Implementation of locally processed neuron-specific enolase is associated with improved guideline-concordant neuroprognostication after out-of-hospital cardiac arrest: a retrospective before-and-after study.

BACKGROUND: For comatose survivors of out-of-hospital cardiac arrest (OHCA), accurate neuroprognostication enables shared decision-making and avoids premature or unnecessarily delayed withdrawal of life-sustaining treatment. Guidelines recommend a multimodal approach requiring at least two independent poor-prognostic criteria, of which neuron-specific enolase (NSE) is currently the primary endorsed serum biomarker. NSE is not consistently used, and guideline adherence remains variable internationally. OBJECTIVES: To evaluate whether locally processed NSE improved guideline-concordant neuroprognostication and decision-making timelines after OHCA. METHODS: A single-centre retrospective before-and-after study at an Australian quaternary intensive care unit (ICU) compared control (January 2022-July 2024) and intervention (August 2024-March 2026) periods. Sixty-nine comatose adults who underwent withdrawal of life-sustaining treatment on neurological grounds after OHCA were included. The primary outcome was guideline-concordant neuroprognostication (≥2 multimodal poor-prognostic criteria). Secondary outcomes were time to clinical determination of a poor neurological prognosis and prognostic disclosure to the family. RESULTS: Guideline-concordant neuroprognostication increased from 35.9% (14/39) to 63.3% (19/30), an absolute increase of 27.4% (p = 0.030). Cumulative sum analysis showed sustained improvement throughout the intervention period. Time to clinical determination of a poor neurological prognosis (87.0 vs 88.4 h, p = 0.66) and prognostic disclosure to the family (91.6 vs 92.4 h, p = 0.69) were unchanged. CONCLUSIONS: Implementation of locally processed NSE was associated with a statistically significant increase in guideline-concordant multimodal neuroprognostication. Decision-making timelines were unchanged, consistent with prognostication being shaped by human and system-level factors beyond diagnostic availability. NSE availability was not associated with earlier withdrawal of life-sustaining treatment.