Examining four blood biomarkers for the detection of acute intracranial abnormalities following mild traumatic brain injury in older adults

Blood-based biomarkers have been increasingly studied for diagnostic and prognostic purposes in patients with mild traumatic brain injury (MTBI). Biomarker levels in blood have been shown to vary throughout age groups. Our aim was to study four blood biomarkers, glial fibrillary acidic protein (GFAP), ubiquitin C-terminal hydrolase-L1 (UCH-L1), neurofilament light (NF-L), and total tau (t-tau), in older adult patients with MTBI. The study sample was collected in the emergency department in Tampere University Hospital, Finland, between November 2015 and November 2016. All consecutive adult patients with head injury were eligible for inclusion. Serum samples were collected from the enrolled patients, which were frozen and later sent for biomarker analyses. Patients aged 60 years or older with MTBI, head computed tomography (CT) imaging, and available biomarker levels were eligible for this study. A total of 83 patients (mean age = 79.0, SD = 9.58, range = 60-100; 41.0% men) were included in the analysis. GFAP was the only biomarker to show statistically significant differentiation between patients with and without acute head CT abnormalities [U = 280, < 0.001, = 0.44; area under the curve (AUC) = 0.79, 95% CI = 0.67-0.91]. The median UCH-L1 values were modestly greater in the abnormal head CT group vs. normal head CT group [U = 492, p = 0.065, r = 0.20; AUC = 0.63, 95% CI = 0.49-0.77]. Older age was associated with biomarker levels in the normal head CT group, with the most prominent age associations being with NF-L ( = 0.56) and GFAP ( = 0.54). The results support the use of GFAP in detecting abnormal head CT findings in older adults with MTBIs. However, small sample sizes run the risk for producing non-replicable findings that may not generalize to the population and do not translate well to clinical use. Further studies should consider the potential effect of age on biomarker levels when establishing clinical cut-off values for detecting head CT abnormalities.