Background and purpose: The brain-specific astroglial protein GFAP is a blood biomarker candidate indicative of intracerebral hemorrhage in patients with symptoms suspicious of acute stroke. Comparably little, however, is known about GFAP release in other neurological disorders. In order to identify potential "specificity gaps" of a future GFAP test used to diagnose intracerebral hemorrhage, we measured GFAP in the blood of a large and rather unselected collective of patients with neurological diseases.
Methods: Within a one-year period, we randomly selected in-patients of our university hospital for study inclusion. Patients with ischemic stroke, transient ischemic attack and intracerebral hemorrhage were excluded. Primary endpoint was the ICD-10 coded diagnosis reached at discharge. During hospital stay, blood was collected, and GFAP plasma levels were determined using an advanced prototype immunoassay at Roche Diagnostics.
Results: A total of 331 patients were included, covering a broad spectrum of neurological diseases. GFAP levels were low in the vast majority of patients, with 98.5% of cases lying below the cut-off that was previously defined for the differentiation of intracerebral hemorrhage and ischemic stroke. No diagnosis or group of diagnoses was identified that showed consistently increased GFAP values. No association with age and sex was found.
Conclusion: Most acute and chronic neurological diseases, including typical stroke mimics, are not associated with detectable GFAP levels in the bloodstream. Our findings underline the hypothesis that rapid astroglial destruction as in acute intracerebral hemorrhage is mandatory for GFAP increase. A future GFAP blood test applied to identify patients with intracerebral hemorrhage is likely to have a high specificity.