers B., Tibbo P., Seres P. and Purdon S. E. (2009) Abnormal prefrontal cortical activity and connectivity during response selection in first episode psychosis, chronic schizophrenia, and unaffected siblings of individuals with schizophrenia. Schizophr. Res. 109, 182190.
  • Yamasue H., Fukui T., Fukuda R. et al. (2002) 1H-MR spectroscopy and gray matter volume of the anterior cingulate cortex in schizophrenia. NeuroReport 13, 21332137.
  • Yasukawa R., Miyaoka T., Mizuno S., Inagaki T., Horiguchi J., Oda K. and Kitagaki H. (2005) Proton magnetic resonance spectroscopy of the anterior cingulate gyrus, insular cortex and thalamus in schizophrenia associated with idiopathic unconjugated hyperbilirubinemia (Gilbert's syndrome). J. Psychiatry Neurosci. 30, 416422.
  • Yoo S. Y., Yeon S., Choi C. H. et al. (2009) Proton magnetic resonance spectroscopy in subjects with high genetic risk of schizophrenia: investigation of anterior cingulate, dorsolateral prefrontal cortex and thalamus. Schizophr. Res. 111, 8693.
  • Yoon J. H., Maddock R. J., Rokem A., Silver M. A., Minzenberg M. J., Ragland J. D. and Carter C. S. (2010) GABA concentration is reduced in visual cortex in schizophrenia and correlates with orientation-specific surround suppression. J. Neurosci. 30, 37773781.
  • Yucel M., Stuart G. W., Maruff P., Velakoulis D., Crowe S. F., Savage G. and Pantelis C. (2001) Hemispheric and gender-related differences in the gross morphology of the anterior cingulate/paracingulate cortex in normal volunteers: an MRI morphometric study. Cereb. Cortex 11, 1725.
  • Yucel M., Lubman D. I., Harrison B. J. et al. (2007) A combined spectroscopic and functional MRI investigation of the dorsal anterior cingulate region in opiate addiction. Mol. Psychiatry 12, 611; 691702.
  • Zaaraoui W., Fleysher L., Fleysher R., Liu S., Soher B. J. and Gonen O. (2007) Human brain-structure resolved T(2) relaxation times of proton metabolites at 3 Tesla. Magn. Reson. Med. 57, 983989.
  • Zabala A., Sanchez-Gonzalez J., Parellada M., Moreno D. M., Reig S., Burdalo M. T., Robles O., Desco M. and Arango C. (2007) Findings of proton magnetic resonance spectometry in the dorsolateral prefrontal cortex in adolescents with first episodes of psychosis. Psychiatry Res. 156, 3342.