To the best of our knowledge, this is the first neuroimaging study that examines the microstructural changes in WM and its associations with psychological symptoms in Shidu parents. Herein, we showed reduced integrity in left UF and SLF in the Shidu group, which was negatively correlated with SCL_90 and its subscales (including anxiety, depression, and interpersonal sensitivity), and such microstructural changes also varied by gender. Taken together, our results may provide insight into mechanisms underlying the early neurobiological response of the WM microstructure to the Shidu trauma in Shidu parents.
DTI is a functional magnetic resonance imaging method, which can indirectly obtain the information on the microstructure integrity of WM fiber by analyzing the diffusion characteristics of water molecules. Interestingly, the quantitative indicators, such as FA, AD, RD, and MD, are used to quantify the directionality and extent of restricted diffusion of water molecules (Alexander et al., 2011). FA can be used to detect early neuropathological changes in the WM microstructure that reflects the fiber density, axonal diameter, and myelination (Beaulieu, 2002). Unlike FA, MD reflects the average diffusion ability of water molecules in all directions and is sensitive (but not specific) to membrane integrity (Song et al., 2003). Other DTI parameters, such as AD and RD, might yield potentially specific information. AD measures the diffusion coefficients along the direction of maximum diffusivity, reflecting changes in the restrictive barriers along a fiber bundle, including axonal diameter and density, while RD measures the diffusion coefficients perpendicular to the direction of maximum diffusivity, mainly reflecting the changes in axonal membranes and myelin (Song et al., 2002). In the present study, the Shidu parents showed reduced FA in left UF and SLF, as well as decreased AD and increased RD/MD in the same WM regions. While both the AD decrease and RD increase tend to be associated with decreased FA, the difference in AD between the two group was not statistically significant (p < 0.05) in current study. Therefore, a decrease in FA could be attributed to the increase in RD, suggesting that the trauma of losing an only child mainly leads to the injury of axon membrane and myelin but has little effect on the integrity of axonal diameter and density. Also, increased MD was observed in the left UF and SLF in the Shidu group, which further indicated the impairment of the axonal membrane.
The UF is a thin fan-shaped WM fiber, which connects the orbital, medial, and lateral prefrontal cortex with rostral temporal areas, including the temporal pole, hippocampal gyrus, and amygdala (Schmahmann et al., 2007). The medial prefrontal cortex (mPFC) is speculated to play a major role in emotion regulation, including the appraisal and expression of negative emotion and modulating the activity of limbic regions involved in generating emotional responses (Etkin et al., 2011). Decreased integrity of UF might lead to an excessive fear response due to the impaired top-down regulation of the mPFC to the amygdala, while also affecting the information communication between the mPFC and hippocampus, a structure vital for contextualizing, thereby resulting in enhanced fear generalization and impaired contextual extinction learning (Liberzon & Abelson, 2016). This observed change of UF is in line with the prevailing neurocircuitry model of PTSD (Rauch, Shin, & Phelps, 2006). Furthermore, Luo et al (Luo et al., 2016) has previously demonstrated that the Shidu parents without PTSD had reduced hippocampal volume as compared to healthy controls. In the present study, the FA value of left UF was significantly decreased in Shidu parents without any mental disorder as compared to HCs, which further suggested that the trauma of Shidu was extremely painful and could independently lead to impaired integrity of WM tract linking the left hippocampus to the ipsilateral mPFC.
FA reductions in UF has also been reported to be associated with increased symptoms such as fear, anxiety, depression, and social-emotional function (Anand et al., 2005; Fani et al., 2015; Phan et al., 2009; Von Der Heide, Skipper, Klobusicky, & Olson, 2013). These findings are supported by results from the present study that shows a negative correlation between the FA values of the left UF and SCL_90 subscores (including anxiety, depression, and interpersonal sensitivity) in the Shidu parents. Anxiety is characterized by chronic, nonspecific apprehension and arousal related to the potential occurrence of future threats (Stober, 1993). Neurobiological theories of anxiety have highlighted the central role of the amygdala in the generation and experience of the fear that can give rise to anxiety, which is typically characterized by the reduced top-down regulation of amygdala activity by the mPFC (Kim et al., 2011). Cortico-limbic activity and connectivity are fundamental to emotion processing and mood regulation; a decreased cortical regulation of limbic activation in response to negative stimuli has been consistently implicated in imaging studies of depressive disorder (de Kwaasteniet et al., 2013). Moreover, interpersonal sensitivity is a strong psychological state of alertness, suspicion, and jealousy due to discomfort and depression in social interaction. Shidu parents tend to narrow their social circles and avoid socializing with families with a living child, mainly manifested as self-isolation and the act of escaping from a familiar region, which is closely related with the interpersonal sensitivity (Jonas, Scanlon, Rusch, Ito, & Joselow, 2018). Although these Shidu parents are free of any diagnosis of psychosis in this study, they share the same neuropathologic changes with depression, anxiety and PTSD, which might partially explain why these parents are more prone to various mental disorders than HCs. Therefore, we speculated that decreased integrity of left UF might influence the information transmission between mPFC and limbic regions, which in turn, impairs the emotion regulation, leading to a range of emotional symptoms in Shidu parents.
Moreover, decreased integrity of left UF was negatively correlated with the duration of Shidu, indicating that the chronic trauma aggravated the damage to WM microstructure integrity. Notably, this correlation only existed in the duration of Shidu within 10 years. Previous studies have shown that the overall WM FA values decrease with age (Hsu et al., 2008), and the intracranial fraction of WM decreases significantly in the age group > 59 years old (Guttmann et al., 1998). However, in the current study, no significant correlation was established between WM integrity and age, which might be due to the influence of the trauma weakening the effect of age. As the duration further extended to > 10 years, the atrophying effect of age on WM increased, thus weakening the effect of trauma.
The decline in FA was also found in the left SLF. As a major tract connecting the parietal, occipital, and temporal lobes with ipsilateral frontal cortices, the SLF facilitates the formation of a bidirectional neural network that is essential for core processes, such as attention, memory, emotions, and language (Kamali, Flanders, Brody, Hunter, & Hasan, 2014). Reportedly, the diminished integrity of the left SLF is associated with aphasia (Ivanova et al., 2016), while disruption of the right SLF is responsible for the attention loss in attention-deficit/hyperactivity disorder (Wolfers et al., 2015). The current study identified FA reductions in the left SLF in Shidu parents as compared to HCs; and it was mainly observed in the temporal part, suggesting that the WM integrity in this area was compromised. Furthermore, being an auditory language center, the temporal lobe is also involved in the learning and memory processes, with a key role in the hippocampus. Additionally, the hippocampus is crucial for contextualizing, and its disruption may result in enhanced fear generalization and impaired contextual extinction learning (Liberzon & Abelson, 2016), manifested as symptoms such as anxiety, fear, and interpersonal sensitivity in Shidu parents.
Gender analysis revealed that Shidu women had reduced integrity in the left UF, while lower integrity was found in Shidu men in the left SLF. Such differences may be related to the different emotion regulation strategies adopted by men and women with respect to Shidu trauma (Nolen-Hoeksema, 2012). Emotion regulation refers to strategies through which the individuals influence their experience and expression of emotions. Emotion dysregulation has been increasingly viewed as central to several types of psychopathology, including depression, anxiety, and PTSD (Etkin et al., 2011). The two common emotion regulation strategies are reappraisal, a cognitive strategy for reframing the context of an emotional experience, and suppression, a behavioral strategy for inhibiting emotional responses (Zuurbier, Nikolova, Åhs, & Hariri, 2013). A social psychology study showed that women use emotion regulation strategies, especially “reappraisal” largely and are more inclined to experience, express, and dwell on their emotions than men (Nolen-Hoeksema, 2012). They constantly pay attention to their negative emotions and their causes and consequences without participating in problem-solving, which might partially explain why women are more prone to PTSD, depression, and general psychological problems. On the other hand, men tend to suppress or avoid both the experience and expression of emotions and hence, might be more likely to engage in problem-solving and reappraisal in attempts to control or change the situations that they believe are driving their emotions than the women (Tamres, 2002). A DTI study on 194 normal adults found that the frequency of using typical “reappraisal” strategies in daily life was positively correlated with the FA of the left UF in the females but not males (Zuurbier et al., 2013). This finding was consistent with the results from the present study, suggesting that the left UF plays a vital role in the emotion regulation of females, and the temporal part of the left SLF may play a role in the “suppression” strategy of males. However, further studies are required to substantiate these findings.
The disrupted integrity of WM microstructure has presented left hemispheric lateralization in this study. Gender and handedness are considered to be the two major factors affecting the structural and functional asymmetries of the human brain (Eviatar et al., 1997). Recently, a functional MRI study has demonstrated that all the right-handers had left hemispheric lateralization in language-related WM tracts (including UF and SLF) (Silva & Citterio, 2017). Moreover, our recent voxel-based morphometry (VBM) study using the same subjects revealed that increased gray matter volume (GMV) was confined to the right hemisphere in Shidu parents without PTSD as compared to the HCs. Therefore, we speculated that the GMV changes in the right hemisphere may occur secondary to the damage of the WM microstructure and serve as an effective structural compensation to resist the development of PTSD. Thus, the impaired WM microstructure integrity might imply an earlier structural change, and the left hemisphere lateralization might be related to the weakened language function caused by the long-term social isolation of Shidu parents.
3.1 Limitations
Nevertheless, the present study had some limitations. First, this study did not set up a “Shidu with PTSD” group. Our previous VBM study using the same subjects has shown significant differences in GMV between Shidu parents with and without PTSD. Thus, by setting up an additional PTSD group, we could gain an in-depth insight into the microstructural changes of WM. Additionally, this was a cross-sectional study and did not allow for causal inferences. Therefore, it was crucial to conduct research in the course of the disease to determine whether the decreased WM integrity in Shidu parents occurred before the trauma onset or subsequent to the trauma of losing an only child. Moreover, we did not conduct an analysis to further examine the emotion regulation strategies or symptom measures within gender groups, thus the results of gender difference in WM integrity should be interpreted more carefully, and future studies would benefit from this consideration.