SA91. Striato–Nigro–Striatal Tract Dispersion Abnormalities in Patients With Chronic Schizophrenia

Abstract

Background: The white matter (WM) connections between the substantia nigra and the striatum are part of a complex neural system involved in reward-based learning, a process that is impaired in patients with schizophrenia. Most diffusion magnetic resonance imaging (dMRI) studies in schizophrenia have focused on the corticocortical and corticosubcortical networks. To our knowledge, however, the striatonigrostriatal (SNS) tract has not yet been explored. Furthermore, the use of conventional dMRI indices to study WM, such as fractional anisotropy (FA), has limitations, as they reflect different biological features of WM. The present study aimed to use dMRI to delineate the SNS tract and to compare the application of 2 dMRI measures, Tract Dispersion (TD), an index of WM morphology, and Fractional Anisotropy (FA), an index of WM integrity, to detect group differences in chronic schizophrenia (CSZ) and healthy controls (HCs). Studying the morphology of WM in CSZ is important, as it is a feature reflective of brain development. Methods: dMRI scans were acquired in 22 male patients with CSZ and 23 age-matched HCs. Two-tensor tractography was used in addition to manually delineated regions of interest to extract the SNS tract. We computed a measure of tract dispersion, which reflects the degree to which tracts deviate from being parallel as in the case of fanning fibers. More specifically, at each point along each fiber, this method computes the variation in a fiber’s tangent vector in all directions orthogonal to the curve, which leads to a measure of local white matter dispersion. A mixed-model analysis of variance was used to investigate differences in mean TD and mean FA between CSZ patients and HCs. The associations between TD and FA and clinical and neuropsychological measures were evaluated. Results: We found a significant group difference in TD (P = .04) but not in FA. Post hoc t-tests revealed that this difference was driven by a reduced TD in the right hemisphere in HCs. In CSZ, there were significant positive correlations between TD and a subtest of apathy in the Scale for the Assessment for Negative Symptoms, r = .47, P = .02. In HCs, but not in CSZ, a trend toward significance was seen between TD and performance in the Iowa Gambling Task, r = .58, P = .07. Conclusion: In the SNS tract, we found the diffusion measure TD to be more sensitive than FA in detecting a group difference between CSZ and HCs. This finding may indicate a morphological distortion of the precise spiraling pattern of the SNS tract as described by Haber et al. (2014). The association between TD and the severity of apathy, a symptom that reflects a diminished response to reward, is consistent with the role of the SNS tract in reward prediction. Together, these findings provide further support for the neurodevelopmental and disconnectivity theories of schizophrenia.