Several studies have evaluated levels of cortisol in children exposed to trauma (Johnson et al. 2013; Shonkoff et al. 2012; McGowan et al. 2009; Carrion et al. 2002; McCrory et al. 2010). In a study of maltreated children with PTSD, 24 h levels of cortisol were increased (De Bellis et al. 2010). Goenjian et al. (1996) reported lower levels of basal cortisol in children 5 years after an earthquake. Gunnar et al. (2001) reported higher diurnal levels of cortisol after 6.5 years post trauma. In a study of girls who had attempted suicide. Girls with early sexual abuse had lower salivary levels of morning cortisol were reported by King et al. (2001). Carrion et al. reported higher levels of cortisol in 33 youth ages 10–16 and, and combined with changes in brain MRI. Youth with post-traumatic stress symptoms (PTSS) (n = 45) had decreased total brain tissue in comparison with healthy controls (n = 15). There was a significant negative association between pre-bedtime cortisol levels and left ventral prefrontal cortex gray volumes (Carrion et al. 2010a). Studies in adults report similar findings in patients with PTSD (Vythilingam et al. 2002). Some developmental differences have been reported in levels of IL-6 in children with PTSD. In both adults and children, IL-6 is elevated in individuals with PTSD. In adults the levels of IL-6 continue to be elevated chronically, while in children the IL-6 levels tend to normalize over time (Pervanidou et al. 2007).

The pathophysiological mechanisms underlying the development of neurobiological consequences of childhood trauma have not been fully elucidated. Disruption of the HPA axis with subsequent cortisol induced toxicity has been proposed as one mechanism for neuroimaging differences (Bremner 1999, 2006; McCrory et al. 2010; Marin et al. 2007). Another proposed mechanism is the induction of an inflammatory response leading to neuronal injury (Miller et al. 2009c; Herberth et al. 2008). A more recent and novel hypothesis is that brain injury could be caused by disruption of the BBB leading to influx of inflammatory cells, as well as toxic compounds which could lead to damage to several brain areas (Neuwelt et al. 2011; Falcone et al. 2015; Andrews and Neises 2012; Angelov et al. 2009; Cucullo et al. 2011; Esposito et al. 2001, 2002, 2003; Janigro 2012; Marchi et al. 2004; Rapoport et al. 1972; Reiber 2001). The last hypothesis has not been investigated as thoroughly due to the of lack of easily available tools to measure the integrity of the BBB until recently (Cucullo et al. 2011; Marchi et al. 2003, 2004; Kapural et al. 2002). Studies in soldiers before and after a parachute jump describe how chronic stress can activate peripheral blood mononuclear cells (PBMC) and increase several inflammatory markers such as nerve growth factor, cortisol, and adrenocorticotropic hormone (Aloe et al. 1994).

Animal model studies suggest that activation of PBMCs can potentially exacerbate CNS damage via inflammation (Ajmo et al. 2008). Lewitus et al. (2008) described the role of T lymphocyte infiltration at the level of the BBB and an increase of intercellular adhesion molecules (ICAM-1) on the BBB. ICAM-1 has also been associated with increased migration of inflammatory cells to the BBB. de Pablos et al. (2006) described increased levels of inflammatory markers in animals secondary to stress that were associated with a decreased number of neurons. A leaky BBB can permit further accumulation of PBMCs, that potentially will produce more inflammation and production of matrix metalloproteinases, further increasing the BBB permeability (Rosenberg 2009).

Studies in animal models have demonstrated that stressful situations, such as immobilization, forced swimming, and training in a water maze produced BBB breakdown (Sharma 2011; Sharma et al. 1990, 1991, 1992, 1995, 2007a, b, 2009, 2010, 2012; Sharma and Dey 1981, 1984, 1986, 1988; Sharma and Johanson 2007). Sharma et al. concluded after multiple investigations on the role of the BBB under stress that specific opening of the BBB occurs under different kinds of stressful situations: “emotional stress,” “heat stress,” “cold stress,” “swimming stress,” “immobilization stress,” and “surgical stress.” Adaptation or resilience is what makes a difference between how these stressors continue to impact the brain overtime. After the initial stress the glucocorticoid system is activated, increasing levels of ACTH that help the organism cope with the increased demands of stress (Sharma 2011; Sharma et al. 1990, 1991, 1992, 1995, 2007b, 2009, 2010, 2012, 2013; Sharma and Dey 1981, 1984, 1986, 1988; Sharma and Johanson 2007; Sharma and Cervos-Navarro 1991). Catecholamines (adrenaline, noradrenaline) are released and alter the BBB function. The initial mild to moderate stress can be an adaptive response that helps the organism to cope with the increasing demands while prolonged, chronic, severe stress can lead to brain changes in selective areas of the brain and potentially later on contributing to the development of PTSD (Shalev et al. 2009) (Fig. 9.2).

There is very strong evidence for the role of inflammation in patients with autism. In children with autism spectrum disorder (ASD), 40 studies reported data on PIM. Al-Ayadhi and Mostafa (2012) in a study of 45 patients with ASD reported increases in levels of IL-17A in a group of ASD patients compared to controls. Manzardo et al. (2012) described 99 ASD patients vs. 40 controls and found decreased levels of IL-1alpha and IL-6 in ASD compared to controls. In a very small sample of ten patients, Bent et al. reported no changes in several cytokines but no control group was available. Also, there was no data on the gender of the participants (Bent et al. 2012). In a study in Saudi Arabia comparing 20 male patients with ASD with 19 controls, levels of IFN gamma were significantly elevated in patients with ASD (El-Ansary et al. 2011). Jyonuchi et al. identified elevated levels of IL-1, IL-6, IL12p40, IL-17, IL-23, IFN gamma, and TNF alpha after stimulation of PBMC. In study of 47 patients with ASD vs. 37 controls, there was decreased production of IL-1, IL-6, IL12p40, IL17, IL23, IFN gamma, and TNF alpha in ASD patients vs. controls (Jyonouchi et al. 2012). In a Brazilian study of 24 patients with ASD vs. 24 controls, levels of IFN gamma were increased in patients with ASD (Tostes et al. 2012). Ashwood et al. reported increased levels of IFN gamma and TNF alpha in a group of 66 patients with ASD vs. 77 controls after stimulation of PBMC. No changes were observed in IL-1B, IL-2, IL-4, IL-5, IL-6, IL8, IL-10, and IL-13 (Ashwood et al. 2011a). One study included 97 patients with ASD, 87 controls, and 39 developmentally delayed patients after plasma multiplex luminex, analysis of levels of IL-1B, IL6, IL-8, IL-12p40 were increased in patients with ASD (Ashwood et al. 2011b). In Saudi Arabia 25 males ages 4–12 vs. 16 controls, there were decreased levels of IL-6 and TNF alpha in ASD. In a study of 6 patients with ASD vs. 6 controls, levels of TNF alpha and IL-6 were increased in the ASD group (El-Ansary et al. 2011). Jyonouchi et al. (2011) reported decreased production of IL-6, IL beta, IL 12, IL 23, IL 10, sTNFRII, and TGF B in 30 ASD patients compared to 28 controls after stimulation of PBMC. Suzuki et al. (2011) in Japan reported levels of 48 plasma inflammatory markers increased in a group of 28 ASD patients vs. 28 controls. While Tobiasova in a study of 21 ASD patients vs. 15 controls, reported serum of IL-13 decreased in ASD patients. ASD patients (n = 77) were treated with risperidone and no effect was observed in the inflammatory markers vs. placebo in ASD patients (Tobiasova et al. 2011). In a postmortem study evaluating expression of IL-6, in the frozen cerebellum of 6 patients of ASD vs 6 controls, increased expression of IL-6 in the cerebellum in ASD was identified in patients compared to controls (Wei et al. 2011). Patients with ASD and fragile X in ASD with fragile X (n = 40) vs. patients with fragile X and no ASD (n = 64) vs. 19 controls, levels of IL1 alpha and IL12p40 were reported increased in patients with ASD/fragile X patients vs. controls (Ashwood et al. 2010). In a group of 17 patients with ASD vs. 16 controls, levels of IL-1 B, IL-6, TNF alpha, were increased after stimulation of the Peripheral Blood Monocytes (PBMC) (Enstrom et al. 2010). In a study of 19 patients with ASD, levels of IL-6 were decreased if their PBMC were treated with BDE-47. Levels of IL-8 were increased in the ASD patients (Ashwood et al. 2009). IL-6, TNF alpha, and IFN gamma were increased in the frontal cortex of ASD patients postmortem compared to controls (Li et al. 2009). Onore et al. (2009) reported decreased IL-23 in 34 patients with ASD. Multiple infections and food allergies are frequently reported in children with ASD. Jyonouchi et al. (2008) studied levels of inflammatory markers in this group and found decreased IL-1 B and IL10 in ASD and increased IL-23 in ASD. Chez et al. evaluated levels of CSF TNF alpha in ASD and found levels increased in 10 patients. No controls were recruited in this study (Chez et al. 2007).

Hyperbaric oxygen treatment is an alternative treatment (non-FDA approved) for ASD. In a group of 18 patients with ASD without controls, levels of high sensitive CRP were elevated in three patients with ASD; after treatment with hyperbaric oxygen, the levels of CRP normalized (Rossignol et al. 2007). Molloy et al. (2006) reported elevated levels of IL-4, IL-5, and IL-13 in 20 patients with ASD vs. 20 controls. Jyonouchi et al. (2005a) analyzed the impact of gastro intestinal (GI) symptoms in children with ASD and the impact on inflammatory markers compared to controls and found increases in TNF alpha and IL-12 in ASD, but without impact on the GI symptoms. The same group in a dietary intervention study evaluated the levels of TNF alpha after stimulation of PBMC, patients with ASD had increased TNF alpha. The restricted diet had no impact on other inflammatory markers (Jyonouchi et al. 2005b). Two studies evaluated intestinal biopsy samples in children with ASD, De Felice found no difference in inflammatory markers in patients vs. controls, and Ashwood found increased CD3 TNF, CD3 IL2, CD3 IFN gamma, and CD3 IL 4 in patients with ASD (Ashwood et al. 2004). Zimmerman et al. (2005) evaluated levels of CSF and serum of several inflammatory markers in patients with ASD and found decreased levels of IL1Ra and IL 6 in patients vs. controls. In a study evaluating brain tissue from area-Brodman 19 and cerebellar cortex, in 9 males with ASD vs. 9 male controls, the inflammatory pathway gene dysregulation was significant in patients with ASD (Ginsberg et al. 2012).

Looking at these studies, there is consistent evidence for increased TNF alpha in patients with ASD. Contradictory findings in patients with ASD on the levels of IL-6. One of the important anti-inflammatory cytokines was decreased in nine studies of children with ASD (Mitchell and Goldstein 2014).

Different animal models of maternal inflammation during the first trimester have been proposed in the pathophysiology of autism (Smith et al. 2007; Patterson 2002). The role of infections during pregnancy has been extensively discussed, especially from the perspective on how early inflammatory changes can potentially alter neural circuitry and pruning. Maternal immune activation can later cause cerebellar abnormalities in the offspring of mice. Similar characteristics are shared in the immunological animal models of schizophrenia and autism (Smith et al. 2007; Zuckerman and Weiner 2005; Meyer et al. 2005, 2006a, b, 2008a, b; Romero et al. 2007; Dammann and Leviton 1997; Zuckerman et al. 2003; Depino 2006; Gilmore et al. 2004; Nawa and Takei 2006).

Several studies report increased inflammatory markers in patients with anxiety disorders. They reported an important relationship between the development of de-novo obsessive symptoms or tic disorder after the exposure to upper respiratory infection (Mitchell and Goldstein 2014).

Since not all suicidal behavior is associated with mood disorder, this section highlights suicide as a separate topic. Suicide rates are high among young people; it is the third leading cause of death in 15- to 24-year-olds and accounts for 13 % of all adolescent deaths annually (Falcone 2008; Nock et al. 2013). Suicidal ideation and attempts are very frequent in adolescents. A survey of adolescents showed that 19 % had seriously considered suicide in the last year, 15 % had made a plan and 8 % reported that they had made an attempt. Since 1950, the suicide rate of adolescents has quadrupled and represents 12 % of the mortality rate in this age group (Brent 2011). Multiple studies evaluating peripheral biological makers have been reported in suicidal patients and postmortem brains of those who have committed suicide (Birmaher et al. 1996). Different systems have been reported to show abnormalities in suicidal patients: the serotoninergic system, the HPA axis, neurotrophin factors, and the neuroimmune system (Serafini et al. 2013a). In the quest for biological markers for different mental illnesses, goals morphed based on genetic studies. If similar genes are involved in different psychiatric disorders, similar specific symptoms may have in common some specific biological pathways. Also independent environmental factors such as stress and adverse childhood experiences can alter the neuroimmune network in youth (Shonkoff et al. 2012). These alterations can predispose to suicide later in life (Dube et al. 2001, 2009; Danese et al. 2007). Special traits such as impulsivity and aggression might be more frequent in adolescent suicides (Nock et al. 2013; Nock 2009).

Although the exact pathophysiology of suicidal behavior has not been unraveled, several studies point to increasing evidence that there is an alteration in the immune system. There are at least 21 published studies of peripheral inflammatory makers in patients after a suicide attempt to date (Serafini et al. 2013a). Although the role of cytokines in suicidal patients has been investigated since the early 1980s and remains ambiguous, the results look more compelling (Johnson et al. 2002; Falcone et al. 2010a, b; Gabbay et al. 2009b; Bayard-Burfield et al. 1996; Carlborg et al. 2013; Dwivedi 2010; Dwivedi et al. 2003; Erhardt et al. 2013; Isung et al. 2012a; Janelidze et al. 2011; Kim et al. 2007; Lee and Kim 2010; Lindqvist et al. 2008, 2011; Serafini et al. 2013b; Steiner et al. 2008; Sublette et al. 2011). Melancholic patients have been consistently reported as having low monocyte counts and cytokine abnormalities (Rothermundt et al. 2001a). Due to the crucial role in the acute inflammatory reaction the monocytes–macrophage system has received the most attention (Serafini et al. 2013a). Cytokines produced by the monocytes like, IL-1beta, activate the complement system.

In adolescents, levels of several inflammatory markers have been identified in patients with suicidal ideation and behavior. Studies have reported alterations of IL-1B, IL-6, TNF-alpha, IFN-gamma, VEGF, S100B, orexin, BDNF, and Kyneurine (Falcone et al. 2010a; Gabbay et al. 2009b; Serafini et al. 2013a; Janelidze et al. 2011).

There are important advances in the field of psychoneuroimmunology in the last 25 years; however, most of the studies on inflammatory markers and psychiatric disorders come from adult populations (Zorrilla et al. 2001; Drexhage et al. 2010; Dowlati et al. 2010; Padmos et al. 2008; Shalev et al. 2009; Serafini et al. 2013a). In this chapter we focus on studies in children and adolescents (Carlton et al. 2009; Falcone et al. 2008a, 2010a, c, 2013; Gabbay et al. 2009b; Ginsberg et al. 2012; Janelidze et al. 2011; Falcone and Janigro 2008). Pediatric populations are key in our process to understand the development and progression of mental illness and in the case of finding novel treatments for those patients who have comorbid increase of inflammatory markers and different psychiatric disorders. The evidence in adult studies points to a developmental advantage when the treatment is started early (Akhondzadeh et al. 2007; Bresee et al. 2006; Muller et al. 2002, 2005; Rapaport et al. 2005; Riedel et al. 2005). The study of inflammatory markers in children is a complex field, important confounding factors have to be taken into account, such as age, Tanner stage, weight, use of medications, sleep, variability of the marker during the day, life stressors, trauma history, and family history of psychiatric disorder (Mitchell and Goldstein 2014; Lucas et al. 2013; Bahn and Schwarz 2011; Gottschalk et al. 2013; Lopresti et al. 2014). Most studies in pediatric populations are cross-sectional and retrospective. This limits our ability to identify pathogenic cause for mental illness, and only permits discussion of an association between inflammation and diverse psychiatric disorders reviewed. Despite this limitation, there is important preliminary evidence for the role of inflammation in psychiatric disorders in pediatric populations (Pardo et al. 2005; Falcone et al. 2010a, 2013; Mitchell and Goldstein 2014; Gabbay et al. 2009b, c; Padmos et al. 2009; Carrion et al. 2002, 2007, 2008, 2010a, b; Chez et al. 2007; Oades 2011; Oades et al. 2010a, b; Pandey and Dwivedi 2012; Carrion and Wong 2012; De Bellis and Kuchibhatla 2006). Taking into account the high incidence of comorbidities in child psychiatry, it is very unlikely that inflammatory markers can be used solely for diagnostic purposes. However they could be very helpful to evaluate treatment response, disease progression and help us monitor high-risk patients that may potentially benefit from preventive treatment (Mitchell and Goldstein 2014; Goldstein and Young 2013). There is progress in identifying the onset, course, and recurrence of behavioral disorders in children and adolescents; however, further research is needed to understand the role of inflammation and the relationship with psychosocial and psychobiological factors.