长程脑电图监测对评估昏迷患者预后的应用价值

Application of long-time EEG monitoring in comatose patients

目的探讨长程脑电图监测和临床检测指标对评估昏迷患者预后的应用价值。方法对26例昏迷患者(男15例,女11例,年龄17～90岁)急性期进行长程脑电图监测、改良 Glasgow 昏迷评分以及半球复杂性的分析,随访患者至昏迷后2个月.并进行 Glasgow 预后评分,将长程脑电图监测和临床检测指标结果和昏迷患者预后进行相关分析;观察昏迷患者半球复杂性随时间的变化及其与预后的关系。结果长程脑电图监测对昏迷患者预后评估的特异度(77.8%)、准确度(92.3%)均较临床检测指标改良 Glasgow 昏迷评分高。脑电图分级与 Glasgow 预后评分进行线性相关分析,相关系数r=0.81,P<0.01;改良 Glasgow 昏迷评分和 Glasgow 预后评分进行线性相关分析,相关系数r=-0.39,P<0.01。患者病情变化时,脑电图改变出现在改良 Glasgow 昏迷评分改变之前。所观察的全部26例昏迷患者其左、右半球复杂性的熵值差异均无统计学意义,熵值随时间的变化呈6种图形,其中4种图形的昏迷患者 Glasgow 预后评分指标不良;2种图形的患者预后指标良好。结论对昏迷患者进行长程脑电图监测,在评价脑功能和预测昏迷患者的预后两方面有较好的应用价值。

Objective To explore the prognostic value of long-time EEG monitoring and the clinical examination index in comatose patients. Methods The long-time EEG monitoring and the modified Glasgow coma scale were done in 26 comatose patients（ 15 patients were male; 11 patients were female; 17-40 years old） in acute phase. The complexity of cerebral hemisphere was analyzed by computer. The patients were followed-up for 2 months and evaluated by Glasgow outcome scale. The correlation between the long-time EEG monitoring, clinical examination index and the prognosis of the patients was analyzed. The time-variation of hemisphere complexity and the relationship between prognosis and entropy were observed. Results The specificity （77.8%） and accuracy rate （92. 3% ） of long-time EEG monitoring in predicting the outcome of the acute comatose patients were higher than those of clinical examination. The EEG index appeared more closed-up with prognosis than clinical examination. The correlation coefficient between the EEG index and the modified Glasgow prognosis scoring was r = 0. 81 （ P 〈 0. 01 ）. The correlation coefficient between the modified Glasgow coma scale and the modified Glasgow prognosis scoring was r = - 0. 39 （ P 〈 0. 01 ）. The change of EEG index appeared earlier than that of clinical examination. Statistically, for the coma patients, there was no difference between the right hemisphere complexity of and the left one. There were 6 kinds of entropy time-variation graphs. The patients whose graphs were in 4 kinds of the graphs had better outcome, and others had the opposite outcome. Conclusions Long-time EEG monitoring might be of certain value in evaluating brain function and predicting the outcome of acute comatose patients. And it might be of some value in clinical therapy.