HIGH-RESOLUTION WHOLE BRAIN MR SPECTROSCOPIC IMAGING IN YOUTHS AT CLINICAL HIGH RISK FOR PSYCHOSIS: A PILOT STUDY.

Background: In general, MR spectroscopy (MRS) studies report alterations of both glutamatergic indices and NAA not only in first episode psychosis and established schizophrenia but also in high risk populations, suggesting that altered excitatory neurotransmission and loss of neuronal integrity are early pathophysiological processes. However, interpretation of these findings is limited by the region-of-interest approach of current MRS techniques, limiting the measurement of metabolites to delimited cerebral volumes, selected by a priori hypotheses. In that context, we developed and implemented a new technique including specific MR sequence and data reconstruction that allows for whole brain high-resolution MRS imaging (MRSI) in two or three dimensions. The results enable the mapping of main metabolites in all brain regions (cortex, white matter, deep grey matter) of youths at clinical high risk for psychosis (CHR-P). Methods: An FID-MRSI (Henning et al. NMR Biomed 2009) sequence with a 3D phase encoding accelerated by compressed-sensing was implemented on a 3T Prisma fit MRI (Siemens, Erlangen, Germany). The echo time (TE) was 0.65 ms, repetition time (TR) was 355 ms and the flip angle 35 degree. FID was acquired with 4 kHz bandwidth. The size of the excited Volume of Interest (VOI) was (A/P-R/L-H/F) 210 mm by 160 mm by 95 mm with a matrix of 42 x 32 x 20 resulting in 5 mm isotropic resolution. After reconstruction (Klauser A et al. Magn Reson Med. 2018), 3D MRSI data were quantified with LCModel to produce 3D metabolite maps. Concentration for total N-acetyl aspartate (tNAA), total creatinine (tCre), choline-containing compounds (Cho), myo-inositol (Ins), glutamate and glutamine (Glx) were calculated in every single voxel. A T1-weighted MPRAGE anatomical scan was acquired for positioning of the 3D MRSI and for the segmentation of the brain. For each participant, brain tissue was segmented into gray and white matter. Cerebral lobes and deep grey mater structures were also delineated using Freesurfer software package. CHR-P individuals were recruited in the service of child and adolescent psychiatry and in the service of general psychiatry, department of psychiatry at Lausanne university hospital. They were help-seeking adolescents and young adults aged between 14 and 35, who presented a psychosisrisk syndrome or basic symptoms as assessed by the Structured Interview for Psychosis-Risk Syndromes (SIPS) and the Schizophrenia Proneness Instrument, Adult (SPI-A) or Child & Youth version (SPI-CY). Healthy controls matched for age and sex were recruited in the general population. Results: Three-dimension MRSI provides spatial specificity by allowing main metabolites (i.e., tNAA, tCre, Cho, Ins and Glx) to be reliably mapped in the volume of the entire brain. The resulting contrast allows the recognition of brain compartments and subcortical structures. Individual brain segments, cerebral lobes and subcortical structures were registered to 3D MRSI data and the mean concentration in each structure was computed to allow group comparisons between CHR-P and HC. Discussion: In general, there is a strong need to develop new tools for the identification and stratification of CHR-P populations. Alterations of gross brain anatomy are relatively late events but early and subtle neurochemical changes and especially those reflecting oxidative stress and concomitant synaptic remodeling are promising candidates. This pilot study illustrates the potential of three-dimension MRSI to detect such alterations in the whole brain and with a good spatial resolution. [ABSTRACT FROM AUTHOR]