When combined with clinical scores, a "game-changing" blood test can expedite the diagnosis and treatment of large vessel occlusion (LVO) stroke, potentially saving many lives, new data suggested.
Using cutoff levels of two blood biomarkers, glial fibrillary acidic protein (GFAP; 213 pg/mL) and D-dimer (600 ng/mL), and the field assessment stroke triage for emergency destination (FAST-ED) (score, > 2), investigators were able to detect LVOs with 81% sensitivity and 93% specificity less than 6 hours from the onset of symptoms.
GFAP has previously been linked to brain bleeds and traumatic brain injury.
The test also ruled out all patients with brain bleeds, and investigators noted that it could also be used to detect intracerebral hemorrhage.
"We have developed a game-changing, accessible tool that could help ensure that more people suffering from stroke are in the right place at the right time to receive critical, life-restoring care," senior author Joshua Bernstock, MD, PhD, MPH, a clinical fellow in the Department of Neurosurgery at Brigham and Women's Hospital in Boston, said in a press release.
The findings were published online on May 17 in Stroke: Vascular and Interventional Neurology.
Early Identification Crucial
Acute LVO stroke is one of the most treatable stroke types because of the availability of endovascular thrombectomy (EVT). However, EVT requires specialized equipment and teams that represent a small subset of accredited stroke centers and an even smaller subset of emergency medical facilities, so early identification of LVO is crucial, the investigators noted.
Bernstock and his team developed the TIME trial to assess the sensitivity and specificity of the blood biomarkers and scale cutoff values for identifying LVO vs non-LVO stroke.
As part of the observational prospective cohort trial, investigators included consecutive patients admitted to the Brandon Regional Hospital Emergency Department in Brandon, Florida, between May 2021 and August 2022 if they were referred for a suspected stroke and the time from symptom onset was under 18 hours.
Patients were excluded if they received thrombolytic therapy before blood was collected or if it was anticipated that blood collection would be difficult.
Investigators gathered information on patients' clinical data, hematology results, time since last known well, and imaging findings to construct a clinical diagnosis (LVO, non-LVO, ischemic stroke, hemorrhagic stroke, or transient ischemic attack [TIA]).
In addition to the National Institutes of Health Stroke Scale, patients were assessed with the FAST-ED, the Rapid Arterial oCclusion Evaluation (RACE), the Cincinnati Stroke Triage Assessment Tool, and the Emergency Medical Stroke Assessment.
Of 323 patients in the final study sample, 29 (9%) had LVO ischemic stroke, and 48 (15%) had non-LVO ischemic stroke. Another 13 (4%) had hemorrhagic stroke, 12 had TIA (3.7%), and the largest proportion of patients had stroke mimic (n = 220; 68%), which included encephalopathy, hyperglycemia, hypertensive emergency, migraine, posterior reversible encephalopathy syndrome, and undetermined.
The Case for Biomarkers
When investigators looked at those with LVO ischemic stroke, they found the concentration of plasma D-dimer was significantly higher than that in patients with non-LVO suspected stroke (LVO suspected stroke, 1213 ng/mL; interquartile range [IQR], 733-1609 vs non-LVO suspected stroke, 617 ng/mL; IQR, 377-1345; P < .001).
In addition, GFAP was significantly increased in the plasma of patients with hemorrhagic stroke vs all other patients with suspected stroke (hemorrhagic stroke, 1464 pg/mL; IQR, 292-2580 vs nonhemorrhagic suspected stroke, 48 pg/mL; IQR, 12-98; P < .005).
Combinations of the blood biomarkers with the scales FAST-ED or RACE showed the best performance for LVO detection, with a specificity of 94% (for either scale combination) and a sensitivity of 71% for both scales.
When investigators analyzed data for just those patients identified within 6 hours of symptom onset, the combination of biomarkers plus FAST-ED resulted in a specificity of 93% and a sensitivity of 81%.
Given that clinical stroke scales in patients with hemorrhagic stroke frequently suggest LVO and that these patients are not candidates for EVT, a tool capable of ruling out hemorrhage and identifying only nonhemorrhagic ischemic LVO is essential, the investigators noted.
"In stroke care, time is brain," Bernstock said. "The sooner a patient is put on the right care pathway, the better they are going to do. Whether that means ruling out bleeds or ruling in something that needs an intervention, being able to do this in a prehospital setting with the technology that we built is going to be truly transformative."
The study was funded by the Innovate UK grant and private funding. Bernstock has positions and equity in Pockit Diagnostics Ltd. and Treovir Inc. and is on the boards of Centile Bio and NeuroX1. Other disclosures are noted in the original article.