Axonal white matter injury is believed to be a major determinant of adverse outcomes following traumatic brain injury (TBI). We hypothesized that measurement of neurofilament light protein (NF-L), a protein found in long white-matter axons, in blood samples, may serve as a suitable biomarker for neuronal damage in TBI patients. To test our hypotheses, we designed a study in two parts: i) we developed an immunoassay based on Single molecule array technology for quantification of NF-L in blood, and ii) in a proof-of-concept study, we tested our newly developed method on serial serum samples from severe TBI (sTBI) patients (n = 72) and controls (n = 35). We also compared the diagnostic and prognostic utility of NF-L with the established blood biomarker S100B. NF-L levels were markedly increased in sTBI patients compared with controls. NF-L at admission yielded an AUC of 0.99 to detect TBI versus controls (AUC 0.96 for S100B), and increased to 1.00 at day 12 (0.65 for S100B). Importantly, initial NF-L levels predicted poor 12-month clinical outcome. In contrast, S100B was not related to outcome. Taken together, our data suggests that measurement of serum NF-L may be useful to assess the severity of neuronal injury following sTBI.

The increase in ventricular CSF (vCSF) total tau probably reflects axonal damage, known to be a central pathologic mechanism in traumatic brain injury (TBI). These results suggest that vCSF total tau may be an important early biochemical neuromarker for predicting long-term outcome in patients with a severe TBI.

No abstract

Severe traumatic brain injury (TBI) may result in widespread damage to axons, termed diffuse axonal injury. Alzheimer's disease (AD) is characterised by synaptic and axonal degeneration together with senile plaques (SP). SP are mainly composed of aggregated beta-amyloid (Abeta), which are peptides derived from the amyloid precursor protein (APP). Apart from TBI in itself being considered a risk factor for AD, severe head injury seems to initiate a cascade of molecular events that are also associated with AD. We have therefore analysed the 42 amino acid forms of Abeta (Abeta1-42) and two soluble forms of APP (alpha-sAPP and ss-sAPP) in ventricular cerebrospinal fluid (VCSF) and Abeta(1-42) in plasma from 28 patients in a serial samples 0-11 days after TBI. The levels of alpha-sAPP, ss-sAPP and Abeta(1-42) were determined using ELISA assays. After TBI, there was a significant stepwise increase in VCSF-Abeta(1-42) up to 1173 % from day 0-1 to day 5-6 and in VCSF-beta-sAPP up to 2033 % increase from day 0-1 to day 7-11. There was also a slight but significant increase of VCSF-beta-sAPP from day 0-1 to day 5-6 and day 7-11. By contrast, the plasma- Abeta(1-42) level is unchanged after injury. The marked increase in VCSFAbeta(1-42) implies that increased Abeta expression may occur as a secondary phenomenon after TBI with axonal damage. The unchanged level of plasma-Abeta(1-42) in contrast to the marked increase in VCSF-Abeta(1-42) after severe TBI, supports the suggestion that plasma Abeta(1-42) does not reflect Abeta metabolism in the central nervous system (CNS).

Both S100A1B and S100BB were related to outcome after severe traumatic brain injury. Even though this study is small, it seems unlikely that separate analyses of the dimers are of any advantage compared with measuring S100B alone.