INTRODUCTION: Early and precise prognosis determination in cardiac arrest survivors remains challenging despite multimodal approach. The aim of our study was to compare prognostic values of NSE with novel biomarkers serum tau protein (Tau), neuro-filament light chain (Nfl) and Galectin-3 (Gal).
METHODS: Eligible subjects were out-of-hospital cardiac arrest survivors. Blood samples for the measurements of NSE, Tau, Nfl and Gal levels were drawn at 24 hrs (D1), 48 hrs (D2), 72 hrs (D3), and 96 hrs (D4) after hospital admission. Thirty-day neurological outcomes according to the Modified Rankin Scale (mRS) were evaluated as clinical endpoints, poor outcome was defined as mRS 4-6. Prognostic values of NSE, Tau, Nfl and Gal for the prediction of poor outcomes were determined using ROC analysis.
RESULTS: A total of 43 patients were enrolled in the present study. The comparison of ROC curves revealed significantly lower area under the curve (AUC) for Gal in comparison to other biomarkers at D2-4. Numerically, the highest AUC at D1 was observed for Tau and at D2, D3 and D4 for NSE. The highest sensitivity for the prediction of poor prognosis with 100% specificity was detected for Tau values at D1 (33.3%) or D2 (70.0%) and for NSE values at D3 (92.9%) or D4 (100%). Multiple logistic regression revealed that combination of all four biomarkers may predict poor prognosis already at D1 with 100% specificity and 53% sensitivity (AUC 0.888, P˂0.001).
CONCLUSIONS: Our results indicate that the novel biomarkers Tau and Nfl have comparable predictive value for clinical outcomes as NSE at 48 to 96 hrs after cardiac arrest. Gal values predict outcomes only at 24 hours. Combination of all four biomarkers could improve prognosis prediction with high predictive value already the first day.