Comparison of the accuracy of predicting prognosis of brain function in patients after cardiopulmonary cerebral resuscitation with two kinds of electroencephalogram techniques combined with neuron-specific enolase

Abstract

Objective: To compare the accuracy of electroencephalography (EEG) grading or amplitude-integrated electroencephalography (aEEG) grading combined with NSE in predicting brain function prognosis after cardiopulmonary cerebral resuscitation (CPR) in adults. Methods: The patients who were admitted to Fujian Medical University Union Hospital after CPR from January 2015 to June 2019 were enrolled. Demographic data, Glasgow coma scale (GCS), blood neuron specific enolase (NSE), EEG grading and aEEG grading were collected. The main clinical outcome was the prognosis of brain function (Glasgow-Pittsburgh cerebral performance category, CPC) in patients at 3 months after CPR. Accordingly, the patients were divided into two groups: favorable prognosis group and poor prognosis group, and relevant parameters were compared between the two groups. The predictive ability of EEG grading or aEEG grading combined with NSE for brain function prognosis was evaluated by receiver operating characteristic (ROC) curve. Results: A total of 57 patients were enrolled, with 34 males and 23 females. The average age was (65±19) years old. In terms of Young EEG scales, there was 16 grade 1 cases (28.1%), 24 grade 2-5 cases (42.1%) and 17 grade 6 cases (29.8%), respectively. As for aEEG grading, there was 11 grade Ⅰ cases (19.3%), 25 grade Ⅱ cases (43.9%) and 21 grade Ⅲ cases (36.8%), respectively. There was no significant difference of age, sex, length of stay between the two groups (all P>0.05). However, there was significant difference of EEG grading scale, aEEG grading, GCS grading and NSE between the two groups (all P<0.05). The area under curve (AUC) of NSE, EEG grading and aEEG grading for predicting brain function prognosis was 0.81, 0.82 and 0.85, respectively (all P<0.01). In aEEG grading combined with NSE group, the AUC of was 0.92, and the optimal cut-off point was 4.5, with a sensitivity of 95.8% and a specificity of 79.0%. In EEG grading combined with NSE group, the AUC was 0.90, and the optimal cut-off point was 3.6, with a sensitivity of 92.1% and a specificity of 77.0%. Conclusions: aEEG grading combined with NSE is more accurate in predicting prognosis in patients with cardiopulmonary cerebral resuscitation when compared to EEG grading. Considering its feasibility, aEEG grading combined with NSE is more suitable for clinical application.