A-359 Comparison of Whole Blood and Plasma GFAP and UCH-L1 Blood-Based Biomarkers for Screening Patients with Mild Traumatic Brain Injury

Abstract

The diagnosis of traumatic brain injury (TBI) in the emergency department can be challenging and typically relies on brain CT scans, which carry associated costs and risks from ionizing radiation. Blood-based biomarkers, such as glial fibrillary acidic protein (GFAP) and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1), have been approved in the last years by regulatory agencies to assist in evaluating patients with mild TBI. While most studies have focused on plasma specimens, there is limited information regarding the correlation between plasma and whole blood due to the reduced stability of whole blood specimens. In this study, we have evaluated the analytical and clinical performance of GFAP and UCH-L1 blood-based tests in patients with suspected TBI admitted to the emergency department at a large academic hospital prior to the implementation of the test in clinical practice. Assay linearity and analytical measurement range was verified in duplicate using linearity material (Abbott). Within-run and day-to day precision (twice over 10 days) were tested using two TBI controls (Abbott). GFAP and UCH-L1 concentrations were measured in plasma and whole blood using multiplex immunoassays with iSTAT Alinity cartridges. Residual specimens were collected from eligible patients (N=80), aged 18 years or older, presenting with head injury or trauma as the chief concern and for whom laboratory tests were ordered as part of routine clinical care. Only 56% of the patients tested had a head CT scan performed. Samples were collected in lavender-top EDTA tubes and tested in parallel: whole blood was loaded first, followed by plasma after centrifugation, all within one hour of collection. Agreement was assessed using percent agreement (for binary outcomes), Bland–Altman plots, Passing–Bablok regression, and Pearson correlation, for both qualitative and quantitative values based on assay-specific cut-offs (Plasma: GFAP = 30 pg/mL; UCH-L1 = 360 pg/mL. Whole blood: GFAP = 65 pg/mL; UCH-L1 = 360 pg/mL). GFAP and UCH-L1 measured in whole blood were linear across 47–10,000 pg/mL (slope = 0.97, y-intercept = 4 pg/mL) and 87–3,200 pg/mL (slope = 0.98, y-intercept = -6.1 pg/mL), respectively. Within-run CVs for GFAP were 8.8% (mean = 169.2 pg/mL) and 3.3% (mean = 4,746.7 pg/mL); for UCH-L1, 7.6% (mean = 451.5 pg/mL) and 6.1% (mean = 1,415.0 pg/mL). Within-lab CVs for GFAP were 5.1% (mean = 164.4 pg/mL) and 4.0% (mean = 4,486.8 pg/mL); for UCH-L1, 5.8% (mean = 453.7 pg/mL) and 4.6% (mean = 1,389.5 pg/mL). Overall qualitative agreement between whole blood and plasma TBI tests was 93.5% (100% negative, 88.5% positive). Three specimens were elevated for GFAP in plasma (=30 pg/mL) but negative in whole blood (=65 pg/mL); all had negative head CTs, indicating possible higher specificity with whole blood due to its higher cutoff. Quantitative comparison showed GFAP in whole blood was 8.3% higher; UCH-L1 showed no bias (0.1%, slope = 1.05, y-intercept = -36.3 pg/mL), supporting a higher cutoff for GFAP in whole blood. GFAP and UCH-L1 demonstrated strong agreement between whole blood and plasma assay supporting the use of whole blood testing to expedite turnaround time.