Interleukin-10 deficiency results in a spectrum of problems with the vasculature, including everything from vascular and endothelial damage from increased oxidative stress and inflammation to deleterious remodeling and an inability of the vasculature to respond to physiological demands such as the requirement for vasodilation. IL-10 inhibits Nox1, a subunit of NADPH oxidase, which plays a role in degenerative vascular remodeling by generating ROS, causing oxidative stress (25). IL-10 knockout (IL-10^−/−) mice have higher levels of Nox1, Nox2, and p22^phox (two additional NADPH oxidase components) and superoxide production and display decreased aortic medial thickness, a loss of smooth muscle cells, and increased vascular collagen deposition, indicating harmful vascular remodeling with IL-10 deficiency (25, 26). Administration of a superoxide scavenger in IL-10^−/− mice prevented vascular remodeling, suggesting the oxidative stress-dependent mechanism (ROS formation by NADPH oxidase) of injurious vascular remodeling with IL-10 deficiency (26). Additionally, IL-10 deficiency led to increased levels of matrix metalloproteinase-9 (MMP-9) in aortic smooth muscle cells, IL-6 in aortas, and the vasoconstrictor endothelin-1 in plasma (25).

Interleukin-10 attenuates endothelial dysfunction and vasoconstriction mediated by ROS, endothelin-1 (27), angiotensin II (28), and by ischemia-reperfusion injury (29), among other mediators, and these protective mechanisms may become more important with age. Old, but not young, IL-10^−/− mice have diminished vasodilatory responses to acetylcholine, while the nitroprusside response is intact, suggesting endothelial rather than vascular smooth muscle dysfunction (26). Additionally, viral transduction of canine basilar arteries with IL-10 increases vasodilatory responses and reduces levels of ICAM-1 and VCAM-1, further supporting the protective role of IL-10 on the endothelium (29). IL-10 has also been shown to act via the AKT pathway to decrease TNFα-directed ceramide synthesis, resulting in lower levels of ROS and ICAM-1 (30). In aortic rings from IL-10^−/− mice, TNFα reduces endothelial nitric oxide synthase expression and vasodilatory ability, and IL-10 administration restores function, providing a protective effect (31). A similar experiment showed that murine aortic rings treated with angiotensin II showed impaired relaxation that was reversible with IL-10 administration (32). High cyclooxygenase-2 activity also plays a role in these stiffer vessels seen in IL-10^−/− mice, ultimately resulting in decreased vascular relaxation, impaired cardiac function, and a larger heart size (33).

The source of endothelial protective IL-10 may come from CD4⁺CD25⁺ regulatory T cells (Tregs) (34), as well as the various other types of IL-10-producing cells, including B cells (see the “IS” section below). For example, hypertensive IL-10^−/− mice transduced with hypertensive wildtype (WT) Tregs have a better vasodilatory response to acetylcholine and lower levels of NADPH oxidase, whereas hypertensive IL-10^−/− Tregs does not confer protection on hypertensive WT mice (34). In middle-aged spontaneously hypertensive rats that display features similar to early-stage human cerebral small vessel disease, there is a decreased level of IL-10 in the cerebrospinal fluid and other immune changes (35).

Of interest, inflammation mediated by toxic bacterial agents, such as lipopolysaccharide (LPS), is more severe in IL-10^−/− mice (36). Surprisingly, cerebral bacteremia may result in so much endothelial damage in the absence of IL-10 as to cause ICH and death shortly following peak bacterial loads (37). These findings are associated with increases in the FAS/FAS-ligand apoptotic pathway, which IL-10 reverses (37). However, the effects of IL-10 seem to be specific to the noxious stimuli generating its production. For example, IL-10 produced in response to both LPS and Borrelia burgdorferi reduces lymphocyte endothelial migration and blunts endothelial production of chemokines; however, IL-10 produced in response to IL-1β and TNFα does not show such effects (38).

Given the intimate tie between the vascular endothelium and pathophysiology of all forms of acute brain injury, it is important to consider the effects of IL-10 on the vasculature. The above studies have shown that following injury, the inflammatory system works on many different levels to cause endothelial cell damage and vascular dysfunction. IL-10 appears to play a central and multifaceted role in attenuating these effects and facilitating the resolution phase of the inflammatory system. In contrast, low or absent IL-10 leads to several changes in gene expression that ultimately results in deleterious vascular remodeling and impaired vascular relaxation in response to physiologic mediators (Figure 1), outcomes that would exacerbate secondary brain damage following acute injury. In summary, IL-10 plays a crucial role in restoring vascular function following injury to the vasculature, similar to that which occurs after IS, SAH, ICH, and TBI.

Cortical neuron cultures from IL-10^−/− mice are more susceptible to neurotoxicity following excitotoxicity and combined oxygen–glucose deprivation, and administration of exogenous IL-10 provides neuroprotection to cultures from both WT and knockout strains (42). This IL-10-mediated protection was later separately shown to be via the IL-10 receptor on cortical neurons and PI3K/AKT and STAT3 signal transduction pathways (87). The transcription factor Nrf2 is expressed widely throughout the body and upregulates the expression of numerous antioxidant genes in response to oxidative stress (88, 89). In astroglia preconditioned to oxidative stress, neuroprotection from oxidative stress–glucose deprivation was reported to be via an Nrf2/IL-10-dependent mechanism (90). Lipocalin-2 is significantly increased in neurons following stroke and is proposed to signal stress from injured neurons to supporting glia (91). Lipocalin-2 treatment of microglia results in glial activation, IL-10 release, and enhanced phagocytosis, and conditioned media from these microglia protect neurons from oxygen–glucose deprivation (91).

In a permanent middle cerebral artery occlusion (MCAO) model, IL-10^−/− C57BL/6 mice have 30% larger infarct volumes compared with WT mice 24 h post-stroke (42). In this model, WT and IL-10^−/− mice display similar degrees of proinflammatory molecules within the first 6 h. However, after 4 days, the IL-10^−/− mice have higher levels of proinflammatory molecules that persist through the end of the first week after stroke (43).

Both systemic intravenous (IV) and central intracerebroventricular (ICV) exogenous administration of IL-10 reduces infarct volumes following permanent MCAO (46). Similarly, in a transgenic C57BL/6J mouse model overexpressing IL-10 (56% increase in the brain and 200% increase in the plasma), infarct volumes were reduced by 40% 4 days following permanent MCAO (48). In the same study, these findings were associated with upregulation of free radical scavengers such as glutathione and manganese superoxide dismutase, reduced activity of the proapoptotic protein caspase 3, and downregulation of IL-1β, IFN-γ, and TNFα 1 day post-stroke (48). Injection of a recombinant adeno-associated viral vector serotype 1 expressing IL-10 into the cerebral artery of rats 3 weeks prior to MCAO results in elevated IL-10 serum levels 3 weeks after injection and decreases neurologic deficit scores, infarct volume, and neuronal injury (49). Systemic IL-10 overexpression in stroke-prone spontaneously hypertensive rats reduces the incidence of stroke, decreases stroke-associated symptoms, and improves survival (50). In Wistar rats, IL-10 is upregulated in viable neurons in the ischemic brain following permanent and transient MCAO, and hypertension blunts this response, potentially contributing to the worse outcomes in the hypertensive setting (41).

Interleukin-10 appears to exert its anti-inflammatory effects in part by downregulation of NFκB (92). Administration of hydrogen sulfide donors at the time of reperfusion protects BBB integrity after ischemia/reperfusion and is accompanied by enhanced IL-10 expression, reduced NFκB nuclear translocation, and MMP-9 and NOX4 activity (51). Moreover, upregulation of IL-10 and the STAT3 pathway and downregulation of NFκB have been proposed as the mechanisms for improved outcomes following cerebral ischemia in mice that overexpress IL-32α (52). A recent whole genome array in C57BL/6 mice showed that stroke upregulated 347 genes 24 h after reperfusion. In this study, ICV injection of IL-10 5 min after occlusion downregulated 341 of these genes (47).

Immune cells, including T and B cells, are important in attenuating neuroinflammation via the modulation of various cytokines and chemokines, with IL-10 playing a central immunomodulatory role (53, 58). Tolerizing mice to oligodendrocyte glycoprotein prior to MCAO reduces infarct size by 70% at 24 h and by 50% at 72 h, but this effect was not seen in IL-10^−/− mice (45). This protection is specifically mediated by IL-10 secreted from CD4⁺ T cells, which were able to protect mice from stroke only when originating from WT mice (44). Elevated levels of IL-10 also reduce the number of CD11b⁺ cells that may contribute to secondary infarct expansion via nitric oxide pathways (45). Histone deacetylase plays a role in activating the expression of Foxp3 on Tregs, which secrete IL-10, and IL-10 was suggested as the main mediator of attenuated infarct volume and behavioral deficits, reduced levels of proinflammatory cytokines, and increased number of Tregs in the brain of MCAO mice with histone deacetylase inhibition (53). Similarly, expansion of the Treg cell population in the CNS via administration of a CD28 superagonist monoclonal antibody at onset of reperfusion reduced infarct size 7 days following MCAO, and its effect was attributed to an increased amount of IL-10 (54). Transferring IL-10-producing B cells into B cell-deficient mice 24 h after MCAO reduced infarct size, the amount of T cells and monocytes in the brain parenchyma, and the peripheral proinflammatory milieu (58). A follow-up to this study showed similar results in B cell-sufficient mice (57). Notably, IL-10-producing B cells also upregulated the number of Tregs (58), perhaps indicating a positive feedback loop between B cells and Tregs, both of which are neuroprotective via IL-10 production.

Neurogenesis is a highly debated topic for potentially enhancing neural recovery following IS. Injection of activated Tregs into the lateral ventricle of C57BL/6 mice after 60 min of transient MCAO promotes neural stem cell proliferation in the subventricular zone in ischemic and normal mouse brains. However, this effect was eliminated by blocking IL-10 with a neutralizing antibody, indicating that activated Tregs act via IL-10 to promote neural stem cell proliferation (59). Hematopoietic cytokines such as GCSF and stem cell factor have been shown to promote neurogenesis (55) and also may be responsible for providing the initial signals to produce IL-10 in the setting of IS (56). Treatment with these cytokines early (1–10 days) and later (11–20 days) after MCAO markedly elevates IL-10 mRNA levels, reduces the levels of activated microglia/macrophages, and does not change proinflammatory cytokine expression in C57BL/6J mice (56). Another study where bone marrow-derived mesenchymal stem cells were transplanted into the lateral ventricle of Sprague-Dawley rats before permanent MCAO afforded similar results, where IL-10 mRNA and protein levels are increased for up to 4 days post-stroke, TNFα is decreased, infarct volumes are smaller, and neurologic function is preserved (60). Adipose tissue-derived stem cells trigger similar effects as mesenchymal stem cells (61) and are most effective when delivered intravenously at 24 h after MCAO (62). Thus, either administration of stem cells themselves or hematopoietic cytokines seem to improve outcomes following IS in part through increasing IL-10 levels.

Several epidemiologic studies have analyzed the genetic contribution of IL-10 to IS risk and outcome. A study in a Chinese population showed that lower basal serum levels of IL-10, an IL-10 promoter single nucleotide polymorphism (SNP) rs1800872 (AA vs. AC + CC), and IL-10 intron SNPs rs1554286 (TT vs. CT + CC) and rs3021094 (CC/CA vs. AA) are all associated with higher risk for IS (70). Similarly, subjects with low IL-10 production in response to LPS stimulation have a higher risk for fatal stroke (74). A study in a South Indian population showed an increased risk of IS with the A allele of the IL-10 promotor SNP rs1800896 (A vs. G), which is associated with low IL-10 production (63). In agreement, a recent meta-analysis showed that the IL-10 1082G/A SNP (rs1800896) is associated with increased risk of IS, but not the 819 C/T SNP (rs1800871) (64, 65). Several other studies further revealed that the IL-10 1082G/A SNP is associated with susceptibility for an IS (66, 67, 93, 94). However, in a different study, the IL-10 1082G/A SNP was not associated with the occurrence of an IS, but the GG genotype predicted early stroke progression and functional dependency independent of other standard risk factors (68). Similarly, another study showed that the IL-10 1082G/A SNP was significantly associated with the risk for specific subtypes of IS, but no significant association was found with the overall risk for an IS (69). A study in a Korean population showed that rs1518111 and rs1554286 are not associated with IS per se, but are associated with hypertension in the risk of IS (71). Similar results were seen in a Russian population, where the IL-10 627*C/*C genotype is protective against hypertension in male patients with stroke (72). One other study in an Eastern Finland population showed that IL-10 plasma concentration independently correlates with cardioembolic high-risk sources, suggesting its usefulness in improving diagnosis of stroke etiology (81). Lastly, in Caucasian patients with peripheral artery disease receiving elective revascularization, the IL-10ATA haplotypes that are associated with lower serum IL-10 levels correlate with a high risk for postoperative cardiovascular events (73).

The relative levels of IL-10 at various times following IS differ from controls and appear to have important implications on IS patient outcomes. For example, at 48 h after IS, mean serum IL-10 is significantly higher than in healthy or at-risk controls, and high IL-10 sera levels independently correlate with severe neurologic deficits (NIHSS ≥12) at 48 h post-stroke and predict major adverse clinical outcomes (recurrent IS, any cause of death, NIHSS ≥12) at 90 days (85). However, in another study, IL-10 levels within the first 72 h after IS were lower than in controls and no significant correlation with NIHSS at admission was observed (78). As compared to those who die, IS patients who improve have been shown to have lower IL-10 at 24 h, but increased levels at 72 h and 144 h (79). In other studies, low IL-10 levels are associated with acute neurologic decline post-stroke (76, 77). Yet another small study found no correlation with IL-10 and other cytokines and IS patient prognosis (86). IS patients with low levels of IL-10 had a higher incidence of hemorrhagic transformation (75), perhaps providing a link between IL-10 and endothelial integrity. At 7 and 28 days, IS patients appear to have lower number of Tregs and IL-1 and higher relative levels of proinflammatory cytokines such as IL-6 and IL-1β (80). In the Leiden 85-plus study, patients with a history of stroke displayed lower IL-10 production in response to LPS stimulation than subjects without stroke (74). No correlation between IL-10 levels and infarct volume has been found (81, 86). Considerable variation exists among the aforementioned studies and is likely due to the different study populations, relatively small cohort sizes, differences in study methodology, and the consistency or lack thereof for identifying and correcting for covariates. Additional large and controlled prospective studies are needed to establish the link between IL-10 and IS severity, recovery, and outcomes.

Ischemic stroke may induce a functional immunosuppressive state that may persist for several weeks, wherein patients may be more susceptible to infection, and infection is a leading cause of the high mortality rates seen long after an IS (82, 95). The temporal IL-10 profile appears to be different between those that develop stroke-associated infection (82, 83). Two studies have shown that high IL-10 levels at admission or 6 h following ischemia is an independent predictor of infection (82, 83). In one of those studies, TNFα and the TNFα/IL-10 ratio was decreased (82). Another study found that the levels of IL-10-secreting mononuclear cells (MNCs) were highly elevated in IS patients compared with healthy individuals (84). Thus, the immunomodulatory role of IL-10 may be an important consideration in understanding infectious processes and mortality after IS.

Ischemic stroke invokes a robust local inflammatory response in the brain that significantly contributes to secondary brain damage and poor outcomes. With IL-10 representing a canonical anti-inflammatory cytokine that intriguingly has a special role in the resolution of inflammation, it is not surprising that many studies investigating the role of IL-10 in IS have been performed. In vitro models have provided evidence for a neuroprotective function of IL-10 under hypoxic and excitotoxic conditions. In IS preclinical models, IL-10 reduces infarct volume via its protective effects on the vascular endothelium and attenuation of inflammatory cascades. Multiple epidemiologic genetic studies have in general shown that polymorphisms in the IL-10 gene that lower IL-10 levels increase the risk of IS. However, clinical studies measuring IL-10 levels after IS are more varied with regard to the role of IL-10 in predicting neurologic outcomes and complications. Even with all of the studies described here, it is clear that additional work is needed to clarify the role of IL-10 in predicting the risk for IS, modulating IS outcomes, and whether IL-10 has biomarker and/or therapeutic potential after IS.

Temporal changes in IL-10 are seen following experimental SAH, although these are not statistically significant (111–113). After autologous blood infusion into the cisterna magna of canines, IL-10 mRNA expression in the basilar artery tends to be increased for up to 14 days post-ictus (111). Similarly, IL-10 tends to be increased in the CSF at 6 h and then declines, but remains elevated through 48 h after autologous blood injection into the cisterna magna of male adult Sprague-Dawley rats (112). Interestingly, although acetylcholine was able to attenuate CV in this model, it had no effect on IL-10 levels (112). Last in an endovascular perforation model of SAH in Wistar rats, trends toward higher IL-10 mRNA in the cortex are seen at 48 h (113). In contrast to the less impressive changes in IL-10, these studies have shown highly increased levels of proinflammatory cytokines such as IL-1α (111), IL-1β (113), IL-6 (111, 112), IL-8 (111), and TNFα (111) at 48 h following experimental SAH. Indeed, Aihara and colleagues postulated that the anti-inflammatory effects of IL-10 were overwhelmed by these proinflammatory cytokines such that IL-10 was unable to counteract the deleterious inflammatory processes involved in CV pathophysiology and thereby improve SAH outcomes (111).

Studies in humans have also not shown significant changes in IL-10 after SAH (114–117). One study reported that IL-10 mRNA expression was virtually non-existent in aneurysm walls, although it was present in the walls of temporal artery controls (117). In microdialysate, when detectable, IL-10 was present in extremely small concentrations and remained at relatively constant levels through 7 days post-bleed (115). Similarly, IL-10 levels were constant through at least 10 days post-bleed in paired plasma and CSF samples, and the levels were not significantly different between the CSF and plasma (means were below 10 pg/ml) (116). In another study, mean IL-10 blood levels throughout the intensive care unit stay were also quite low (mean ≈11 pg/ml) (114). However, a study in an Indian population showed that IL-10 SNPs rs1800871 and rs1800872 are significantly associated with the incidence of intracranial aneurysms, independent of epidemiological factors, although it is unclear whether this increases the risk for aneurysm rupture and thus the incidence of SAH (118). Additional larger controlled studies are necessary to clarify the association of IL-10 with aneurysm formation and rupture and the role of IL-10 after SAH, including its prognostic and diagnostic predictive potential for SAH outcomes.

In comparison with the other forms of acute brain injury described herein, substantially fewer studies have explored the contribution of IL-10 in the setting of SAH. Preclinical studies have shown non-significant increases in IL-10 after SAH, accompanied by robust changes in proinflammatory cytokines, suggesting an imbalance in the inflammatory milieu that could possibly contribute to prolonged disease course and worse SAH outcomes. In addition, clinical studies have not shown significant changes in IL-10 after SAH, but no studies have investigated whether IL-10 affects clinical outcomes or complications such as CV and/or DCI, or whether IL-10 has therapeutic or biomarker potential. It is clear that additional work is needed to understand how IL-10 changes in the various biocompartments (serum, CSF, brain) after SAH and the resulting implications on SAH pathophysiology.

Microglia, a primary player in immune surveillance and the initiation of inflammation in the CNS (127), become activated shortly after ICH (12) and secrete proinflammatory cytokines such as TNFα, IL-6, and IL-1β (127). These microglia also secrete IL-10, which acts on astrocytes in the brain to redirect their focus from the production of a proinflammatory cytokine profile toward the production of TGFβ (127). TGFβ then feeds back to act on the microglia and attenuate their proinflammatory response. IL-10 can also increase CD36 expression and thereby enhance the erythrophagocytic ability of microglia, and with CD36 deficiency, mRNA levels of TNFα and IL-1β are higher and IL-10 levels are significantly lower in the perihematomal tissues (123). Additionally, IL-10 levels are significantly increased at 6 h and 7 days after ICH with respect to sham, suggesting that IL-10 may have early and late influences on ICH outcomes (121).

Interestingly, a study in IL-10^−/− mice suggests that the presence of IL-10 is protective against the development of cerebral hemorrhage (128). Cerebral hemorrhage and edema are attenuated by anti-TNFα therapy (128), indicating that the main protective effect of IL-10 is derived from its inhibition on the production of proinflammatory cytokines, namely, TNFα (129–131). This same effect is found in adult male Sprague-Dawley rats, where statins (2, 5, 10 mg/kg) produce dose-dependent increases in IL-10 levels and are associated with dose-dependent decreases in TNFα, as well as fewer activated microglia, sensorimotor deficits, and inflammation and less edema 3 days after ICH (122).

Only a couple of studies have looked at IL-10 genetic polymorphisms and the risk for developing ICH. The IL-10 1082G/A SNP has been identified as a risk factor for ICH in a North Indian population (132). This SNP was not associated with the risk for hemorrhagic presentation of brain arteriovenous malformations (133).

Interleukin-10 and the amount of IL-10-secreting MNCs are elevated in peripheral blood 2 days following ICH (84, 125). Another study found that systemic IL-10 release was significantly higher in ICH patients with left hemisphere hemorrhage compared with those with right hemisphere hemorrhage at 2 days, whereas no such correlation was found for IL-6 (126). Because dysfunction of the autonomic nervous system is a common ICH complication (134), the autonomic nervous system can be regulated asymmetrically, and the sympathetic nervous system can induce a notable increase in systemic IL-10 release, this study posits autonomic dysfunction as a potential mechanism for asymmetrical regulation of IL-10 after ICH (126).

At admission, IL-10 levels are negatively correlated with the Glasgow Coma Scale (GCS), although a stronger negative correlation was found for IL-6 and GCS (125). At 2 days post-bleed, IL-10 levels correlate with IL-6 levels, and IL-6 levels correlate with hematoma volume and mass effect, but IL-10 does not (125). Hematoma expansion is a major cause of morbidity and mortality after ICH, and inflammation may be associated with its pathogenesis (119). In spontaneous ICH, higher plasma IL-10 levels on admission are associated with hematoma expansion and worse 30-day outcomes (119). Another study also found that plasma IL-10 levels were higher in ICH patients with poor outcome (modified Rankin Scores of 3–6) at 1 month (124), although this study did not utilize a multivariate model.

Similar to IL-10 and SAH, few studies have explored the contribution of IL-10 in the setting of ICH, although IL-10 appears to be elevated following experimental and clinical ICH. In preclinical models, IL-10 exerts a protective effect against spontaneous cerebral hemorrhage by downregulating TNFα, a key proinflammatory cytokine, and some weak evidence exists suggesting that IL-10 is neuroprotective after ICH. Clinically, very little is evident regarding the contribution of IL-10 polymorphisms. However, other studies generally have found that IL-10 levels correlate with the extent of initial brain injury and have prognostic value, suggesting both diagnostic and prognostic biomarker potential for IL-10 after ICH.

Understanding the temporal relationship between the injury and elaboration of IL-10 provides insight into its immunomodulatory role after TBI. A study in adult male Sprague-Dawley rats found that brain IL-10 mRNA expression increases immediately following TBI, whereas IL-10 protein levels are stable initially, and only begin to rise rapidly after 2 h (135). These findings suggest that the surge in IL-10 levels is due to an increase in local IL-10 synthesis rather than systemic IL-10 entering through a leaky BBB. IL-10 protein levels continue to rise through 24 h, although at a slower rate (135). No change in plasma IL-10 levels is seen throughout this timeframe (135). Following repetitive mild TBI, TNFα and IL-6 are higher, and the expression of IL-10 is lower when compared to a single TBI group on days 1, 3, 7, 14, and 30 post-injury (145).

Studies with IL-10^−/− mice have suggested that IL-10 is beneficial after TBI, although IL-10 administered after TBI as a therapeutic agent has provided mixed results depending on the experimental paradigm and administration route. At 4 weeks after TBI, 8-week-old IL-10^−/− C57BL/6 female mice have larger lesion volumes, increased edema and inflammation, worse motor and cognitive function, and higher levels of BBB dysfunction and apoptosis after controlled cortical impact-induced TBI (142). Additionally, using the lateral fluid percussion TBI model, SQ IL-10 (100 µg) administered 1 h after TBI results in a significant reduction in lesion volume and edema at 5 days, as well as improved recovery of motor and cognitive function (142). In another study using the lateral fluid percussion model and adult male Sprague-Dawley rats, both SQ and IV IL-10 (100 µg) improved neurological recovery, but ICV dosing (1 or 6 µg) did not (147). Although not statistically significant, survival rates in the SQ and IV groups were higher than in the controls, and the ICV group had dose-dependent lower survival rates (147). IV administration also reduced TNFα and IL-1β expression in the injured cortex and hippocampus (147). Although these studies have shown that systemic IL-10 treatment reduces neuroinflammation and improves neurological recovery (142, 147), a different study found that while systemically administered IL-10 (5 µg) does reduce inflammation (75% reduction in neutrophil accumulation) at 5 h, there is no improvement in motor or cognitive recovery (143). Last, central IL-10 administration following corticectomy in adult female CD1 mice results in a reduction in the number of reactive astrocytes and TNFα levels at 4 days (141), corroborating previous studies showing a reduction in neuroinflammation with IL-10 treatment.

A couple of studies with drugs or therapeutic regimens have demonstrated improved TBI outcomes through increased IL-10 protein levels and IL-10-dependent mechanisms. TBI has a significant ischemic injury component associated with poor cerebral blood flow (157). Hyperbaric oxygen (HBO) therapy has been suggested to provide some therapeutic benefit in conditions where poor blood flow and hypoxia lead to secondary tissue injury, and HBO therapy following TBI improves outcomes (142, 158). HBO therapy increases IL-10 at 4 days in adult Sprague-Dawley rats and mice above the levels induced by TBI alone in both the lateral fluid percussion (146) and the controlled cortical impact (142) models. The reported beneficial effects of HBO therapy after TBI are thought to be through an IL-10-dependent mechanism, as the smaller lesion volumes, less edema, improved motor and cognitive recovery, decreased inflammation, reduced apoptosis and BBB dysfunction 4 weeks after TBI with HBO therapy are only seen in WT mice and not in IL-10^−/− mice. Triptolide, an anti-inflammatory molecule, given immediately after TBI significantly increases IL-10 levels in the brain after 1 day and attenuates increases in proinflammatory cytokines; improves neurobehavioral outcomes; and reduces edema, contusion volume, and apoptosis (144).

The temporal profile of IL-10 in clinical TBI studies is not conclusive. Early reports found that plasma IL-10 levels peak within the first 3 h (151, 152), while a later study showed the peak is between 5 and 6 days post-injury (148). IL-10 levels may remain elevated for up to 22 days (131, 139) or even up to 6 months (137), and in some cases, there is a second peak in concentration (131, 152). In contrast to this general pattern of a rise in IL-10 levels followed by a gradual decline in the CSF and plasma (153), others have been unable to identify any pattern of IL-10 levels in microdialysate (115). Another study in patients with a severely disrupted BBB found high initial CSF IL-10 levels that decreased over time and serum IL-10 levels that increased over time (156), supporting a heightened intrathecal IL-10 synthesis after TBI with overflow of the cytokine into the systemic circulation (137), augmenting baseline systemic IL-10 levels. However, there is inconsistency in the literature as to whether IL-10 levels are more dramatically increased in the CSF or serum following TBI (131, 139), rendering it difficult to determine the source of increased IL-10 levels. This discrepancy likely relates to isolated head injury versus multiple injuries, as one study found that serum levels were higher than CSF levels in patients with additional injuries; however, in patients with isolated TBI, CSF concentrations were greater than or equal to the corresponding serum levels (155). Isolated head injuries have also been shown to result in either no difference in serum IL-10 levels (129) or lower IL-10 levels than that seen with multiple injuries (151, 152). Moreover, serum levels of IL-10 are elevated in trauma patients with and without brain injury (150, 159). Thus, the heterogeneous nature of clinical TBI patients may be in part responsible for the inconclusive patterns and biocompartmental distribution of IL-10 following TBI.

Similarly, it seems that clinical severity measures and TBI complications such as GCS, various Injury Scores/Scales, and BBB dysfunction/disruption have not been definitively correlated with IL-10 levels. Although most studies found no correlation with GCS (129, 130, 153–155) or Injury Severity Score (129, 155) and IL-10 levels, a minority endorse a significant association (149, 151). Blood–brain barrier BBB dysfunction/disruption is one component of secondary brain damage after TBI (8, 135, 139, 147, 160), and although more severe BBB dysfunction is evident in non-survivors of TBI (156), no significant correlation can be made between IL-10 levels in either the serum or CSF and the degree of BBB disruption (131, 139, 156). Disruption to BBB integrity often contributes to elevated intracranial pressure (ICP), another common complication of TBI (140, 161) that is associated with poor outcomes (136, 140). Although high ICP coupled with sympathetic activation has been shown to increase systemic IL-10 levels (162), and some studies have demonstrated higher IL-10 levels in patients with both high ICP and unfavorable outcomes (155, 156), not all studies have found such an association between ICP and IL-10 levels (131, 150). Although some interleukins exhibit a stronger and more prolonged response in females (115), no studies have found a clear association between IL-10 levels and gender (115, 130). One study has shown that increased IL-10 levels after TBI are associated with age, such that increased levels after TBI are found with ages less than 4 years (154).

Interleukin-10 is elevated after TBI and exhibits a more prolonged response when compared to other cytokines (115), and although many have attempted to correlate IL-10 levels with outcomes, this effort has not been reliably successful (129, 130, 136, 139, 150, 153, 163–166). However, a few studies did find that elevated IL-10 concentrations are associated with unfavorable outcomes (137, 149, 154–156). Higher CSF IL-10 in pediatric patients (154) and elevated serum and CSF IL-10 levels in adult patients were significantly associated with mortality (156). Non-survivors demonstrate higher serum IL-10 levels at admission that continued to rise over the next 24 h, whereas survivors have lower IL-10 serum levels at admission that subsequently decrease (156). Additionally, non-survivors have higher CSF IL-10 levels at admission; however, both the non-survivors and survivors demonstrate decreasing CSF IL-10 levels over time (156). In another study, elevated IL-10 levels at 10 or 30 h after TBI were 6 and 5 times more frequently associated with hospital mortality, independent of GCS, age, and systemic trauma (149). It appears that early IL-10 correlates best with outcome, as both admission CSF and serum IL-10 levels within the first 30 h correlate with mortality, whereas later time points do not (149, 156). Increased CSF and serum levels correlate with a GCS score less than 4 (137, 155). Additionally, in a multivariate analysis, one study found that a higher IL-6/IL-10 ratio from 2 weeks to 3 months was significantly associated with a GCS score less than 4 at 6 months (78). Hypothermia is known to attenuate the proinflammatory response following TBI¹⁴⁷; however, monitoring of IL-10 levels in severe TBI patients divided into hypothermia and normothermia treatment groups revealed no significant differences in CSF or serum IL-10 levels, nor outcomes in pediatric or adult patients (130, 153, 167).

It is well documented that IL-10 levels in humans and experimental models increase in both the serum and CSF shortly after TBI (8, 129, 135, 151, 152, 168) and remain elevated for many days (131, 152) followed by a slow decline. However, due to methodological variations in studies, the prognostic and diagnostic value of this cytokine remains unclear.

Traumatic brain injury is a heterogenous type of acute brain injury that involves a complex interplay of both direct primary injury and secondary injury, the latter of which is closely linked to neuroinflammatory processes. Preclinical studies have shown that IL-10 reduces neuroinflammation following brain trauma and, in general, IL-10 treatment improves neurological outcomes after TBI. Clinically, several studies have shown that IL-10 levels increase after TBI, although the temporal profile of IL-10 levels and whether IL-10 correlates with initial injury severity is less clear. Regarding the prognostic potential, it appears that IL-10 has the most utility in predicting mortality after TBI, although given the controversy in the literature, future work is necessary to further define the role of IL-10 in predicting other outcomes.