Neurofilament Light Outperforms Traditional Biomarkers in Post-Cardiac Arrest Prognosis: TTM2 Study Finds

With approximately 4 million global out-of-hospital cardiac arrests occurring annually and high mortality rates, traditional neuroprognostication has long relied on clinical examinations and markers like neuron-specific enolase (NSE) that are limited by factors like hemolysis; thus, Dr. Marion Moseby-Knappe from Lund University aimed to evaluate which brain injury biomarker most accurately predicts functional outcomes using automated assays suitable for immediate clinical implementation.

Therefore, the prospective international multicenter observational investigation was conducted within the Targeted Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest (TTM2) trial, enrolling 819 unconscious adult participants from 24 European intensive care units. Serial blood samples were analyzed at 0, 24, 48, and 72 hours post-admission using automated electrochemiluminescence immunoassays (ECLIA) to evaluate NSE, S100, NfL, and glial fibrillary acidic protein (GFAP) against 6-month functional outcomes while excluding those with insufficient data.

Key Clinical Findings of the Study Include:

• Superior Predictive Accuracy: NfL reached a significantly higher AUROC of 0.93 at 48 hours compared to 0.85 for NSE and 0.87 for GFAP, establishing it as the most precise biomarker for evaluating long-term neurological recovery.

• Increased Clinical Sensitivity: At a stringent 98% specificity, NfL achieved 60% sensitivity at 24 hours, which is 25% to 44% higher than sensitivities found using routine clinical methods like electroencephalogram (EEG) or computed tomography (CT) scans.

• Mortality Correlation: Patients with serum NfL concentrations in the highest tercile at 24 hours faced a 43.2-fold higher hazard of death within 180 days compared to the lowest tercile, where survival was 94%.

• Subgroup Reliability: The prognostic accuracy of NfL remained remarkably robust across diverse patient subgroups regardless of age, sex, or initial cardiac rhythm, and it was not susceptible to interference from common clinical confounders like hemolysis.

The results suggest that neurofilament light is a highly accurate predictor of long-term outcomes after cardiac arrest, with 92% of patients correctly classified based on their functional status as early as 24 hours post-admission.

Thus the study concludes that clinicians may find that integrating these automated neurofilament light assays into multimodal prognostic protocols can help individualize patient care and improve family communication regarding potential neurological recovery.

Although the study was limited by the reliance on local markers at some sites and a mostly cardiac-specific population, future research is recommended to establish standardized cutoff values across diverse global populations and different causes of arrest.

Reference

Moseby-Knappe M, Levin H, Ullén S, et al. Blood biomarkers for the prediction of outcome after cardiac arrest: an international prospective observational study within the Targeted Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest (TTM2) trial. Lancet Respir Med 2026; 14: 215–24.