A Specific Inflammatory Profile Underlying Suicide Risk? Systematic Review of the Main Literature Findings

1Suicidal behavior, which is frequently linked to mood disorders, in particular major depression, is a fundamental public health concern associated with significant disability and psychosocial impairment worldwide. Major depression may be considered one of the leading causes of suicide worldwide with a more than 20-fold increased risk and at least half of all completed suicides which are linked to depressive disorders. Generally, suicidal behavior associated with major depression occurs predominantly in the elderly due to the emergence of psychotic symptoms.

2Abnormalities of the immune system have been reported to significantly contribute to the pathophysiology of both major depression and suicidal behavior based on existing evidence. Inflammatory mediators and oxidative stress leading to excitotoxicity may play a critical role in the pathophysiology of either major depression and suicidal behavior.

3Studies in the current literature supported the abnormal activation of cell-mediated immunity and different immune biomarkers in the pathophysiological pathways underlying major affective disorders and suicidal behavior. Specifically, abnormally elevated granulocytes and monocytes numbers, enhanced C-reactive protein (CRP) and haptoglobin, chemokines, abnormalities of regulatory T-cells, and inflammatory cytokines have been documented in both patients with major depression and suicidal behaviors. Thus, central and peripheral immune dysregulation are proposed as important pathways underpinning the pathophysiology of major depression and suicidal behavior identifying peripheral inflammatory mediators as promising candidate immune biomarkers.

4In the last few decades, a growing interest emerged for inflammatory cytokines, which are known as key modulators of neuroinflammation, representing a heterogeneous group of molecules released by immunocompetent cells, such as lymphocytes and macrophages. Inflammatory cytokines include a heterogeneous group of messenger molecules released by immunocompetent cells playing a key role in stress-related conditions and major depression. One of the most relevant characteristics of inflammatory cytokines is the pleiotropy through which cytokines may bind to different cell target types. These molecules are generally divided into two subgroups: pro-inflammatory, including interleukin (IL)-1β, IL-6, interferon (IFN)-γ, and tumor necrosis factor (TNF)-α, and anti-inflammatory cytokines (e.g., IL-10, transforming growth factor-beta (TGF-β)).

5Several studies recently postulated the existence of abnormally elevated levels of inflammatory cytokines in subjects at risk for suicide or those who die by suicide.

6For instance, Lindqvist and colleagues reported increased levels of IL-6 in the cerebrospinal fluid of individuals who attempted suicide when compared to controls. Janelidze et al. found abnormally elevated IL-6 and TNF-α concentrations together with reduced IL-2 levels in suicide attempters suggesting that cytokine levels in the blood may distinguish suicide attempters from depressed patients. Moreover, suicide attempters reported higher peripheral levels of both inflammatory cytokines and CRP, that is a pentameric protein in blood plasma, the circulating levels of which are higher in response to inflammation than depressed individuals who did not attempt suicide, independently of the time interval between suicidal behavior and cytokine quantification. Relevantly, existing meta-analyses documented that circulating levels of IL-1β and IL-6 are abnormally elevated and IL-2 concentrations are significantly reduced in suicidal patients relative to non-suicidal patients and healthy controls promoting the hypothesis of inflammatory markers in suicidal patients (including those with active suicidal ideation as well as those with a positive history of suicide attempts). These studies supported the assumption that determining the inflammatory marker profile of suicidal patients may be crucial in order to better predict suicidal behaviors and identify novel therapeutic strategies.

7In their postmortem study, Shelton et al. investigated the expression of inflammatory genes in the Brodmann area of 10 depressed patients and found that genes coding for IL-1α and IL-2 were upregulated in the prefrontal cortex of depressed subjects. Furthermore, Pandey et al. found that both the protein and messenger ribonucleic acid (mRNA) levels of IL-1β, IL-6, and TNF-α were abnormally altered in the prefrontal cortex of young adults who died by suicide, when compared to controls. Therefore, the available evidence supports the notion of existing abnormalities in pro-inflammatory cytokines in individuals who die by suicide as well as suicide attempters.

8Environmental, interpersonal stressors (e.g., social and familial threats, negative life events, etc.) may be triggering conditions underlying most suicidal acts. These precipitating/contributing factors may significantly interact with predisposing factors increasing vulnerability to suicide according to a stress-diathesis perspective, although the main pathophysiological mechanisms underlying this link are still poorly understood.

9Such stressful conditions can activate inflammatory signaling pathways (e.g., nuclear factor kappa-lightchain-enhancer of activated B cell DeoxyriboNucleic Acid (NF-kB DNA) binding) in specific immune cells such as monocytes and macrophages linked to abnormally enhanced pro-inflammatory IL-1β, IL-6, and TNF-α levels on neuroglial cells. Classically, neuroglial cells include astrocytes, oligodendrocytes, microglia cells and NG2 glia. In response to inflammatory and injury stimuli, microglia may develop different phenotypes and, when abnormally activated, microglia usually releases pro-inflammatory cytokines or chemokines such as IL-1, IL-6, IL-12, and TNF-α. Pro-inflammatory cytokines released by microglia may stimulate astrocytes having phagocytic properties and secreting abnormally elevated cytokines such as IL-6 and TNF-α. Finally oligodendroglia, commonly derived by oligodendrocytes progenitor cells, may release specific cytokines (e.g., IL-1) and chemokines (e.g., monocytes chemoattractant protein 1) playing a fundamental role in myelination. Importantly, abnormally elevated inflammatory cytokines levels are able to determine a cytotoxic action on neuroglial cells, presumably leading to apoptosis and demyelination with the final result of increased glutamate release impairments in neuroplasticity mechanisms and enhanced excitotoxicity. Given the mentioned background, the present manuscript aimed to systematically review the current literature about the role of pro-inflammatory cytokines in suicidal behavior.

10Relevant studies about the main topic have been identified using a detailed search strategy reported in Figure 1. Aiming to provide a systematic and critical review on inflammatory cytokines and their role in suicidal behavior, a detailed Pubmed/Medline, Scopus, Science Direct, and PsycInfo search has been conducted. All relevant studies within the time interval between 1990 and June 2019 (exclusively in English language) upon this topic have been searched. The search used a combination of the following terms: “Neuroinflammation” AND “Inflammatory Cytokines” OR “pro-inflammatory cytokines” OR “anti-inflammatory cytokines” AND “Suicid*” (including suicide ideation OR suicidal thoughts OR deliberate self-harm OR suicide attempts OR completed suicide). We examined all the full-text articles on the main topic in case a title or abstract seemed to describe a study eligible for inclusion. Two independent researchers (VMP, GS) carried out a two-step literature search. Consultations with senior authors (AF, MP, MA) have been used in case of any discrepancies between the two reviewers who, blind to each other, examined the studies for their possible inclusion. Relevant studies were also searched using the reference lists of the articles included in the review (VMP, GS). We included all English language full-text articles reporting original data about the main topic. The main search strategy and criteria for study selection (identification, screening, eligibility, inclusion process) in the current review are reported in Figure 1.

11In order to obtain a high standard of reporting, we adopted the ‘Preferred Reporting Items for Systematic Reviews and Meta-Analyses’ (PRISMA) guidelines. PRISMA may be considered an evidence-based minimum set of items that is frequently used in both systematic reviews and meta-analyses. PRISMA usually allows the reporting of reviews assessing randomized clinical trials, but it may be similarly adopted for reporting systematic reviews of other research, such as reports referring to evaluations of interventions. The PRISMA statement includes a 27-item checklist and a four-phase flow diagram for reporting in systematic reviews and meta-analyses. PRISMA is usually adopted to include a reliable report of different types of health studies and it mainly aims to improve the quality of research used in decision-making in healthcare.

12Inclusion criteria were papers reporting a link between inflammatory cytokines and suicidal behavior in both depressed and nondepressed individuals. Whether a title/abstract appears to refer to a study eligible for inclusion, the full-text article was down-loaded and carefully assessed to ascertain its possible relevance to the main topic according to inclusion/exclusion criteria. The following criteria have been used for the aims of this study: (a) being an original paper in a peer-reviewed journal and (b) describing the experimental association between (pro- or anti-) inflammatory cytokines and suicidal behavior. Exclusion criteria were: manuscripts referring exclusively to suicide attempts/suicidal ideation, articles with samples including individuals <18 years of age, articles that did not mention how the diagnosis has been performed, meta-analyses, main texts without abstracts, manuscripts that focused only on completed suicides, and those reporting only incomplete laboratory data.

13The recorded variables for each article about inflammatory cytokines and suicide were: author(s), year, sample features, study design, type of inflammatory cytokines which have been investigated, main results, major shortcomings, and conclusive remarks (Table 1 and Table 2).

14Although we did not include a risk of bias assessment, according to the PRISMA statement the following criteria have been considered in order to detect the fundamental limitations potentially affecting the validity and generalization of the included studies: (a) representativeness of the main study sample; (b) existence/representativeness of a control group; (c) study design; (d) existence of a long-term follow-up period; (e) evidence-based measures to assess suicide risk; and (f) existence of at least two raters who blind to each other identified the most relevant studies for inclusion.

15Initially, a total of 312 full-text articles (149 from Pubmed, 46 from Scopus and 117 from ScienceDirect, and PsycInfo) emerged using the combined search strategy; after a thoroughly analysis, 60 full-text articles have been screened and 191 completely excluded after removing duplicates. Later, 61 additional full-text articles were further deleted as they were: (1) articles not published in peer reviewed journals, (2) articles not in English language, and (3) articles without abstracts. In addition, nine further articles were subsequently excluded as: (1) their abstracts did not specify in detail the association between inflammatory cytokines and suicidal behavior, (2) they were published before 1990, (3) include unclear information about materials and methods or number of recruited patients. Finally, of the fifty-one articles which were evaluated for eligibility, 22 additional full-texts were excluded due to the low-relevance to the main topic. Therefore, the final result consists of 29 studies (including 1525 patients and 1400 controls) that fulfilled inclusion criteria.

16Overall, 11 studies analyzed the existence of inflammatory cytokines abnormalities in suicidal vs. nonsuicidal healthy controls. Recently, Conejero et al. reported that IL-1β was negatively associated with right orbitofrontal cortex (OFC) activation during the explicit social exclusion vs. social inclusion, whereas IL-2 was positively associated with activation of the right anterior cingulate cortex (ACC), insula and OFC even after controlling independently for the suicidal status. Moreover, Pandey et al. found that IL-1β, IL-6, TNF-α, and lymphotoxin A mRNA and protein levels were significantly increased, while IL-10 and IL-1 receptor antagonists (IL-1ra) were reduced in the prefrontal cortex (PFC) of depressed individuals who died by suicide compared to healthy controls. The same research group previously documented that both protein and gene expression of glucocorticoid receptor-α (GR-α) were reduced in the PFC and amygdala of teenage suicide victims (n = 24) compared to healthy controls.

17The GR inducible target gene glucocorticoid-induced leucine zipper (GILZ) mRNA levels were decreased in PFC and amygdaloidal nuclei of these subjects as well. Based on another recent study, TNF-α expression was significantly higher in the dorsolateral prefrontal cortex (dlPFC) of suicide subjects regardless of psychiatric diagnosis, although its expression was enhanced even in major depressive disorder (MDD) subjects who died by causes other than suicide. Moreover, in a two-year follow-up study, Bay-Richter et al. demonstrated that quinolinic acid (QA) was increased and kynurenic acid (KYN) decreased in suicidal patients vs. healthy controls. A significant association between lower KYN and severe depressive symptoms and IL-6 levels and more severe suicidal symptoms were also documented. Another additional study analyzed the levels of specific inflammatory markers in suicidal (n = 47), non-suicidal depressed patients (n = 17) and healthy controls (n = 16) and found greater IL-6 and TNF-α levels coupled with decreased IL-2 concentrations in the group of suicide attempters after adjusting for potential confounders.

18Moreover, Sublette et al. found that KYN was higher in the MDD suicide attempter subgroup compared with MDD non-attempters who did not differ from healthy controls. Only suicide attempters showed a positive correlation of the cytokine activation marker neopterin with the KYN: tryptophan (TRP) ratio, concluding that the production of KYN may be influenced by inflammatory processes. Cerebrospnal fluid (CSF) levels of IL-6 were also found to be higher in suicide attempters than healthy controls, with subjects having a history of violent suicide attempts showing the highest levels of IL-6. A significant positive correlation between Montgomery–Åsberg Depression Rating Scale (MADRS) scores and CSF IL-6 levels were reported. Importantly, CSF IL-6 and TNF-α correlated with 5-hydroxyindolacetic acid (5-HIAA) and homovanillic acid (HVA). Significantly reduced plasmatic TNF-α concentrations compared to non-suicidal MDD adolescents (n = 18) were also reported by Gabbay et al. (2009) in a sample of 12 suicidal MDD adolescents. Even after controlling for age and gender, IFN-γ was increased in both suicidal MDD and non-suicidal adolescents compared to controls. Additionally, in the study of Kim et al., non-suicidal MDD patients (n = 33) had higher IL-6 production than suicidal MDD patients (n = 36) and healthy controls (n = 40). Suicidal MDD patients had lower IL-2 compared to non-suicidal patients and normal controls as well, while both MDD patients with and without a history of suicide attempts had lower levels of IFN-γ and IL-4 and a higher TGF-β1 production. Finally, Lee and Kim found that in vitro TGF-β1 levels resulted significantly higher in suicidal MDD patients (n = 48) and non-suicidal MDD patients (n = 47) than controls (n = 91).

19Four studies analyzed the existence of inflammatory cytokines abnormalities in suicidal patients (having both a positive history of suicide attempts and suicidal ideation) and eventually healthy controls. First, Knowles et al. tried to identify a possible link between increased inflammatory cytokines and suicide risk and reported that IL-8 and IL-6 shared a significant genetic overlap with risk of suicide attempts (for IL-6 this risk was attenuated when body mass index (BMI) was included as a covariate). Relevantly, the genetic overlap between IL-8 and risk for suicide attempts was significant in females but not males. According to Keaton et al., the biological profile of patients considered to be at increased risk of suicide differed from that of depressed individuals. Importantly, blood cell count and polymorphonuclear leukocyte count had a significant impact on suicide risk. The authors reported that IL-8 was negatively associated with increased suicide risk after adjusting for confounders. Furthermore, Melhem et al. found lower hair cortisol concentrations in first-time suicide attempters compared to patients with suicidal ideation and normal controls.

20Moreover, patients with suicide attempts showed lower GR or the DNA methylation of human glucocorticoid receptor gene (NR3C1) (α isoform) mRNA, higher CRP, and higher TNF-α mRNA supposing that suicide attempters demonstrated a distinct biological profile on the identified biomarkers. Finally, it has been already suggested that T-cells of depressed suicidal depressed patients may have T-helper1 (Th1) characteristics, while T-cells of non-suicidal depressed patients may have Th2 characteristics based on the concept that Th1 activation in suicidal depression may reflect, similarly to other autoimmune disorders (e.g., autoimmune diabetes, multiple sclerosis, and autoimmune thyroiditis), a specific form of self destructive activation of the immune system.

21Two studies explored the existence of inflammatory cytokines abnormalities in depressed patients with suicidal ideation vs. those without. First, O’Donovan et al. reported that MDD patients with suicidal ideation had higher inflammatory index scores than both controls and MDD patients with lower suicidal ideation. MDD patients also had higher levels of IL-6 and IL-10 than normal controls at the follow-up analyses. A trend toward higher levels of CRP in MDD patients was reported as well. Additionally, patients with suicidal ideation had higher levels of IL-6 and a trend toward lower levels of IL-10 than healthy controls. In addition, thirty female outpatients with recurrent MDD (18 with suicidal ideation and 12 without) together with 16 healthy controls were also analyzed in the study of Grassi-Oliveira et al.. After multivariable analysis of covariance adjusted for age, BMI, and depression severity, the authors documented that MDD patients with suicidal ideation presented lower levels of monocyte chemoattractant protein-1/chemokine C-C motif ligand 2 (MCP-1/CCL2) and normal T-cell expressed and secreted (RANTES/C-C motif ligand 5 CCL5 or RANTES/CCL5) and higher levels of Eotaxin/C-C motif chemokine 11 (CCL11) when compared to healthy controls.

22Seven studies investigated the existence of inflammatory cytokines abnormalities in patients with a history of suicide attempts vs. those without. Coryell et al., in a sample of MDD patients with a history of suicide attempts (n = 79), reported that IL-1β levels were lower when compared with those of subjects without a history of suicide attempts (n = 123). Moreover, IL-1β levels correlated inversely with aggression measures. Kim et al. (2013) reported that the GG genotype of the TNF-α-308G>A polymorphism increased the risk for suicide attempts. In addition, IFN-γ +874A>T and IL-10 -1082A>G were not associated with suicide risk. In the longitudinal study of Janelidze et al., CSF eotaxin-1, macrophage inflammatory protein-1β (MIP-1β), monocyte chemoattractant protein-1 (MCP-1), monocyte chemoattractant protein-4 (MCP-4), and thymus and activation-regulated chemokine (TARC) were lower in the group of suicide attempters than healthy controls. Moreover, lower chemokine levels were associated with psychotic symptoms and pain. After follow-up analyses, TARC was significantly lower in suicide attempters compared to psychiatric patients who had never attempted suicide. A positive correlation between blood TARC and brain-derived neurotrophic factor (BDNF) levels emerged. In 2013, Vargas et al. reported that subjects with a history of suicide attempts (n = 141) had significantly higher levels of nitrogen oxide (Nox) and lipid hydroperoxides and lowered plasma total antioxidant potential (TRAP) when compared to individuals without suicide attempts (n = 201). After logistic regression analysis, both unipolar and bipolar disorder, female gender, smoking behavior, and lipid hydroperoxides were associated with a history of suicide attempts independently of specific socio-demographic and clinical risk factors.

23Isung et al. aimed to identify new inflammatory biomarkers for suicide prediction and found lower levels of vascular endothelial growth factor (VEGF) in the seven patients who completed suicide after a follow-up period of 13 years. VEGF also showed a trend for negative correlation with the planning subscale of the Suicide Intent Scale. A trend emerged for lower IL-2 and higher IFN-γ levels in suicide victims as well. The same research group found the existence of a significant negative correlation between CSF VEGF and depression severity in a cross-sectional study, including medication-free suicide attempters and 20 healthy male volunteers. The authors hypothesized that lower CSF levels of VEGF may reflect a lack of trophic support to neurons and downregulation of neurogenesis in the hippocampus associated with more severe depressive states. Finally, in 1993 Nassberger and Träskman-Bendz had already identified an association between S-IL-2R and the ratio of norepinephrine-epinephrine in 24-h urine together with plasma and cerebrospinal fluid 4-hydroxy-3-methoxymethylglycol in suicide attempters.

24Three studies explored cytokines abnormalities in patients who died by suicide vs. those who died for other causes. Boehm et al. analyzed the lungs of burn victims and showed a greater extent of intra-alveolar edema than the other groups. The authors added that macrophages in all groups mostly showed a distinct expression of TNF-α, but not of IL-8 or intercellular adhesion molecule-1 (ICAM-1), that the intravascular erythrocytes positivity of TNF-α was strongest in the group of burn victims. Tonelli et al. reported an elevated expression of IL-4 in female suicide victims and IL-13 in male suicide victims. Abnormally elevated, although not significant, cytokine expression was also observed for TNF-α in female suicide victims. Finally, Torres-Platas et al. reported that blood vessels surrounded by a high density of macrophages were more than twice higher in depressed suicides than healthy controls. Moreover, gene expression of ionized calcium-binding adapter molecule 1 (IBA1) and MCP-1 was significantly upregulated in depressed suicides, and mRNA for CD45 was significantly enhanced in depressed suicides.

25Two studies reported the existence of inflammatory cytokines abnormalities in depressed, suicidal patients with suicidality vs. healthy individuals. IL-6 was significantly elevated in melancholic depressive patients (n = 29) compared to healthy controls (n = 39), while no differences were found between patients with atypical depression (n = 18) and healthy volunteers (n = 39). Lower TNF-α serum level was found both in melancholic and atypical depressed individuals compared to healthy subjects. A positive correlation between cytokine levels and atypical depression was also found. Importantly, the duration of lifetime exposure to antidepressants correlated with IL-6 serum levels in both melancholic and atypical depressed subjects. In addition, Li and colleagues reported that plasma TNF-α levels were significantly decreased following venlafaxine treatment in a sample of 64 first-episode drug-naïve MDD patients when compared with 64 matched healthy controls. Relative to non-responders, responders had a greater reduction in TNF-α levels, which was associated with a greater reduction rate of depressive symptoms. The plasma TNF-α levels were equally higher in both suicidal and non-suicidal MDD patients compared to healthy controls at admission.

26According to the main findings of this systematic review, neuroinflammation may play a relevant role in the pathophysiology of suicidal behavior. However, based on the selected studies and in line with a previously published paper of our research group upon the same topic, the existence of an association between inflammatory cytokines abnormalities and suicidal behavior does not necessarily reflect a specific causal link. Among the selected studies, eleven reports analyzed inflammatory cytokines abnormalities in suicidal patients, four explored cytokines abnormalities in patients with both a positive history of suicide attempts and suicidal ideation, two investigated the existence of inflammatory cytokines abnormalities levels in depressed patients with suicidal ideation vs. those abnormally without, seven examined the inflammatory cytokines abnormalities in patients with a history of suicide attempts vs. those without, three analyzed patients died by suicide vs. those died for other causes, and two focused on depressed suicidal patients with suicidality.

27There are studies showing the existence of a distinct biological profile underlying suicidal behavior with inflammatory dysregulation that may be significantly involved in the physiopathology of mood disorders with suicidality. Overall, the existence of an imbalance among pro-inflammatory (e.g., IL-1β, IL-2, IL-6, IFN-γ, and TNF-α) and anti-inflammatory cytokines (e.g., IL-4 and IL-10) has been reported even in untreated depressed patients.

28Systematic reviews and meta-analyses confirmed the relation between circulating inflammatory markers, major affective disorders, and suicidal behavior with significant inflammatory changes observed in the periphery, CSF, and brain tissues of suicidal patients.

29In particular, the meta-analysis of Enache et al. supported the existence of abnormally increased IL-6 and TNF-α concentrations in CSF and brain parenchyma, together with an increased microglia activity and abnormally reduced astrocytes/oligodendrocytes markers levels in MDD. The authors supposed that the reduced number of astrocytes can induce impairments in the integrity of blood-brain barrier with enhanced monocyte recruitment and inflammatory infiltration, which has been confirmed by post-mortem studies in MDD populations.

30What are the neurobiological mechanisms underlying the link between abnormally inflammatory cytokines levels and suicidal behavior? The first demonstration of a causal link between inflammation and subsequent onset of depression comes from the initial treatment trials using IFN in patients with specific infections or IFN-β in multiple sclerosis with depressive symptoms and suicidal behavior; according to these studies, depressive symptoms may emerge approximately one month after the first administration of this medication.

31Based on most recent evidence, neuroinflammation may determine abnormalities in the kynurenine pathway, resulting in increased quinolinic acid production, which agonizes N-methyl-D-aspartate (NMDA) receptors resulting in neurotoxic effects potentially contributing to the pathophysiology of both major affective disorders and suicidal behavior. Specifically, inflammatory cytokines are able to activate the indolamine 2,3-dioxygenase (IDO), a specific enzyme which is located on microglia and astrocytes. IDO is implicated in the catabolism of TRP, which is involved in serotonin neurotransmission. Kynurenine pathway usually produces kynurenic and quinolinic acid, which are highly neurotoxic metabolites able to activate NMDA receptors. An altered production of both kynurenic and quinolinic acid are supposed to be related to a severe unbalance between microglia and astrocytes activation.

32According to our findings, the most consistent link between inflammatory changes and suicidal behavior may be found in studies directly analyzing cytokines abnormalities and suicidal patients (completed suicides or those with a history of suicide attempts) when compared to healthy controls. Most of these studies documented abnormally higher levels of IL-1β, IL-6, TNF-α, TGF-β1, VEGF, and KYN, and lower levels of IL-2, IL-4, and IFN-γ predominantly in specific brain regions such as OFC, right ACC, insula, and dlPFC of suicidal patients (mainly suicide attempters). Abnormally elevated IL-6 and IL-10 levels have also been reported in patients with active suicidal ideation when compared to healthy controls.

33Based on functional neuroimaging studies, social exclusion may pathologically activate brain regions like the insula, OFC, and ACC playing a fundamental role in suicidal vulnerability, which may be enhanced by neuroinflammation. In addition, inflammatory cytokines may lead to significant behavioral and emotional changes according to their direct effects in specific brain regions. However, other studies support the notion that the inflammatory modulation of the neural response to social exclusion may be directly related to specific psychiatric conditions, independently of suicidal behavior. To this specific regard, the authors found that IL-1β was negatively associated with right OFC activation in explicit social exclusion vs. social inclusion, whereas IL-2 was positively associated with activation of the right ACC insula and OFC in explicit social exclusion vs. social inclusion, even after controlling for group, indicating that these changes were independent of suicidal status.

34Not all studies supported the association between abnormally elevated cytokines concentrations and suicidal behaviors. For instance, the finding of abnormally reduced TNF-α in suicidal depressed young adults compared to non-suicidal controls is not in line with what most adult studies usually indicate. According to a systematic review on this topic, this can be due to the existence of different underlying pathophysiology concerning suicidal behaviors in adolescents and young adults. Increased suicide risk has also been found in children and adolescents treated with antidepressant medications, which should be closely monitored, especially in the initial treatment phases. The increased suicidal risk in this patient population may be related to initially induced antidepressants effects in youths such as enhanced activation, hyperarousal associated with impulsivity, restlessness, and/or insomnia.

35We firmly suggest that further additional studies using longitudinal designs should test the association between abnormal inflammatory response and suicidal behaviors in homogeneous populations after controlling for the presence of major depression.

36This study needs to be considered according to the following shortcomings. First, the majority of studies included in the present review (76.7%) adopted a cross-sectional study design and lack of appropriate longitudinal follow-up periods. In addition, patients with different diagnostic subtypes (e.g., melancholic, atypical, etc.) and heterogeneous treatment and disease duration have been selected. Moreover, the selection of mixed samples of patients (e.g., patients with different psychiatric diagnoses or comorbid psychiatric conditions) does not permit the generalization of the main results of the selected studies. In addition, some studies lack control groups or healthy controls may not be appropriately matched for socio-demographic and clinical characteristics. Furthermore, both psychiatric symptoms and suicidal behavior may have been evaluated only clinically and not using specific psychometric measures. Most studies (e.g., postmortem studies) did explore the effects of specific inflammatory cytokines on some brain regions (such as OFC, ACC, and PFC) rather than others; the interaction with other significant neurobiological variables may not have been taken into account as well. Importantly, the short- and long-term immunoregulatory effects of psychoactive treatments taken by patients may represent additional confounding factors in specific studies. Furthermore, although conducted on the most recent evidence-based contributions in the current literature, the reviewed studies are based on the authors’ choice and do not necessarily represent the most representative studies upon the main topic. Notably, we were not able to include within the main text a risk of bias assessment that would have helped us to establish transparency of evidence synthesis results and findings for each of the included studies.

37Finally, most studies could not exclude the exact contribution of MDD as a mediator/moderator of the link between inflammatory cytokines abnormalities and suicidal behavior. Inflammatory changes have been reported to occur with great variability not only in MDD, but also in other psychiatric conditions such as schizophrenia and related psychoses in which the role of the immune abnormalities has been consistently reported. To this specific regard, schizophrenia has been linked to the disruption of the cytokine milieu and the enhanced propensity for the production of specific proinflammatory cytokines. In addition, given the most frequent cross-sectional nature of the studies’ designs, it is not possible to exactly determine whether inflammation precedes depression/suicide or vice-versa. Thus, how and to what extent inflammatory cytokines abnormalities are related to MDD or other comorbid psychiatric conditions and not exclusively related to suicidal behavior is not quite unclear.

38Based on the most relevant results of the present review and in line with previous findings, neuroinflammation may play a crucial role in the pathophysiology of suicidal behavior. Specific inflammatory changes may be identified in the periphery, CSF and central nervous system of patients at risk for suicide. More detailed knowledge of the pathophysiological mechanisms underlying suicidal behavior, including possible mediators/moderators of the inflammatory response that are able to enhance vulnerability or resilience to suicide is really desirable for both clinicians and researchers. Importantly, it would be appreciated to identify specific subgroups of suicidal patients who also manifest an increased vulnerability to inflammation in order to more directly detect specific at-risk subjects as well as target homogeneous populations to test the precision of possible anti-inflammatory agents.

39Currently, we are only at the beginning of this exciting journey that might lead clinicians in the next future to use immunomodulatory treatments alone or in combination with existent available psychotropic medications in order to manage psychiatric conditions such as major affective disorders and suicidality. Further additional studies are required in order to clarify the complex pathophysiological mechanisms of the immune pathways underlying suicidal behavior.