**3. Results**

#### *3.1. Results from the Literature Search*

A total of 572 articles were identified after the literature search conducted under the databases PubMed and Web of Science. Upon the search and identification of records, duplicate records between the two databases were removed, followed by the screening of titles and abstracts of the studies. The screening process excluded a total of 254 studies as according to the eligibility criteria mentioned above in the Materials and Methods section. After careful consideration and full-text review of the remaining records according to the eligibility criteria, 75 studies were removed. After the inclusion of 11 studies through cross-referencing and reviewing of the records, and a total of 50 studies were included in this review. Individuals with PTSD who were participants of studies that were included in this review were exposed to various types of trauma, including accidents such as surviving a car crash, fire, or natural disaster, victims of various crimes including physical assault, sexual violence, or domestic abuse, exposure to combat such as military personnel or police o fficers, and other forms of trauma that were not specified. Among the 50 articles selected for this review, 43 were studies that explored serum inflammatory cytokine markers as measures of inflammation, 4 were studies exploring PTSD- and oxidative stress-related genes, while the remaining 3 were neuroimaging-based studies that explored the neural correlates of inflammation in PTSD. A detailed summary of the literature search process and the screening and inclusion of the eligible studies for the current review is shown in Figure 1.

**Figure 1.** Flowchart of literature search and identification of eligible articles for the current review.

#### *3.2. Alterations of Serum Inflammatory Markers in Association with PTSD*

Growing literature emphasizes the importance of regulation of mental health and stress in various disorders including PTSD due to the high prevalence of comorbidity with other chronic metabolic diseases [31]. For instance, the presence of PTSD has been noted to co-occur or associate with risks of chronic medical diseases including cardiovascular diseases and type 2 diabetes mellitus [31]. Considering the significance of cytokines as regulators of immune responses [32], the majority of studies explore the alteration and influence of inflammation in PTSD through the measurement of peripheral cytokine levels. In particular, serum cytokines are often measured and used as peripheral markers of inflammation in various psychiatric disorders including depression [33], panic disorder [34], bipolar disorder [35], and obsessive-compulsive disorder [36].

Cytokines are largely divided into two classes, which are the proinflammatory cytokines and the anti-inflammatory cytokines. As demonstrated via animal models, proinflammatory cytokine levels tend to reflect the promotion and occurrence of inflammation whereas anti-inflammatory cytokine levels are related to the suppressing of inflammation as well as the protection of further damage from inflammation [37]. For studies investigating inflammation in PTSD according to peripheral inflammatory markers, the current review identified and selected 24 studies that explored serum levels of proinflammatory cytokines including one or more of IL-1β, IL-6, TNF-<sup>α</sup>, and IFN-γ as well as CRP, and 19 studies that explored serum levels of anti-inflammatory cytokines IL-4 and/or IL-10.

#### 3.2.1. Roles of Proinflammatory Cytokines in PTSD: IL-1β, IL-6, TNF-<sup>α</sup>, IFN-γ and CRP

Proinflammatory cytokines tend to be the more prevalently used outcome variables, as increased proinflammatory cytokines have consistently demonstrated to indicate worsening of the immune system, and may therefore be reliable and direct indicators of disease progression [32]. In addition, despite serum proinflammatory cytokines being peripheral in nature, they have been described as mediators of many pathways underlying the CNS through its important roles in inflammatory response [38], therefore may provide valuable insight with regards to the pathophysiology of PTSD from a CNS perspective. Proinflammatory cytokines increase in concentration with age but also

may be elevated with stress as well as secondary symptoms of stress such as sleep and fatigue [24]. The current study summarizes the collective findings from three of the most prevalently explored proinflammatory cytokines in PTSD, which are IL-1β, IL-6, and TNF-<sup>α</sup>, all of which are often paired as they are all induced by the same endotoxin, lipopolysaccharides (LPS) [39], along with other inflammatory measures. Furthermore, among the 4 proinflammatory cytokines and CRP selected as part of this review, IL-6 downregulates the synthesis of IL-1β and TNF-α [40]. Considering that a distinct regulatory relationship is present among the inflammatory markers of interest of the current review, it may be presumed that alteration in the markers provides implications with regards to the potential direction of alteration in the other remaining cytokines as well.

Through the current literature selection, studies have demonstrated that individuals with PTSD have significantly elevated serum levels of proinflammatory cytokines than the respective control group without PTSD. Specifically, individuals with PTSD demonstrated as having higher levels of serum IL-1β [15,25], IL-6 [15,16,25,41–47], TNF-α [15,16,25,42–44,46–48], IFN-γ [15,47,49], and CRP [42,50] than individuals who have been exposed to trauma but have never developed PTSD. In the current review, IL-6 demonstrated to be the most documented proinflammatory cytokine in human models of PTSD, and the majority of studies demonstrated increased IL-6 levels in the serum sample of individuals diagnosed with PTSD as compared with their respective control group. This may also be in alignment with previous research which noted IL-6 as the most considerable marker for inflammation [21]. Specifically, studies have shown that IL-6 is crucial in the relationship between immune system and CNS in inflammatory states [44]. Furthermore, 4 studies have demonstrated that individuals with PTSD have elevated proinflammatory cytokine levels when compared with healthy individuals who have not been exposed to any trauma [51–54]. In addition, studies have also demonstrated elevated CRP levels which are known to be influenced by proinflammatory cytokines, where individuals with PTSD showed increased CRP levels compared to healthy controls [42], and PTSD symptom severity were positively associated with CRP levels [50,55].

Participants with PTSD resulting from surviving a fire and burn injury were assessed for proinflammatory cytokines including IL-1β, IL-6, IL-8, and TNF-αin a study by Jiang and colleagues [51]. Here, Jiang and colleagues (2018) have also noted that, individuals with severe burn injury and diagnosed with PTSD demonstrated decreased clinical symptoms of PTSD three months after the trauma exposure, while having remained elevated levels of proinflammatory cytokines [51]. In addition, a study by Gill and colleagues (2013) found that women who recover from PTSD have statistically similar IL-6 levels as healthy control individuals who have never been exposed to any trauma [52]. This may then indicate that, for individuals who are able to recover from PTSD, both the normalization of clinical symptoms of PTSD as well as the influences of oxidative stress and inflammation may be possible. Furthermore, the findings from Jiang and colleagues (2018) on the sustained elevation of proinflammatory cytokine levels despite having reduced in clinical symptoms of PTSD [51] may indicate that those who are susceptible or possibly at risk of chronic PTSD may have distinct inflammatory responsiveness, regardless of clinical symptomatology. Therefore, the distinct pattern between inflammatory measures and clinical measures may warrant the incorporation of both physiological data as well as clinical assessments to better understand the pathophysiology of PTSD.

The study by Wang and colleagues (2016) may provide supportive evidence to the potential distinct inflammatory pathways between individuals who are at risk for versus resilient to PTSD [46]. In particular, their findings indicated that the severity of the trauma experienced may not necessarily mediate or influence the relationship between level of serum proinflammatory cytokine IL-1β and post-traumatic stress symptom severity in the case of combat-related PTSD [46]. Among veterans all of whom have been exposed to similar combat, those who are diagnosed with PTSD demonstrated significant imbalance in their inflammatory state as compared to those without PTSD [46], implicating that one's risk or resilience towards PTSD may be largely explained by biological factors and their ability to regulate their inflammatory state, as opposed to the nature or severity of the traumatic event.

Another interesting finding from the studies identified is the supportive evidence of stress influencing proinflammatory cytokine levels even at a younger age group, as well as the potential predictive role of proinflammatory cytokine levels in the comorbidity of PTSD and other chronic metabolic diseases. In particular, Blessing and colleagues (2017) demonstrated increased indicators of systemic inflammation in individuals with PTSD as compared to the respective control group, where the inflammatory markers were also indicators of higher cardiovascular risk as well as insulin resistance [43], which may then provide additional perspective with regards to related diseases such as type 2 diabetes mellitus.

Furthermore, two studies also explored the additional factor of traumatic brain injury (TBI) in PTSD and found significant alteration in proinflammatory cytokine levels in association with PTSD. The first is the study by Devoto and colleagues (2017) who found that TBI may have an additive e ffect to the alteration of proinflammatory cytokine levels in the case of PTSD, as individuals with PTSD as well as TBI had demonstrated greater elevation in proinflammatory cytokine levels than individuals with PTSD but without TBI [44]. The study also demonstrated that, among individuals with both PTSD and TBI, the levels of proinflammatory cytokines were correlated with post-traumatic stress symptom severity, where serum proinflammatory cytokine levels of IL-6 and TNFα were higher in individuals with higher post-traumatic symptom severity. The second study is by Kanefsky and colleagues (2019), which further demonstrated that, even within individuals with both PTSD and TBI, the presence of loss of consciousness at the time of the traumatic brain injury may also influence the alteration in proinflammatory cytokine levels of IL-6 [45].

Contrasting findings were also reported in this literature search. For instance, studies have reported that victims of assault who are exposed to trauma and have been diagnosed with PTSD have statistically similar IL-6 cytokine levels as those who have been exposed to trauma but have not developed PTSD [56]. In addition, one study reported insignificant alteration in three of the serum proinflammatory cytokine levels between individuals with PTSD versus those without PTSD [57], and another study reported insignificant alteration in IL-6 in relation to PTSD [58]. Moreover, a study reported small but significant decrease in the levels of IL-6 and IFN-γ in individuals with PTSD as compared to those without PTSD [59]. However, it may be noteworthy that these studies were primarily focused on stress as measured by symptoms of insomnia, respectively, rather than the alteration in the general symptoms PTSD through oxidative stress as measured by the proinflammatory cytokines targeted in this review.

Similarly, McCanlies and colleagues (2011) reported that there was a lack of association between proinflammatory markers of IL-6 and CRP with onset of PTSD symptomatology in urban police o fficers with high PTSD symptoms [60]. There were also opposite findings reported in one study by Bruenig and colleagues (2018), which reported a positive correlation between post-traumatic stress symptom severity and reduced levels of serum TNFα [61]. Despite the numerous reports of contrasting findings in terms of levels of proinflammatory cytokines according to the diagnosis of PTSD, however, it is noteworthy that the majority of the literature selected for review upon the screening and assessing for eligibility were largely based on male populations. Considering that between-sex di fferences in the levels of proinflammatory cytokines IL-6 have been reported [54], where men with PTSD show significantly elevated IL-6 levels as compared to their respective healthy control group while the women did not, this may be an important factor to consider in the summary of the studies selected in this review.

A detailed description of the studies reviewed that investigated alteration in serum proinflammatory cytokine levels in relation to PTSD is provided in Table 1.




**Table 1.** *Cont.*

† Indicate studies that included a trauma-unexposed healthy control group. ‡ Depicts the sample size of the PTSD group only, as gender distribution for the control group was not provided. Abbreviations: BMI, body mass index; CRP, c-reactive protein; f, female; IFN, interferon; IL, interleukin; m, male; N, sample size; PTSD, post-traumatic stress disorder; Ref., reference; TBI, traumatic brain injury; TNF, tumor necrosis factor.

#### 3.2.2. Roles of Anti-Inflammatory Cytokines in PTSD: IL-4 and IL-10

In addition to proinflammatory cytokines, anti-inflammatory cytokines also play a pivotal role in the regulatory processes of oxidative stress and inflammation. In contrast to proinflammatory cytokines, which promote and induce inflammatory responses, the primary role of anti-inflammatory cytokines is to inhibit the synthesis of proinflammatory cytokines, and is therefore also described as inhibitors of inflammatory mediators [32].

While numerous anti-inflammatory cytokines are present, the current study reviewed the alteration in levels of IL-4 and IL-10, as it has been previously suggested as an important marker in psychosocial stress [27] as well as other chronic medical diseases including type 2 diabetes mellitus [62] and cardiovascular disease [63], both of which are more often than not closely associated with PTSD. For instance, previous studies have consistently reported that level of serum anti-inflammatory cytokine IL-10 is significantly associated with insulin sensitivity in young individuals, which may later develop towards type 2 diabetes mellitus [64] or chronic pain [65]. In addition, a recent study has also reported that chronic inflammation as observable by altered levels of anti-inflammatory cytokine levels including IL-4 and IL-10 may play a predictive role in disorders such as type 2 diabetes mellitus development [66].

Furthermore, IL-10 enables the suppression of synthesis of TNFα as well as IL-1β [34], which are proinflammatory cytokines primarily explored in models of PTSD, and may therefore provide a comprehensive overview of the specific inflammatory biomarkers that may represent the pathophysiology of PTSD. Considering that anti-inflammatory cytokines have also been suggested to be potential options of possible treatment of chronic disorders [67], an overview of the alteration in IL-4 and IL-10 may reveal important findings with regards to the distinct role of anti-inflammatory cytokines in PTSD.

A total of 19 studies were selected from the literature search process that explored the alteration in serum levels of IL-4 or IL-10 in PTSD. Contrary to expectations of the opposition roles between proinflammatory and anti-inflammatory cytokine alterations in response to oxidative stress and inflammation, findings from the selected studies have mostly demonstrated statistically similar levels of serum IL-10 between individuals with PTSD and trauma-exposed controls who have never been diagnosed with PTSD [15,16,25,44,45,57,68–70]. Among these studies that demonstrated no alteration in serum IL-10 also included those from Table 1 which demonstrated significant increased proinflammatory cytokine levels in the PTSD group as compared to their respective healthy controls [15,16,25,44,45], indicating that elevated proinflammatory cytokines in association with PTSD may not necessarily reflect reduced anti-inflammatory cytokines. Furthermore, for studies that included a healthy control group who have not been exposed to any type of trauma, consistent results were shown where individuals exposed to trauma – regardless of their diagnosis for PTSD—did not show any between-group di fferences in IL-10 levels as compared to the control [69,71].

However, it is noteworthy that contrasting findings were reported by de Oliveira and colleagues (2018) as well as Guo and colleagues (2012), where both proinflammatory cytokines including IL-6 and anti-inflammatory cytokines including IL-4 and IL-10 were positively correlated and elevated in individuals with PTSD as compared to the respective healthy control group [54,72]. Dennis and colleagues (2018) also found that PTSD symptom severity is associated with higher IL-10 levels, which are then mediated by vagal activity, smoking, and alcohol dependence [48]. However, this may be partially explained by the well-distributed sex ratio in this particular study of 87 males and 80 female, considering that previous studies have noted significant sex di fferences in the patterns of inflammatory marker levels. Elevated IL-10 levels in respect to the diagnosis of PTSD were also found in two studies [52,54], although these studies were heavily male-dominant and female-dominant in their populations, respectively. Considering that many studies demonstrated findings of reduced IL-10 levels in association with PTSD diagnosis [46,56,61], this may then sugges<sup>t</sup> that perhaps this anti-inflammatory cytokine are not a suitable marker that provides insight towards the pathophysiology of PTSD or presence of trauma exposure.

*Antioxidants* **2020**, *9*, 107

A detailed description of the studies reviewed that investigated alteration in serum anti-inflammatory cytokine levels of IL-4 and IL-10 in relation to PTSD is provided in Table 2.


**Table 2.** Studies exploring serum anti-inflammatory cytokines IL-4 and IL-10 alteration in PTSD.


**Table 2.** *Cont.*

† Indicate studies that included a trauma-unexposed healthy control group. ‡ Trauma types of the individuals in the PTSD group of this study were described as *other* for two individuals, *car accident* for five individuals, and miscellaneous forms of assault for the remaining individuals. § Depicts the sample size of the PTSD group only, as gender distribution for the control group was not provided. Abbreviations: CRP, c-reactive protein; f, female; IFN, interferon; IL, interleukin; m, male; mTBI, mild traumatic brain injury; N, sample size; PTSD, post-traumatic stress disorder; Ref., reference; TBI, traumatic brain injury; TNF, tumor necrosis factor.

#### *3.3. Roles of PTSD- and Oxidative Stress-Related Genetic Markers in PTSD*

Five studies explored specific genes that have been previously suggested to be of significance in PTSD and oxidative stress, which included the investigation of the *ALOX12* and *ALOX15* [74], *BDNF* [75,76], and *RORA* [77,78]. In summary, studies based on individuals who were exposed to combat-related trauma demonstrated inconsistent findings, where one reported that veterans with PTSD have higher frequencies of the Met/Met and 66Met alleles as compared with veterans without PTSD [75], while another demonstrated that the frequencies of the two alelles were similar between the veteran groups [76]. While the findings from the two studies are contrasting, it is noteworthy that both studies strictly included trauma-exposed veterans without the inclusion of a healthy control group that has not been exposed to trauma, and the studies vary in sex distribution. Furthermore, Bruenig and colleagues (2016) have noted that the low Met/Met genotype frequencies in their sample may have influenced the results also [76].

Two other studies have explored the association between single nucleotide polymorphism (SNP) in the *RORA* gene with PTSD, where both studies found the same SNP within the *RORA* gene (rs8042149) to be significantly associated with the presence of PTSD [77,78]. Considering that the *RORA* gene is expressed in neurons and brain structures as to serve neuroprotective roles within the brain such as in response to oxidative stress [77], these findings indicate that *RORA* may be a marker for assessing one's capability in neuroprotection towards oxidative stress or inflammation, therefore a potential marker for an individual's risk or resilience towards developing PTSD. The two genes *ALOX12* and *ALOX15* were also assessed in one study [74], which demonstrated that the *ALOX12* locus (rs1042357 and rs10852889) significantly moderated the association between PTSD diagnosis and reduced cortical thickness in the brain, whereas the *ALOX15* locus did not moderate in this association.

A detailed description of the studies exploring PTSD- and oxidative stress-related genetic markers in relation to PTSD is provided in Tables 3 and 4.


**Table 3.** Studies exploring PTSD- and oxidative stress-related genetic markers in PTSD.

Abbreviations: BDNF, brain-derived neurotrophic factor; m, male; N, sample size; N/A, not applicable; PTSD, post-traumatic stress disorder; Ref., reference; SNP, single nucleotide polymorphism.

#### *3.4. Brain Alterations in Relation to Inflammation and Oxidative Stress in PTSD*

Another growing field of interest within the study of inflammation and oxidative stress in PTSD is the neuroinflammatory responses within the brain. Neuroinflammation refers to the physiological reactivity of the brain in response to various types of inflammation or oxidative stress, including but not limited to external injury or trauma [79,80], and is often observed in individuals with chronic diseases during their early stages of the disease as part of a coping mechanism [81]. Considering that investigating neuroinflammatory responses are often invasive in nature [79], the utilization of neuroimaging techniques to target inflammatory responses within the brain may potentially bridge the association between altered levels of peripheral biomarkers such as proinflammatory and anti-inflammatory cytokines and the clinical symptoms that arise in PTSD.

In the case of proinflammatory cytokine, previous literature have noted that IL-1β and IL-6 are secreted at stress-induced states and take part in the catecholaminergic pathways [82]. Considering that the catecholaminergic pathway has been demonstrated to play an important role in specific regions of the brain including the hippocampus and amygdala [82], it has then been suggested that these peripheral inflammatory markers may represent alteration in the function or activity of the hippocampus and amygdala, influencing symptoms including but not limited to emotion lability [83], fear reactivity, and retrieval of traumatic memories [82], all of which are crucial factors in PTSD. Previous studies have also shown that IL-1β is known to be consistently increased in major depression after psychosocial stress [84].

Over the years, emergence of novel technologies in neuroimaging have allowed the indirect correlative investigation of neuroinflammation without the invasive nature of directly measuring neuroinflammatory responses, as neuroimaging methods allow the direct measurement of the structure and functional state of the brain. In particular, neuroimaging techniques including MRI and PET have been largely utilized as they can obtain detailed information with regards to the structure of the brain such as cortical thickness, subcortical volume, or the metabolism of targeted brain regions as measured by glucose uptake. The current search included 3 neuroimaging-based studies on oxidative stress and inflammation in PTSD. The selected studies all reported findings that were consistent with previous literature, where markers of oxidative stress or inflammation are significantly associated with alterations of the brain.

The first study is by Miller and colleagues (2015), which explored the influence of the gene *ALOX12* on the association between PTSD and cortical thickness of the brain [74]. This study takes on an approach that is distinct from peripheral measures of proinflammatory cytokines, in that it investigated the *ALOX12* locus and its potential moderating roles, based on previous knowledge that the *ALOX12* enzyme plays a key role in an oxidative-related neuronal death program [85,86]. Findings

from this study first indicated that the *ALOX12* locus—through two single nucleotide polymorphisms rs1042357 and rs10852889—plays a significant moderating role in the association between PTSD and reduced cortical thickness of the brain. Specifically, brain areas with reduced cortical thickness included the middle frontal gyrus, superior frontal gyrus, rostral anterior cortex and medial orbitofrontal cortex, all of which were close in proximity and part of the frontal cortex of the brain.

The second study is by O'Donovan and colleagues (2015), which provided findings that are consistent to previous literature, where elevated proinflammatory cytokine levels of soluble receptor II for tumor necrosis factor (sTNF-RII) were significantly associated with PTSD [87]. The study also explored the influence of inflammation from a neurological perspective, and found that elevated levels of sTNF-RII is associated with reduced hippocampal subcortical volume in the case of veterans who have been exposed to combat and are diagnosed with PTSD, while this association was not present in veterans without PTSD [87]. Here, it is noteworthy that while the study also explored the potential roles of IL-6 by observing for alterations in serum IL-6 level, findings reported that IL-6 was not statistically associated with the reduced hippocampal volume in PTSD. In fact, findings indicated that higher post-traumatic stress symptom severity is associated with increased sTNF-RII and reduced IL-6 levels, which may contrast to previous literature as shown in Table 1, where the majority of individuals with PTSD were characterized with elevated serum levels of IL-6. This contrasting finding may be explained by the fact that, while IL-6 is often described as a proinflammatory cytokine marker of inflammation, it also does have anti-inflammatory properties [88,89]. Specifically, the primary roles of IL-6 are largely coupled with the regulation of the other two proinflammatory cytokines IL-1β and TNFα, and studies have reported that IL-6 often inhibits the expression of IL-1β and TNFα while promoting IL-10 [90]. This has been suggested to be especially so when in an anti-inflammatory environment, where the integrity of the blood-brain barrier needs to be regulated [91], such as during the occurrence of neuroinflammatory responses.

The last of the neuroimaging-based studies reviewed is a pilot study using 18F-fluorodeoxyglucose (FDG)-PET by Toczek and colleagues (2019), which investigated the association between the levels of proinflammatory cytokines and the amygdala of the brain as to explore the relationship between oxidative stress and inflammation measurable by peripheral markers and the alteration of the amygdala [92]. The study reported that there were no significant associations between levels of IL-1β, IL-6, and TNFα with FDG signal of the amygdala. The FDG signal has been used as a tool to detect vascular inflammation in specific regions [93] and FDG signal within a brain region may provide important contextual information such as the metabolism and glucose intake within the region [94,95], ultimately enabling the inference regarding the functional activity of the respective brain region. Although there had been no associations found between levels of IL-1β, IL-6, and TNFα and amygdala activity as measured by FDG signal, findings from the study by Toczek and colleagues (2019) reported a significant correlation among FDG signal in the amygdala, spleen and bone marrow [92]. The spleen and bone marrow are closely related organs that are also described as lymphoid organs [96,97] and have been previously described as key factors in the immune system as immune cells migrate across the spleen [97], while the bone marrow is the production site for lymphocytes [98]. As such, while proinflammatory cytokine levels did not alter in association with FDG signal of the amygdala in this study, correlation between FDG signals within lymphoid organs and the amygdala in individuals with PTSD may implicate inflammatory responses within the brain that results from trauma. Considering that FDG signal within the spleen and bone marrow have been suggested as potential approaches to measuring systemic inflammation in other chronic metabolic diseases [99], this finding may reflect a link between amygdala activity and systemic inflammation in PTSD. The findings by Toczek et al. (2019) are also consistent with a portion of previous literature, which provided evidence for brain alterations in specific regions including the amygdala, hippocampus, and prefrontal cortex in patients diagnosed with PTSD [100,101]. In particular, previous literature suggested that the amygdala, hippocampus, and prefrontal cortex each has a high density of glucocorticoid receptors, which are then related with the activation of the hypothalamic-pituitary-adrenal (HPA) axis [102]. Considering that the HPA axis

is a significant contributor in responsiveness towards psychosocial stress [103] and various forms of trauma exposure [104], studies that bridge the relationship among alteration in inflammatory markers from a genetic or peripheral level, structural and functional alterations of brain regions associated with PTSD, and the clinical symptoms of PTSD, may provide a more comprehensive overview of the pathophysiology of PTSD as a non-communicable disease.

A detailed description of the neuroimaging-based studies reviewed that investigated the relationship between inflammatory and oxidative stress markers and alteration within the brain in PTSD is provided in Table 4.


**Table 4.** Neuroimaging-based studies exploring oxidative stress and inflammation in PTSD.

Abbreviations: CT, computed tomography; f, female; FDG, 18F-fluorodeoxyglucose; IL, interleukin; m, male; N, sample size; PET, positron emission tomography; PTSD, post-traumatic stress disorder; Ref, reference; sMRI, structural magnetic resonance imaging; SNP, single nucleotide polymosphism; sTNF-RII, soluble receptor II for tumor necrosis factor; TNF, tumor necrosis factor.
