At the time of injury, in children with both sTBI and mTBI, IL-6 was significantly elevated compared to controls. When stimulated ex-vivo with LPS, in sTBI and not mTBI, significant elevations of IL-8 and IL-10 were seen. The mTBI cohort showed significantly lower IL-8, IL-10 IL-17A and TNF-α at the time of injury. However, when stimulated with LPS, these children with mTBI had greater IL-8 and TNF-α responses, both in serum from the time of injury and serum at 2 weeks, showing persistent immune activation, at a time when the majority of children show clinical recovery. Serum taken two weeks from injury showed the same pattern of significantly decreased IL-17A, IL-8, IL-10 and TNF-α levels, while IL-6 had returned to baseline.
There are gaps in our understanding in how to modify the innate immune response to trauma to better the outcomes in TBI across the spectrum, particularly in children where the innate immune responses are different. Limiting the pro-inflammatory response and promoting anti-inflammatory properties is key to therapeutics. Defining these responses and timelines are important for secondary injury, particularly in the milder TBI children, not previously known to have persistent immune dysregulation. Early measurement of innate immune function may better help to risk stratify in mTBI.
Interleukin-6 was elevated in all children with TBI and predicted sTBI. IL-6/IL-10 ratio enhanced the prediction of mTBI relative to IL-6 alone. In sports related concussion IL-6 levels correlated with symptom burden in young adults [8]. While IL-6 is elevated in mTBI in adult studies it has not correlated with cognitive impairment[24], a common component of post concussive syndrome. Gill and colleagues showed significant increases in IL-6, IL-10 and TNF-α during blast exposure in military training [25]. In our study IL-6 elevations did not persist at two weeks. Interestingly exercise prior to injury in animal models have been shown to attenuate levels of IL-6 in response to mTBI [26], and also the effect of the injury [27].
In childhood sTBI is associated with increased serum IL-8 and predicted a worse outcome [7] In neonatal brain injury elevated IL-8 and IL-10 correlated with poor outcome and severity of injury [28]. Most TBI studies of cytokine responses have been in sTBI but not mTBI, where there is no overt intracranial bleeding, and the mechanism of injury is different. We demonstrated an opposite effect in this cohort with significantly lower levels of IL-8 and IL-10 in our mTBI cohort at baseline when compared to controls. An adult study looking at blast-related TBI in the military found similar responses to our study: IL-8 decreased from pre-deployment levels in those who suffered blast injury[29]. Interestingly this effect was more profound in those younger than 25 years. This may be an effect of age or may also suggest the response to all brain injury is not on a continuum.
There was a significant and sustained decrease in IL-17A following TBI. Only in sTBI did LPS mediate an IL-17A response to stimulation. Interleukin-17A is associated with TBI severity and also progression in multiple sclerosis. It has a role in mediating the blood brain barrier [30]. A severe TBI model in rats showed increased IL-17 production [31]. The role of the adaptive immune system in TBI, particularly in repeated mTBI needs further clarification.
Two weeks from injury there was a sustained decrease in IL-8, IL-10 IL-17A and TNF-α in the mTBI cohort. Animal studies demonstrate persistent immune dysfunction in severe TBI [3] and chronic changes associated with repeated head injury [32]. The mouse model of head injury showed chronic persistent changes in thymic and T cell function, neutrophil and monocyte function, and proinflammatory cytokine production including TNF-α, at 60 days [3]. In moderate/severe blast injury there was a reduction in IL-10, IL-17A and TNF-α even at a year from injury reflecting persistent inflammation [33]. Following neonatal encephalopathy children have persistent immune dysfunction much longer, at school-age[4]. Jeschke and colleagues demonstrated altered immune response in the three years following a severe physical burn injury in children [34]. The response of the immune system over time to one TBI and repeated insults may be amenable to immunomodulation.
Immunomodulatory effects of TBI are demonstrated in both cohorts. Provocative testing with LPS stimulation of samples from children with TBI unmasks immune dysregulation and may aid diagnostics. sTBI had significantly elevated IL-8 and IL-10 response to LPS while mTBI had a significant and sustained IL-6 and IL-8 response to LPS challenge, at time of injury and at two weeks. The greater TNF-α response at 10–14 days was specific to mTBI, suggesting priming. Prior studies of TNF-α responsive to LPS looked only at sTBI and not mTBI. A blunted TNF-α response in sTBI to LPS stimulation has been previously demonstrated in sepsis and trauma in children [35, 36]. This phenomenon results in increased nosocomial infections[37]. Our study is the first to look at children with mTBI/concussion which shows priming of the TNF-α response to LPS at two weeks, with an opposite and greater response to LPS challenge, at time when the majority of children appear to have recovered.
Innate immune function in children is distinctly different to adults. Ageing is associated with low grade chronic systemic inflammation with elevated IL-6 and C-reactive protein [38]. Studying the effects of TBI in children naïve to chronic disease offers a unique opportunity to look at TBI in isolation. Understanding the pathophysiological and post inflammatory changes associated with paediatric TBI may facilitate the understanding of potential new drug therapies in the initial and chronic phases of TBI. Even the use and timing of basic anti-inflammatory medication is poorly understood. This study was limited by the small number of sTBI, but the large number in the mTBI cohort was a strength in this under studied population.