85. Diffusion Imaging of White Matter Pathways in Schizophrenia: Are Illness-Linked Changes Progressive?

Abstract

Listen

Translate article

Background: The presence of altered white matter microstructure in disorders involving psychosis is well established, but the nature and developmental course of these changes are not well understood (Peters & Karlsgodt, 2015). It is not clear whether the presence of attenuated frontal white matter anisotropy at any phase of illness indexes a pathophysiological change that occurs only within the prodromal and/or first episode phases, or whether early white matter changes are part of a lifespan trajectory of accelerated white matter degeneration in patients (Schwehm et al., 2016; Wright et al., 2014). It also remains to be determined whether schizophrenia-linked anisotropy changes are the result of axon damage or if they also reflect organizational changes. Methods: Diffusion MRI data were collected in 81 schizophrenia patients (54 First Episode and 27 Chronic), and 65 controls (Mean age = 26, 98 males, N = 146). Tract Based Spatial Statistics (TBSS; Smith et al., 2006) were performed to examine group differences in Fractional Anisotropy (FA; a widely used measure of white matter coherence) on a whole-brain voxel-based level. Average FA was extracted from a whole-brain-derived ROI, and subsequently used to examine whether the association between Age and FA differed as a function of illness. Fiber density (“fixel-based”) analysis was also performed (Raffelt et al., 2016). Fixel-based analysis leverages anatomical and crossing-fiber information to assess whether group differences are likely to be attributable to loss of white matter fiber bundle density or cross-section. Results: Compared to controls, the schizophrenia group displayed a pattern of reduced FA in the bilateral frontal lobes and the corpus callosum (FWE-corrected, P < .05). ROI analysis indicated a lack of significant interaction between Group and Age on FA in whole-brain significant regions (P = .81). Preliminary results from whole-brain analyses of fiber density indicate a trending inverse association between fiber density in the postcentral corpus callosum (FWE-corrected P = .08), but no such trend was evident in the frontal regions identified by TBSS analysis of FA. Conclusion: These results replicate the oft-reported finding that frontal and callosal FA is lower in schizophrenia. Cross-sectional analysis of Age and FA indicates that patients do not display a steeper rate of age-related FA decline compared with controls, suggesting that the group difference in FA manifests before the onset of psychosis, but does not progress across phase of illness. Preliminary assessment of fiber density raises the possibility that axon loss within postcentral callosal fiber bundles may be partly responsible for FA changes in postcentral callosal fibers. The lack of group differences in frontal fiber density supports the notion that schizophrenia-linked FA changes in these white matter regions are unlikely to be due to axon loss, but rather, to organizational changes.