Prediction of Neurological Recovery After Cardiac Arrest Using Neurofilament Light Chain is Improved by a Proteomics-Based Multimarker Panel

Background: Continuous advances in resuscitation care have increased survival, but the rate of favorable neurological outcome remains low. We have shown the usefulness of proteomics in identifying novel biomarkers to predict neurological outcome. Neurofilament light chain (NfL), a marker of axonal damage, has since emerged as a promising single marker. The aim of this study was to assess the predictive value of NfL in comparison with and in addition to our established model.

Methods: NfL was measured in plasma samples drawn at 48 h after cardiac arrest using single-molecule assays. Neurological function was recorded on the cerebral performance category (CPC) scale at discharge from the intensive care unit and after 6 months. The ability to predict a dichotomized outcome (CPC 1-2 vs. 3-5) was assessed with receiver operating characteristic (ROC) curves.

Results: Seventy patients were included in this analysis, of whom 21 (30%) showed a favorable outcome (CPC 1-2), compared with 49 (70%) with an unfavorable outcome (CPC 3-5) at discharge. NfL increased from CPC 1 to 5 (16.5 pg/ml to 641 pg/ml, p < 0.001). The addition of NfL to the existing model improved it significantly (Wald test, p < 0.001), and the combination of NfL with a multimarker model showed high areas under the ROC curve (89.7% [95% confidence interval 81.7-97.7] at discharge and 93.7% [88.2-99.2] at 6 months) that were significantly greater than each model alone.

Conclusions: The combination of NfL with other plasma and clinical markers is superior to that of either model alone and achieves high areas under the ROC curve in this relatively small sample.