目的 评估不同昏迷量表和中央区95%频谱边缘频率（SEF）预测急性昏迷患者近期死亡的价值。方法  研究对象为2014年8月至2016年10月南通大学附属医院神经内科重症监护病房收治的52例急性昏迷患者。在患者发病72 h内进行格拉斯哥昏迷量表（GCS）、格拉斯哥匹兹堡昏迷量表（GCS-P）、全面无反应性量表（FOUR）评分，同时行脑电图（EEG）监测，记录中央区95% SEF数据。随访1个月，患者分为生存组和死亡组，比较两组年龄、性别、既往史、GCS评分、GCS-P评分、FOUR评分和中央区95%SEF，采用单因素和多因素Logistic回归分析影响近期死亡的相关因素。采用受试者工作特征（ROC）曲线比较GCS评分、GCS-P评分、FOUR评分和中央区95%SEF对近期死亡的预测价值。采用McNemar χ²检验对GCS评分、GCSP评分、FOUR评分联合中央区95%SEF与单独使用上述评分在预测近期死亡的敏感性和特异性方面进行比较。结果 52例患者中，生存组39例，死亡组13例。与生存组相比，死亡组的GCS评分、GCS-P评分、FOUR评分以及中央区95%SEF均明显降低（P<0.05, P<0.01，P<0.01和P<0.01）。多因素Logistic回归分析显示，GCS评分、GCS-P评分、FOUR评分以及中央区95%SEF均为近期死亡的独立预测因素。ROC曲线分析显示，GCS评分、GCS-P评分、FOUR评分以及中央区95%SEF对近期死亡均具有中等预测价值。与单独用昏迷量表预测相比，联合GCS评分、GCS-P评分、FOUR评分和中央区95%SEF预测在敏感性上差异无显著性（均P>0.05），而特异性明显提高（均P<0.05）。结论G  CS评分、GCS-P评分、FOUR评分以及中央区95%SEF均可用于急性昏迷患者近期死亡的预测，联合昏迷量表和中央区95%SEF能更有效地预测急性昏迷患者的近期死亡。

Objective  To evaluate the value of different coma scales and 95% spectral edge frequency(SEF) in the central area in predicting the recent death in patients with acute coma. Methods  Fifty-two patients with acute coma admitted in the neurointensive care unit in Affiliated Hospital of Nantong University from August 2014 to October 2016 were included. Glasgow coma scale (GCS), Glasgow-Pittsburgh coma scale (GCS-P), and full outline of unresponsiveness score (FOUR) were performed within 72 hours of the onset of the patients. EEG monitoring was performed and the data of 95% SEF in the central area were recorded simultaneously. The patients were divided into two groups: survival group and death group according to one-month followup. The age, sex, previous history, GCS score, GCS-P score, FOUR score and 95% SEF in the central area were compared between two groups. Single factor and multivariate logistic regression analysis were used to investigate relevant factors related to the recent death. The predictive value of GCS score, GCS-P score, FOUR score and 95% SEF in the central area were compared with the receiver operating characteristic (ROC) curve. The GCS score, the GCS-P score, and the FOUR score combined with the 95% SEF in the central area were compared with the above scores used alone by McNemarχ2 test in the sensitivity and specificity of predicting the recent death. Results   Of the 52 patients, 39 of them were in the survival group and 13 of them were in the death group. Compared with the survival group, the GCS score, GCS-P score, FOUR score and 95% SEF in the central area were significantly decreased in the death group (P<0.05, P<0.01, P<0.01, P<0.01, respectively). Multivariate logistic regression analysis showed that GCS score, GCS-P score, FOUR score and 95% SEF in the central area were independent predictors of one-month death. ROC curve analysis showed that GCS score, GCS-P score, FOUR score and 95% SEF in the central area had moderate predictive value for one-month death. There was no significant difference in sensitivities between the GCS score, the GCS-P score, the FOUR score combined with the 95% SEF in the central area vs the above scores used alone (P>0.05, P>0.05, P>0.05, respectively), but the specificities were significantly higher (P<0.05, P<0.05, P<0.05, respectively). Conclusion  GCS score, GCS-P score, FOUR score, and 95% SEF in the central area can be used to predict the recent death of patients with acute coma. Coma scales combined with 95% SEF in the central area are more effective in predicting the recent death of patients with acute coma.