Prehospital GFAP Test Detected 32 of 33 Brain Bleeds in Nontraumatic Coma

A prehospital whole-blood point-of-care test for glial fibrillary acidic protein (GFAP) detected 32 of 33 intracranial hemorrhages (ICH) among adults in acute nontraumatic coma, according to a diagnostic study published in JAMA Network Open.

The study used data from East Anglian Air Ambulance, a UK physician-paramedic helicopter emergency medical service that measured GFAP and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1) with the i-STAT Alinity point-of-care blood analyzer (Abbott Laboratories), with results available within 15 minutes. According to the authors, the FDA authorized a whole-blood point-of-care combined GFAP and UCH-L1 assay in 2024, but its prehospital performance was uncertain.

The analysis included 65 consecutive adults with a Glasgow Coma Scale score of 12 or less who underwent prehospital rapid sequence induction of anesthesia with clinical suspicion of ICH. Computed tomography, the reference standard, showed ICH in 33 patients (50.8%). At the manufacturer's 65 pg/mL threshold, GFAP had a sensitivity of 97.0% (95% CI, 84.2%-99.9%) and a specificity of 53.1%, with an area under the receiver operating characteristic curve of 0.94.

UCH-L1 performed less well, with a sensitivity of 78.8% and a specificity of 18.8%. Counting a positive result on either biomarker as positive produced a sensitivity and negative predictive value of 100%, although specificity was 15.6% and the confidence interval for the negative predictive value was wide (47.8%-100%). The single GFAP false negative occurred in a patient with isolated subarachnoid hemorrhage, which the authors said suggests GFAP may be less sensitive for that form of bleeding.

The authors cautioned that low specificity would produce substantial overtriage if biomarker testing were used alone. They cited limitations including a single physician-led helicopter service, a highly selected cohort with a high ICH prevalence, and a GFAP cutoff predetermined by the manufacturer and optimized for mild traumatic brain injury. Further prospective external validation is needed to determine the test's association with triage, treatment decisions, and patient outcomes, they concluded.

Source: Price J, Lachowycz K, Palti G, et al. Prehospital point-of-care GFAP for detection of intracranial hemorrhage in acute nontraumatic coma. JAMA Netw Open. 2026;9(10):e2639337. doi:10.1001/jamanetworkopen.2026.39337