The FOUR score predicts outcome in patients after cardiac arrest

Background: Prognostication for survivors of cardiac arrest is a frequent challenge to neurologists. Our aim was to determine whether the FOUR (Full Outline of UnResponsiveness) score is an accurate predictor of outcome in patients after cardiac arrest and to compare its performance to the Glasgow Coma Scale (GCS). Methods: We prospectively identified patients surviving cardiac arrest from June 2006 to October 2009. Neurologic exams were grouped into two time intervals following cardiac arrest: 1-2 days and 3-5 days. The FOUR score and the Glasgow coma scale (GCS) were determined for each examination. Primary outcome was in-hospital mortality. Results: Of 136 patients, 112 (82%) were examined on days 1-2 after cardiac arrest and 87 (64%) on days 3-5. Forty-seven patients (35%) survived to hospital discharge and 89 (65%) died during hospitalization. No patients with a sum FOUR score 4 at exam days 3-5 survived (false positive rate [FPR] 0% C.I. 0.000-0.0345), whereas one patient (2%) with sum GCS score of 3 survived to discharge (FPR 2.2%, C.I. <0.0001-0.1758). At days 3-5 after arrest, 41 of 45 (91%) patients with a sum FOUR score >8 survived (P < 0.0001), while 39 of 45 (87%) with a sum GCS > 6 survived (P < 0.0001). A 2-point improvement in FOUR score, but not GCS, in serial exams was associated with survival. Sensitivities, specificities, positive, and negative predictive values were comparable between both scales. Conclusion: The FOUR score, a simple clinical tool, is an accurate predictor of outcome in patients surviving cardiac arrest.