Inferior Frontal Volume Reduction in Chronic Schizophrenia: An MRI Study Markant, D.(1,2), Levitt, J.J.(1,2), Cohen, A.(1,2), Bouix, S.(1,2), Niznikiewicz, M.(2), McCarley, R.W.(2), Shenton, M.E.(1,2) (1) Psychiatry Neuroimaging Laboratory, Brigham and Women’s Hospital, Harvard Medical School (2) Clinical Neuroscience Division, Laboratory of Neuroscience, Boston VA HealthCare System, Department of Psychiatry, Harvard Medical School Background Manual Tracing  The left (IFG) is a prefrontal region that includes Broca’s area, which is involved specifically in To confirm the volume difference in the left inferior frontal gyrus found using VBM, the IFG was manually traced in 52 subjects from the same syntactic language expression and verbal memory. The left IFG has been shown to be structurally (Suzuki 2005) sample (27 NC and 25 HC) in both hemispheres. The IFG was then subdivided into three subregions: pars orbitalis, pars triangularis, and pars and functionally (Kubicki 2003, Ragland 2004) abnormal in schizophrenia. opercularis. The ROIs were traced on an automated tissue segmentation to improve the reliability of the volume measurement. Tracing was done using 3D Slicer (www.slicer.org) using boundary definitions adapted from previous volumetric studies of the IFG (Suzuki 2005, Tomaiuolo 1999).  The IFG is composed of three subregions separable by sulcal landmarks: pars orbitalis, pars triangularis, and pars opercularis. Though the functional differences of the subregions are not entirely clear, they are ROI Definition Inferior Frontal Inferior Frontal Sulcus Precentral Sulcus cytoarchitectonically distinct.

 Voxel-Based Morphometry is a whole-brain comparison of brain tissue composition between groups. A VBM comparison of healthy controls (HC) and chronic schizophrenic (SZ) patients showed volume reduction of the right thalamus and left inferior frontal gyrus, specifically the pars opercularis subregion, in the SZ group.

 To confirm the volume reduction in the left inferior frontal gyrus, the ROI was manually traced on a subset of the same sample. The left and right inferior frontal gyri were divided into three subregions, the pars orbitalis, pars triangularis, and pars opercularis, based on sulcal boundaries. Voxel Based Morphometry Voxel-Based Morphometry (VBM) provides a whole-brain, voxel by voxel comparison of brain tissue Lateral Circular Insular Sulcus Horizontal Ramus Vertical Ramus composition that is rapid and unbiased. After MR images of subject brains are warped to a standard template to correct for global anatomical differences, a voxel-wise comparison is created in the form of a statistical Superior and inferior boundaries: parametric map (SPM) to assess group differences in density. The inferior frontal sulcus was the superior boundary for the entire IFG. The inferior boundary was the circular insular sulcus in all slices where the insula was visible, and the lateral orbital sulcus in all other slices. In Optimized VBM, the method is improved by the use of: 1) a study-specific template brain for more accurate spatial normalization, 2) study-specific tissue class priors and skull removal for better tissue segmentation, Anterior and posterior boundaries: Pars Opercularis and 3) modulation of voxel intensity by the spatial warping parameters, allowing for a comparison of volume The anterior IFG was traced on coronal slices from the most anterior coronal slice where the inferior frontal sulcus was change between groups. visible to the plane of the vertical ramus. The posterior IFG was drawn on axial slices from the plane of the vertical ramus to the precentral sulcus. An implementation of Optimized VBM was run using MR images of a group of 30 healthy controls and a group of 30 chronic, medicated Pars Triangularis schizophrenics. Scans were acquired using an SPGR sequence on a 1.5-T General Electric scanner. All subjects were males between the Subregions: ages of 20 and 54 and matched for IQ and parental socioeconomic status. Gray matter VBM analysis was conducted using SPM2 The pars orbitalis was bounded superiorly by the horizontal ramus of the sylvian fissure. The pars triangularis was framed (Statistical Parametric Mapping, Wellcome Institute, London). The VBM software was adapted from scripts written by Christian Gaser and by the horizontal and vertical rami. The pars opercularis was bounded by the vertical ramus and the precentral sulcus. John Ashburner (2004). Regions of the brain were localized using Pickatlas in SPM2 (Maldjian et. Al 2003). The volumes of the pars orbitalis (pink in figure), pars triangularis (orange), and pars opercularis (green) were recorded Pars Orbitalis VBM Results and corrected for total intracranial volume. A one-way analysis of variance (ANOVA) was used to evaluate differences in volume of the subregions of the IFG for both left and right sides. MNI Coordinates Voxel-level p- Cluster-level p- of Cluster value (FDR- value Region Maximum corrected) (corrected) Left inferior frontal -50, 4, 16 0.051 < 0.05 Manual Tracing Results gyrus Right thalamus 4, -22, 4 0.066 < 0.05 Pars Opercularis Pars Triangularis Pars Orbitalis

 Modulated analysis demonstrated a significant decrease in volume in the right thalamus and left inferior frontal gyrus in * the schizophrenic group compared to normal controls (see table above):

 Non-modulated analysis did not show a significant decrease (p >  The left pars opercularis was significantly smaller in SZ than in HC (p = 0.027). The right side was not significantly different (p = 0.37).  SPM showing regions of decreased gray matter 0.4, FDR-corrected) or increase (p = 1.0, FDR-corrected) in gray There were no significant volume differences between diagnostic groups in the pars triangularis (left: p = 0.416; right: p = 0.816) or pars orbitalis (left: p = 0.161; right: p = 0.897). volume in schizophrenia group compared to matter density in schizophrenics compared to normal controls control group  There was no significant volume difference in whole IFG volume (left: p = 0.252; right: p = 0.341). after correcting for multiple comparisons. Conclusions References  Using two different methods, we found convergent evidence of a volumetric difference in the left pars opercularis of the left inferior  Ashburner J., Friston K. Voxel-Based Morphometry--The Methods. NeuroImage, 2000; 11:805-821. frontal gyrus in schizophrenia.  Honea, R., Crow T., Passingham D., Mackay C. Regional Deficits in Brain Volume in Schizophrenia: A Meta-Analysis of Voxel-Based Morphometry  Previous manual tracing studies of schizophrenia have found a volume reduction of the whole left IFG, but did not measure the three Studies. Am J Psychiatry, 2005; 162:2233-2245.  Kubicki M, McCarley RW, Nestor PG, Huh T, Kikinis R, Shenton ME, Wible CG. (2003). “An fMRI study of semantic processing in men with subregions separately. Unlike those reports, in this sample there was no overall difference in whole IFG volume between groups, but a schizophrenia.” NeuroImage 20; 1923—1933. regional reduction confined to the left pars opercularis.  Maldjian J.A., Laurienti P.J., Kraft R.A., Burdette J.H. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of  Previous studies of schizophrenia have shown widespread dysfunction in the language network, including structural and functional fMRI data sets. NeuroImage, 2003; 19(3):1233-1239.  Ono M, Kubik S, Adernathey CD. (1990). Atlas of the Cerebral Sulci. Thieme, New York. differences in the superior temporal, dorsolateral prefrontal, and inferior frontal cortices. In addition, there is growing evidence of  Ragland JD, Gur R, Valdez J, Turetsky B, Elliott M, Kohler C, Siegel S, Kanes S, Gur RE. (2004). “Event-Related fMRI of Frontotemporal Activity impaired connection of these regions via frontotemporal white matter tracts. Measurement of the subregions of the IFG can help to During Word Encoding and Recognition in Schizophrenia.” Am J Psychiatry 161:1004-1015. define more precisely the structural differences in schizophrenia and their relation to the disorganized thought and other cognitive  Suzuki M, Zhou SY, Takahashi T, Hagino H, Kawasaki Y, Niu L, Matsui M, Seto H, Kurachi M. (2005). “Differential contributions of prefrontal and impairments found in the disorder. temporolimbic pathology to mechanisms of psychosis.” Brain 128: 2109—2122.  Tomaiuolo F, MacDonald JD, Caramanos Z, Posner G, Chiavaras M, Evans AC, Petrides M. (1999). “Morphology, morphometry and probability  Future work will include 1) further comparison of VBM and manual tracing results, 2) assessment of sulcal variability in the inferior mapping of the pars opercularis of the inferior frontal gyrus: an in vivo MRI analysis.” frontal gyrus, and 3) correlation of the volume reduction with clinical and cognitive measures.