定量脑电图对意识障碍患者预后的预测价值

目的 探讨定量脑电图(quantitative dectroencephalography,qEEG)中95%频谱边界频率(95% spectral edge frequency,SEF95)和全功率(total power,TP)对意识障碍患者预后的预测价值.方法 研究对象为2008年1月至2010年6月南方医院神经内科重症监护病房收治的意识障碍患者.患者入院后进行格拉斯哥昏迷量表(Glasgow Coma Scale,GCS)评分,同时行脑电图监测.以患者出重症监护病房时的存活情况分为存活组和死亡组,比较两组年龄、性别、高血压、糖尿病、GCS评分、SEF95和TP,并对上述因素进行多变量logistic回归分析.对预后相关指标进行接受者操作特征(receiver operating characteristic,ROC)曲线分析,明确qEEG对意识障碍患者死亡的预测能力.结果 共纳入109例幕上病变患者,其中存活组79例,死亡组30例.死亡组GCS评分[(5±3)分对(9±3)分,P=0.000]和SEF95(7.0±4.0对10.0±4.0,P=0.002)均显著低于存活组.多变量logistic回归分析显示,GCS评分(优势比0.100,95%可信区间0.029～0.353]和SEF95(优势比0.853,95%可信区间0.740～0.983)为近期预后的独立预测因素.ROC曲线分析显示,GCS评分和SEF95越低,患者死亡的可能性越大.SEF95＜7.75时,判断死亡的敏感性为60.O%,特异性为72.2%,阳性预测值为82.6%,阴性预测值为45.0%;当GCS评分＜8分时,判断死亡的敏感性为83.3%,特异性为73.4%,阳性预测值为82.6%,阴性预测值为45.O%.结论 SEF95有助于意识障碍患者的预后判断,有望成为重症监护病房意识障碍患者床旁预后评价的重要手段.

Abstract：

Objective To investigate the predictive value of prognosis of the 95% spectral edge frequency (SEF95) and total power (TP) in quantitative electroencephalography (qEEG) in patients with disturbance of consciousness. Methods The patients with disturbance of consciousness admitted in the neurointensive care unit (NICU) in Nanfang Hospital from January 2008 to June 2010 were included. Glasgow Coma Scale (GCS) scores were performed on admission and EEG monitoring was performed simultaneously. The patients were divided into either a survival group or a death group according to the survival status of the patients at the time of leaving NICU. The age, sex, hypertension, diabetes, GCS scores, SEF95, and TP were compared between the two groups. A multivariate logistic regression analysis was performed for the above factors. The prognostic indicators were analyzed with the receiver operating characteristic (ROC) curves and the of qEEG predictive ability of death in patients with disturbance of consciousness were determined. Results A total of 109 patients with supratentorial lesions were enrolled in the study, 79 of them were in the survival group and 30 of them were in the death group. The GCS scores (5 ±3vs. 9 ±3, P =0. 000) and SEF95 (7. 0 ±4.0 vs. 10. 0 ±4. 0, P = 0. 002) in the death group were significantly lower than those in the survival group. Multivariate logistic regression analysis showed that GCS scores (odds ratio 0. 100, 95% confidence interval 0. 029-0. 353) and SEF95 (odds ratio 0. 853, 95% confidence interval 0. 740-0. 983) were the independent predictors of recent prognosis. ROC curve analysis showed that the lower the GCS scores and SEF95 were, the greater the likelihood of death in patients. When SEF95 was ＜7. 75, the sensitivity to determine the death was 60. 0%, the specificity was 72. 2%, the positive predictive value was 82. 6%, and the negative predictive value was 45. 0%; when the GCS score was ＜8, the sensitivity to determine the death was 83. 3%, the specificity was 73. 4%,the positive predictive value was 82. 6%, and the negative predictive value was 45. 0%. Conclusions SEF95 helps determine the prognosis of patients with disturbance of consciousness, and it is expected to become an important means of bedside assessment of prognosis in patients with disturbance of consciousness in NICU.

Predictive value of prognosis in patients with disturbance of consciousness using quantitative electroencephalography

Objective To investigate the predictive value of prognosis of the 95% spectral edge frequency (SEF95) and total power (TP) in quantitative electroencephalography (qEEG) in patients with disturbance of consciousness. Methods The patients with disturbance of consciousness admitted in the neurointensive care unit (NICU) in Nanfang Hospital from January 2008 to June 2010 were included. Glasgow Coma Scale (GCS) scores were performed on admission and EEG monitoring was performed simultaneously. The patients were divided into either a survival group or a death group according to the survival status of the patients at the time of leaving NICU. The age, sex, hypertension, diabetes, GCS scores, SEF95, and TP were compared between the two groups. A multivariate logistic regression analysis was performed for the above factors. The prognostic indicators were analyzed with the receiver operating characteristic (ROC) curves and the of qEEG predictive ability of death in patients with disturbance of consciousness were determined. Results A total of 109 patients with supratentorial lesions were enrolled in the study, 79 of them were in the survival group and 30 of them were in the death group. The GCS scores (5 ±3vs. 9 ±3, P =0. 000) and SEF95 (7. 0 ±4.0 vs. 10. 0 ±4. 0, P = 0. 002) in the death group were significantly lower than those in the survival group. Multivariate logistic regression analysis showed that GCS scores (odds ratio 0. 100, 95% confidence interval 0. 029-0. 353) and SEF95 (odds ratio 0. 853, 95% confidence interval 0. 740-0. 983) were the independent predictors of recent prognosis. ROC curve analysis showed that the lower the GCS scores and SEF95 were, the greater the likelihood of death in patients. When SEF95 was ＜7. 75, the sensitivity to determine the death was 60. 0%, the specificity was 72. 2%, the positive predictive value was 82. 6%, and the negative predictive value was 45. 0%; when the GCS score was ＜8, the sensitivity to determine the death was 83. 3%, the specificity was 73. 4%,the positive predictive value was 82. 6%, and the negative predictive value was 45. 0%. Conclusions SEF95 helps determine the prognosis of patients with disturbance of consciousness, and it is expected to become an important means of bedside assessment of prognosis in patients with disturbance of consciousness in NICU.