Longitudinal loss of gray matter volume in patients with first-episode schizophrenia: DARTEL automated analysis and ROI validation

Abstract

Region of Interest (ROI) longitudinal studies have detected progressive gray matter (GM) volume reductions in patients with first-episode schizophrenia (FESZ). However, there are only a few longitudinal voxel-based morphometry (VBM) studies, and these have been limited in ability to detect relationships between volume loss and symptoms, perhaps because of methodologic issues. Nor have previous studies compared and validated VBM results with manual Region of Interest (ROI) analysis. In the present VBM study, high-dimensional warping and individualized baseline-rescan templates were used to evaluate longitudinal volume changes within subjects and compared with longitudinal manual ROI analysis on the same subjects. VBM evaluated thirty-three FESZ and thirty-six matched healthy control subjects (HC) at baseline (cross-sectionally) and longitudinally evaluated 21 FESZ and 23 HC after an average of 1.5 years from baseline scans. Correlation analyses detected the relationship between changes in regional GM volumes in FESZ and clinical symptoms derived from the Brief Psychiatric Rating Scale, as well as cognitive function as assessed by the Mini-Mental State Examination. At baseline, patients with FESZ had significantly smaller GM volume compared to HC in some regions including the left superior temporal gyrus (STG). On rescan after 1.5 years, patients showed significant GM volume reductions compared with HC in the left STG including Heschl's gyrus, and in widespread brain neocortical regions of frontal, parietal, and limbic regions including the cingulate gyrus. FESZ showed an association of positive symptoms and volume loss in temporal (especially STG) and frontal regions, and negative symptoms and volume loss in STG and frontal regions. Worse cognitive function was linked to widespread volume reduction, in frontal, temporal and parietal regions. The validation VBM analyses showed results similar to our previous ROI findings for STG and cingulate gyrus. We conclude FESZ show widespread, progressive GM volume reductions in many brain regions. Importantly, these reductions are directly associated with a worse clinical course. Congruence with ROI analyses suggests the promise of this longitudinal VBM methodology.

Figure 1

Image processing of the longitudinal study using a Diffeomorphic Anatomical Registration Through Exponentiated Lie algebra (DARTEL) tool in Statistical Parametric Mapping (SPM) 5. Abbreviations: GM, gray matter; MNI, Montreal Neurological Institute Step 1: Both baseline and follow-up scan T1-weighted images were realigned manually according to the AC-PC line and midsagittal plane. The baseline image was manually coregistered to the follow-up scan image without reslicing. Step 2: All the T1-weighted images were segmented into probability maps of GM, white matter (WM) and cerebrospinal fluid by the unified segmentation approach in SPM5. The resulting GM and WM probability maps were automatically rigidly aligned (3 rotations and 3 translations) to MNI space and resampled into 1.5mm isotropic voxels. Note: We did not describe WM maps in this figure because WM maps were only used to achieve better registration. Only GM maps were used for statistical analysis. Step 3: To evaluate the longitudinal morphometric changes within each subject, a template was created for each subject using the information of both GM and WM maps. This template was obtained by combining the GM/WM maps of the baseline and follow-up scans into average GM/WM maps using an automated unbiased template building, non-linear registration program (DARTEL). The baseline and second scan GM/WM maps were then spatially normalized onto the corresponding subject-specific template non-linearly. The signal intensity of the normalized maps was modulated by the determinant of the Jacobian of the transformation to account for expansion and/or contraction of brain regions. Step 4: A population template was created by simultaneously non-linearly registering all subject-specific GM/WM templates using DARTEL. The baseline and follow-up scan normalized maps of each subject were spatially non-linearly normalized to the population template, and then modulated. Step 5: In order to bring the final analysis into standard MNI space, the population GM template was registered automatically to the MNI space through an affine transformation. All the individual GM maps residing in the population template space were then co-registered to MNI using the same affine transformation. Finally, these GM images were smoothed with an 8-mm FWHM Gaussian kernel. The information of the WM maps was not used in this step as it was only used to achieve better registration in steps 1-4, but not incorporated in the statistical analysis.

Figure 2

Regions of reduced gray matter volume at baseline in the 33 patients with first-episode schizophrenia compared with the 36 healthy control subjects. Glass brain images include all the volume reduced regions (A). A coronal slice shows volume reductions of left superior temporal gyrus (STG) including some insula, hippocampus, and bilateral thalamus (B). Sagittal slices show volume reductions of the left STG including some insula (C), left amygdala-hippocampus complex (D), right anterior cingulate gyrus (ACG, rostral subregion) with some superior frontal gyrus (SFG) overlap (E). Uncorrected threshold of p<.001 with an extent threshold of 70 voxels is applied for graphical reporting (see supplemental Table S1 for specific coordinates). Abbreviations: L/lt, left; R/rt, right; A, anterior; P, posterior

Figure 3

Figure 3-A, B, C and D show progressive gray matter volume reductions greater in the 21 patients with first-episode schizophrenia compared with the 23 healthy control subjects over the 1.5 year follow-up interval. These regions include the bilateral superior temporal gyrus (STG), precentral gyrus, and left anterior cingulate gyrus (ACG) (A), left ACG (dorsal, rostral, and subgenual subregions), posterior cingulate gyrus (PCG), and superior frontal gyrus (SFG), and right ACG (dorsal and rostral) and SFG (B), left and right insula (C). Figure 3-D shows widespread gray matter volume reductions in the bilateral frontal, temporal and parietal regions (D) (see Table 3 for specific coordinates). Uncorrected threshold of p<.001 with an extent threshold of 70 voxels is applied for graphical reporting. Abbreviations: L/Lt, left; R/Rt, right; A, anterior; P, posterior

Figure 4

Correlation between percentage change of regional gray matter (GM) volumes and absolute changes in thinking-disturbance (positive symptom), hallucinatory behavior, and unusual thought contents scores of Brief Psychiatric Rating Scale (BPRS). The lesser improvement of the thinking-disturbance (positive symptom) score was correlated with the greater longitudinal GM volume loss of the bilateral Heschl's gyrus (4-1). Lesser improvement of the hallucinatory behavior score was correlated with the greater longitudinal GM volume loss of the bilateral Heschl gyrus (4-2) and left Heschl's gyrus (4-3). Moreover, the lesser improvement of the unusual thought content score was correlated with the greater longitudinal GM volume loss of the right Heschl's gyrus (4-4). Abbreviations: lt, left; rt, right; bil, bilateral

Figure 5

Correlation between percentage change of regional gray matter (GM) volumes and absolute changes in withdrawal-retardation (negative symptom) score of Brief Psychiatric Rating Scale (BPRS). Lesser improvement of the withdrawal-retardation (negative symptom) score was correlated with the greater longitudinal GM volume loss of the left and right inferior frontal gyrus (IFG, 5-1, -2), bilateral insula (5-3), and left supramarginal gyrus (5-4). Abbreviations: lt, left; rt, right; bil, bilateral