Abstract
Diagnosing traumatic brain injury (TBI) from body fluids in cases where there are no obvious external signs of impact would be useful for emergency physicians and forensic pathologists alike. None of the previous attempts has so far succeeded in establishing a single biomarker to reliably detect TBI with regards to the sensitivity: specificity ratio in a post mortem setting. This study investigated a combination of body fluid biomarkers (obtained post mortem), which may be a step towards increasing the accuracy of biochemical TBI detection. In this study, serum and cerebrospinal fluid (CSF) samples from 30 acute lethal TBI cases and 70 controls without a TBI-related cause of death were evaluated for the following eight TBI-related biomarkers: brain-derived neurotrophic factor (BDNF), ferritin, glial fibrillary acidic protein (GFAP), interleukin 6 (IL-6), lactate dehydrogenase, neutrophil gelatinase-associated lipocalin (NGAL), neuron-specific enolase and S100 calcium-binding protein B. Correlations among the individual TBI biomarkers were assessed, and a specificity-accentuated threshold value analysis was conducted for all biomarkers. Based on these values, a decision tree modelling approach was performed to assess the most accurate biomarker combination to detect acute lethal TBIs. The results showed that 92.45% of acute lethal TBIs were able to be diagnosed using a combination of IL-6 and GFAP in CSF. The probability of detecting an acute lethal TBI was moderately increased by GFAP alone and considerably increased by the remaining biomarkers. BDNF and NGAL were almost perfectly correlated (p = 0.002; R(2) = 0.944). This study provides evidence that acute lethal TBIs can be detected to a high degree of statistical accuracy using forensic biochemistry. The high inter-individual correlations of biomarkers may help to estimate the CSF concentration of an unknown biomarker, using extrapolation techniques.