Voxel-based diffusion tensor analysis reveals aberrant anterior cingulum integrity in posttraumatic stress disorder due to terrorism

Abstract

Recent functional neuroimaging work has suggested that interregional functional connectivity between the anterior cingulate cortex (ACC), other limbic, and prefrontal regions may be involved in the pathophysiology of posttraumatic stress disorder (PTSD). However, less attention has been paid to the white matter network. Voxel-based analysis enables an exploration of morphological or functional changes throughout the entire brain. Here we undertook the first application of this technology to diffusion tensor data in patients with PTSD. Participants were 9 victims of the Tokyo subway sarin attack with PTSD and 16 matched victims of the same traumatic event without PTSD. The voxel-based analysis showed a significant fractional anisotropy increase in the left anterior cingulum, subjacent to the left ACC gray matter where we previously found a volume decrement, in PTSD subjects. Moreover, the severity was positively, but not significantly associated with the fractional anisotropy of the left anterior cingulum in the victims with PTSD, using the region of interest defined in the native space with the inverse normalization technique. The present study demonstrated further evidence of abnormalities of both the ACC, a structure that is pivotally involved in attention, emotional regulation, and fear conditioning, and of subjacent white matter in the pathology of PTSD.

Introduction

Posttraumatic stress disorder (PTSD) is characterized by recurrent trauma-related memories, and increased fear response and physiological reactivity to reminders of the trauma, coupled with sleep disturbances, nightmares, avoidance, increased startle, and other symptoms that can persist for many years after the original traumatic event (Bremner et al., 1995a, Pitman, 1989). Previous neuroimaging studies have revealed that limbic, paralimbic, and prefrontal structures may play a key role in the pathophysiology of PTSD (Bremner et al., 1999a, Bremner et al., 1999b, Lanius et al., 2001, Shin et al., 1999, Yamasue et al., 2003), although some studies have suggested otherwise (Bonne et al., 2001, Fennema-Notestine et al., 2002). In our recent study, voxel-based morphometry (VBM) showed a significant gray-matter volume reduction in the left anterior cingulate cortex (ACC) in the victims of the Tokyo subway sarin attack with PTSD, compared with those without PTSD. Together with previous functional neuroimaging studies showing a dysfunction of this region (Bremner et al., 1999a, Lanius et al., 2001, Shin et al., 2001), our study provided further support for the important role of the ACC, which is pivotally involved in attention, emotional regulation and fear conditioning, in the pathology of PTSD.

A growing body of functional neuroimaging work has suggested involvement of interregional functional connectivity between the ACC and other limbic and prefrontal regions in the pathophysiology of PTSD (Gilboa et al., 2004, Lanius et al., 2004, Shaw et al., 2002, Shin et al., 2004). For example, Lanius et al. reported significant between-group differences in interregional brain activity correlations between the ACC and several prefrontal and parietal regions during the recall of traumatic memories in traumatized subjects with and without PTSD (Lanius et al., 2004). These studies may implicate disrupted neural networks through white matter fiber tracts in the pathophysiology of PTSD. However, relatively little attention has been paid to the possible role of the white matter network in the pathophysiology of PTSD.

A newer MRI method such as diffusion-weighted imaging, which is sensitized to the translational motion of water molecules, can play a pivotal role in investigating structures of cells, membrane permeability, transport processes and temperature in vivo—data which are not accessible with conventional MRI (Le Bihan et al., 1986, Stejskal and Tanner, 1965). In biological systems, the diffusional motion of water is impeded by tissue structure such as cell membranes, myelin sheaths and intracellular microtubules and associated proteins. An important and potentially informative consequence of the interaction between water molecules and sub-cellular structures is “diffusional anisotropy,” which refers to the finding that diffusion is less hindered when parallel relative to an axon or myelin sheath than when perpendicular (Moseley et al., 1990). Anisotropy measurement using diffusion tensor analysis is a promising method to non-invasively detect the degree of fiber damage in various diseases or maturation processes affecting white matter (Abe et al., 2002, Klingberg et al., 2000, Sach et al., 2004, Werring et al., 2000).

To our knowledge, there has been no diffusion tensor study investigating subjects with PTSD using a voxel-based approach. The use of a voxel-based approach has an advantage in searching for abnormalities through the entire brain, compared with manually drawn region of interest (ROI) analysis. There is also the question of the validity of investigator-determined ROIs, given the inter-individual variability, even though ROIs can be defined so that their measurement is highly reliable. Voxel-based morphometry has provided more objective and reliable results, independent from operator bias (Kubicki et al., 2002, Richardson et al., 1997, Suzuki et al., 2002, Vargha-Khadem et al., 1998, Wright et al., 1999), although some studies have suggested otherwise (Tisserand et al., 2002). In contrast, voxel-based diffusion tensor analysis is an emerging method (Abe et al., 2004, Burns et al., 2003, Eriksson et al., 2001, Foong et al., 2002, Rugg-Gunn et al., 2001, Sach et al., 2004).

In this voxel-based diffusion tensor study, it was hypothesized that terrorism victims with PTSD would show abnormal diffusional anisotropy in the ACC, hippocampus, and adjacent cortex or subcortical white matter compared with victims without PTSD on the basis of the findings of previous studies investigating morphology and functional alteration in subjects with PTSD (Bremner et al., 1995b, Gilbertson et al., 2002, Rauch et al., 2003, Yamasue et al., 2003). Furthermore, voxel-based analysis with statistical parametric mapping (SPM) could not provide a quantitative value but a statistical parametric map of the t-statistic in a voxel-wise manner. When the voxel-based diffusion tensor analysis revealed a significant difference between the PTSD and non-PTSD groups, we also performed an ROI analysis, using the inverse normalization technique supplied with SPM99.