Usefulness of early plasma S-100B protein and Neuron-Specific Enolase measurements to identify cerebrovascular etiology of out-of-hospital cardiac arrest.

Abstract

While S-100B protein and Neuron-Specific Enolase (NSE) dosages have been extensively investigated for neurological prognostication after cardiac arrest (CA), there is no data about their ability to detect a cerebrovascular cause of CA. We assessed the utility of plasma S-100B protein and NSE measurements for early diagnosis of primary neurological cause in resuscitated CA patients. Case control study based on two prospectively acquired CA databases. Patients with a primary cerebrovascular etiology were compared with randomly selected CA of non-neurological cause. S-100B protein and NSE were measured at ICU admission in all patients. CA was due to a cerebrovascular etiology in 18 patients (subarachnoid hemorrhage, n = 15; ischemic stroke, n = 3), with an ICU mortality of 100%. Comparative group was constituted with 66 patients (cardiac etiology n = 45, respiratory etiology n = 21), with an ICU mortality of 71%. Admission S-100B protein concentration was 2.0 [0.63-7.15] μg/L in the cerebrovascular group and 0.45 [0.24-1.95] in the non-cerebrovascular group (p < 0.001). In contrast, NSE concentration was similar in cerebrovascular and non-cerebrovascular etiologies (35 [25-103] μg/L vs. 27 [19-47] respectively, p = 0.16). Area under ROC curves for S-100B protein and NSE to predict cerebrovascular cause of CA was 0.75 [95% CI: 0.64-0.87] and 0.61 [95% CI: 0.45-0.76], respectively. Even if S-100B protein dosage performs slightly better than NSE, early dosages of these biomarkers are poorly predictive of a cerebrovascular etiology of CA. Our results suggest that early measurement of brain biomarkers should not be recommended to tailor the imaging strategy employed to investigate the CA cause.