By using the method of graph theory analysis, the regional similarity of cortical thickness was calculated to ascertain the changes in topological characteristics of morphological networks of trauma-exposed patients. The relationship between these differences and clinical symptoms was also analyzed. As far as we know, this work is the first one to incorporate the structural covariance network of PTSD, TEC and HC groups simultaneously. We found that PTSD and TEC groups have different pathophysiological mechanisms, and the topological characteristics of brain network in both groups have changed significantly at different levels.
Human brain network is characterized by small-world attribute (Bullmore & Sporns, 2009), which is associated with high-efficiency information dissemination with low construction cost (Achard & Bullmore, 2007). The results of whole brain graph theory analysis showed that compared with HC group, PTSD patients retained the characteristics of small world topology. It is by now generally accepted that the small-world organization balances network isolation and network integration for information processing (Watts & Strogatz, 1998). Other results have also reported the maintenance of small-world properties in PTSD group (Zhu et al., 2019). It was notable that the small-world attribute was found to be impaired in TEC group. This suggested that trauma could destroy the brain network topology attributes to a certain extent. Although trauma exposure in itself does not meet the diagnostic criteria, it still disrupts the balance of brain isolation and integration. However, it is still unclear whether the changes in small-world properties caused by trauma are temporary or permanent. A study of earthquake survivors found that within 25 days after the earthquake, the survivors' small-world attributes was found decreased compared with control group, indicating a shift towards randomized network, but the damage returned to normal after two years (Du et al., 2015).
Although the overall small-world structure of PTSD patients was similar to that of TEC subjects, their local efficiency and clustering coefficient were higher compared to TEC group, which was in line with previous studies (Lei et al., 2015; Proessl et al., 2020). The increase of clustering coefficient denotes higher fault tolerance of the brain when under attack from outside (Brust et al., 2012). In view of the inverse relationship between local efficiency and characteristic path length, the increase in local efficiency of PTSD may reflect more effective communication within network. The increase in local efficiency limits the transmission of information to closely connected specific areas, indicating the increase of network separation in the brains of PTSD patients. The changes of these global topology indexes were positively correlated with the severity of symptoms, indicating that the increased network separation has clinical significance.
The randomization process of neural networks is thought to be a common pattern in some neurological diseases (Lynall et al., 2010; Stam et al., 2009). In contrast, the increased local efficiency and clustering coefficient in current study indicated a more regular brain network in patients with PTSD. Regular patterns of brain networks have also been found in children with PTSD (Suo et al., 2015). While the networks of TEC subjects became increasingly random.
In addition to the indicators of the whole brain network, we also explored the changes at nodal level. The nodal efficiency of the right isthmus cingulate cortex of PTSD was higher than HC subjects. The alterations in the isthmus cingulate cortex reflect the symptoms of general neurological diseases, namely the impairments in emotional stimulation processing and response capability (Nielsen et al., 2005). Research has shown that PTSD patients have difficulty in emotional response (Wei et al., 2017). It is interesting to note that the nodal efficiency of the right isthmus cingulate cortex in patients with PTSD was closely related to the severity. Moreover, the nodal centrality in the right superior frontal gyrus of PTSD patients was greater than that of HC group. A similar pattern of results was obtained in previous studies (Sun et al., 2018; Suo et al., 2015). The superior frontal gyrus belongs to the prefrontal cortex and plays a critical role in the emotional brain circuit (Lee et al., 2004). In PTSD group, the higher nodal centrality of the right superior frontal gyrus may represent the neural changes related to emotional response in the brain. It is not clear whether these changes in nodal centrality reflect a pre-onset abnormality of PTSD or a consequence of trauma, since no such change was found in TEC group.
Compared with TEC subjects, PTSD patients demonstrated increased nodal centrality in the left rostral middle frontal gyrus, bilateral supramarginal gyrus and right temporal pole. The rostral middle frontal gyrus is the key brain area of emotion regulation, which is related to fear conditioning (Michalski et al., 2017; Minamoto et al., 2015). The supramarginal gyrus, which belongs to the parietal cortex, is related to emotional processing (Lee et al., 2004). The temporal pole is thought to be an extension of the limbic system (Giordana et al., 2005) that plays a key role in emotional processing (Ezzyat, 2007). In addition, it was also observed that these increased nodal centralities were positively correlated with CAPS scores, suggesting that these nodal lesions may be related to the severity of the disease.
The nodal centrality of the left postcentral gyrus in PTSD group and TEC group was lower than that of HC group, which suggested that this area was related to trauma. The postcentral gyrus is activated during the induction of basic negative emotions (Gilead et al., 2016; Ochsner et al., 2004). Previous studies found that the thickness of the postcentral gyrus decreased in PTSD patients (Lindemer et al., 2013; Liu et al., 2012). The nodal centrality in the left insula in the TEC group was higher than that of PTSD and HC group, suggesting that the left insula may be a protective factor against PTSD. Previous studies also found that the nodal degree of bilateral insula in trauma survivors increased significantly (Du et al., 2015). The insula is related to traumatic memory (King et al., 2009). The increase of nodal centrality of the left insula may improve the resistance of network topology to some extent, thereby reducing the damage of trauma. A previous study found that higher regional cerebral metabolic rate of glucose (rcMRglu) in the anterior insula was associated with stronger resilience, which suggest a role of rCMRgu in the anterior insula in resisting negative effect of trauma (Jeong et al.).
Limitations
There were some limitations in this study. Above all, the results of nodal attributes didn’t survive correction, which might be due to sample constraints and limited capacity for statistical analysis. Expanding the sample size ought to be premeditated in future investigations. Secondly, the current research is only for typhoon victims, so it is necessary to be cautious in extending it to other types of trauma. Furthermore, as a cross-sectional study, it was uncertain whether the reported results existed pre-onset or were consequences of the diagnosis. In the future, longitudinal researches are expected to define the causal connection between brain network topology changes and PTSD.