Blood Biomarkers for Prediction of Outcome After Cardiac Arrest

A study published in The Lancet Respiratory Medicine has identified neurofilament light (NFL) as the most accurate blood biomarker for predicting long-term neurological outcomes in patients who remain unconscious following cardiac arrest. This research, conducted within the Targeted Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest (TTM2) trial, represents the largest prospective international study of brain injury biomarkers in cardiac arrest patients to date.

The study analysed blood samples from 819 unconscious adult patients admitted to intensive care units across 24 European hospitals between April 2018 and January 2020. The mean age was 64 years, and 81% of participants were male. All patients had experienced out-of-hospital cardiac arrest from presumed cardiac or unknown causes. Blood samples were collected at admission (0 hours) and at 24, 48, and 72 hours after ICU admission.

The primary outcome was functional recovery at six months, classified as either good. Approximately half the patients (51%) experienced poor outcomes, while 49% achieved good functional recovery.

You might also like: ESAIC Guidelines for Use of Cardiac Biomarkers in Perioperative Risk Evaluation

Neurofilament light demonstrated exceptional accuracy in predicting functional outcomes, with area under the receiver operating characteristic curve (AUROC) values of 0.77 at admission, rising to 0.92 at 24 hours and remaining at 0.93 at both 48 and 72 hours. These results significantly outperformed the other three biomarkers evaluated: glial fibrillary acidic protein (GFAP), neuron-specific enolase (NSE), and S100.

Neurofilament light showed statistically significant superiority over GFAP, the second-best performing marker, at all timepoints from 24 hours onwards. At 24 hours, neurofilament light showed 60% sensitivity and 98% specificity, correctly identifying overall functional outcomes in 92% of patients. When normal neurofilament light concentrations were present, they predicted good outcomes with 93–96% negative predictive value.

Current guidelines recommend neuron-specific enolase as the primary biomarker for neuroprognostication after cardiac arrest, despite known limitations, including false elevations from haemolysis, commonly occurring in patients requiring mechanical circulatory support or renal replacement therapy. The study confirmed that neurofilament light substantially outperformed neuron-specific enolase at all time points.

Neurofilament light demonstrated 25–44% higher sensitivity than routine prognostic methods, including CT imaging, electroencephalography, and brainstem reflex assessments, when matched for specificity. This translated to identifying 60–110 additional patients with poor outcomes compared to conventional methods. Critically, adding routine prognostic assessments to neurofilament light measurements between 24 and 72 hours did not enhance predictive accuracy, suggesting neurofilament light alone provides comprehensive prognostic information during this period.

The study revealed a correlation between neurofilament light concentrations and survival probability. Patients with low neurofilament light concentrations at 24 hours had 94% survival and good functional outcomes at 6 months. Only 9% of patients with high concentrations survived to six months.

Overall, this study establishes neurofilament light as a highly accurate, clinically feasible biomarker for predicting neurological outcomes after cardiac arrest, superior to currently recommended markers. The standardisation of cutoff values across different assay platforms and prospective validation in diverse patient populations remain important next steps towards full clinical implementation. Nevertheless, these findings suggest neurofilament light could soon become integral to multimodal prognostication strategies, supporting clinicians in tailoring treatment and informing families about likely outcomes following cardiac arrest.

Source: The Lancet

Image Credit: iStock