Early Prediction of Neurological Outcome After Cardiac Arrest-Rationale and Design of the Prospective International Observational EARLY-NEURO, a STEPCARE Substudy

Abstract

Background: Guidelines discourage prediction of neurological outcome in comatose patients within the first 72 h after cardiac arrest. Increasing evidence suggests that patients with the most severe brain injury and those with no or minimal brain injury may be identified before 72 h using novel methods. We present a protocol for the EARLY-NEURO study, which aims to evaluate whether good and poor outcomes can be reliably predicted already from 24 h after cardiac arrest using the most commonly available methods. Methods: Protocol for a prospective international multicenter substudy within the Sedation, TEmperature and Pressure after Cardiac Arrest and REsuscitation (STEPCARE) trial where adults post-arrest are randomized to minimal or deep sedation, fever treatment with or without a temperature management device and to two different targets of mean arterial blood pressure. Patients sedated or still unconscious at 24 h are examined with head computed tomography (CT) and electroencephalogram (EEG). Blood samples are collected at 24 h after randomization, and stored for analysis of the brain injury marker neurofilament light. CT and EEG examinations will be centrally evaluated for signs of a likely poor or good outcome applying standardized criteria by raters blinded to treatment allocations and patient outcomes. Intensive care treatment, neurological prognostication, and criteria for withdrawal of care will be according to the STEPCARE protocol. Timepoint and reasons for withdrawal of life-sustaining therapy (WLST) will be recorded. WLST prior to 72 h after randomization based on a presumed futile neurological prognosis is strongly discouraged. Primary outcome will be good or poor functional outcome, assessed by the modified Rankin Scale (dichotomized as 0–3 versus 4–6) at 6 months. Results will be reported in accordance with the Standards for Reporting Diagnostic Accuracy (STARD). Conclusions: Earlier prognostication aims to balance the avoidance of premature treatment withdrawal in patients with favorable potential against the prevention of unnecessary intervention in patients with a definitely poor prognosis.