Correlation of levels of neuronal and glial markers with radiological measures of infarct volume in ischaemic stroke: a systematic review

Abstract

Background: A blood test that quantified the extent of brain damage following ischaemic stroke might be a useful surrogate outcome measure in trials of acute stroke treatments. Measures of neuronal and glial damage, such as neuron-specific enolase (NSE), glial fibrillary acidic protein, tau-protein, myelin-basic protein and S100-β are potential candidate biomarkers.

Aim: We systematically reviewed the relevant literature to find studies that correlated blood levels of neuronal and glial damage markers with imaging measures of infarct volume.

Methods: We identified studies with a comprehensive search of databases and the reference lists of relevant studies. We included studies that: (1) measured the highest level, or area under the curve (AUC) over time of markers of cerebral damage, (2) calculated infarct volume, and (3) correlated the two measures.

Results: Seventeen studies met the criteria for the systematic review. There were sufficient data to provide summary estimates for S100-β and NSE. The peak level and AUC over time of both markers correlated with subacute infarct volume. Measurements of S100-β later than 24 h after stroke were better correlated with subacute infarct size than earlier measurements. However, scan times varied, and none was later than 8 days after stroke.

Conclusion: Peak and AUC levels of NSE and S100-β levels correlated with subacute infarct volume. Correlations of S100-β with infarct volume were stronger when measured after 24 h than closer to admission. Exploratory studies within clinical trials are necessary before blood markers of cerebral tissue damage can be recommended as surrogate endpoints.