Introduction: Neuron specific enolase (NSE) is a marker of brain injury that is predictive of poor outcome in cardiac arrest patients when greater than 33 µL/ml with a 0% false positive rate1,2. The validity of NSE prognosis in cardiac arrest patients treated with hypothermia has not been validated1. We sought to validate NSE values in predicting clinical outcome in patients with cardiac arrest who received therapeutic hypothermia.
Methods: The medical records of all patients who had cardiac arrest and received therapeutic hypothermia at Mayo Clinic Hospital from July 2006 to August 2009 were retrospectively reviewed. NSE and simultaneous glomerular filtration rate (GFR) values were obtained between days 0-3 post cardiac arrest per American Academy of Neurology guidelines. We measured outcome by Glasgow Pittsburgh Cerebral Performance Category (CPC) by death or by hospital discharge; with good outcome defined as CPC 1-2, and poor outcomes CPC 3 to 5.
Results: During a two year period, 41 patients received therapeutic hypothermia for coma after cardiac arrest. Nine patients did not have recorded NSE values and were excluded from analysis. Of the remaining 32 patients who received therapeutic hypothermia post-cardiac arrest, the initial rhythm found at the arrest included 9 patients with asystole, 10 with pulseless electrical activity (PEA), 8 with ventricular fibrillation, 4 with bradycardic arrest, and 1 from ventricular tachycardia. 11 patients had NSE values greater than 33 µL/ml, 10 of which (10/11, 91%) were associated with poor outcome (Table 1). However, one patient with a NSE value greater than 33 had a good outcome. Markedly high NSE values of 256 and 525 were seen in patients with a glomerular filtration rate (GFR) between 34 and 36 ml/min/m2, suggesting that renal failure may give falsely elevated NSE values. 15 patients had NSE values obtained in less than 24hrs after cardiac arrest, 11 of which had poor outcome (27%, 3/11 had NSE >33).
Conclusion: NSE values greater than 33 µL/ml in hypothermic post-cardiac arrest patients were associated with a 91% chance of poor outcome and a false positive rate (FPR) of 9%. Thus, NSE appears falsely elevated in renal failure (GFR <40). Clinicians should be aware that NSE values greater than 33 µL/ml may occur when GFR is less than 36 ml/min/m2 and may be invalid for prognostication purposes. In those cases, other means of prognostic testing in hypothermic patients after cardiac arrest is suggested for more accurate neurologic prognostication.