脑电双频指数早期评估急性脑损伤预后的可行性研究

Feasibility study on early assessment of prognosis of acute brain injury by bispectral index

目的:分析脑电双频指数(bispectral index,BIS)早期评估急性脑损伤(acute brain injury,ABI)预后的可行性,探讨扩展格拉斯哥预后评分(extended Glasgow outcome scale,GOSe)和BIS的相关性。方法:采用前瞻、双盲方法,收集ABI患者起病24 h内的BIS和6个月后的GOSe值,分别计算未存活、存活、不良预后、良好预后的BIS平均值(BISmean)。对BIS和GOSe数值行一元线性回归,分别绘制预测未存活(GOSe <2分)和预测预后不良(GOSe <5分)的受试者工作特征曲线(receiver operating characteristic curve,ROC)。结果:共纳入309例患者,未存活、存活、不良预后、良好预后的BISmean分别为27.11±14.68、67.43±15.71、46.63±15.36、72.31±13.66,未存活与存活、不良预后与良好预后间差异均有统计学意义( P <0.01)。BIS和GOSe数值呈显著线性正相关(R=0.759,R 2=0.576, P <0.01),回归方程GOSe =0.093×BIS-0.288。BIS预测未存活的ROC曲线下面积(area under the curve,AUC)为0.981,BIS最佳截断点为40.7时,预测敏感度+特异度最大(敏感度=0.952,特异度=0.944);预测预后不良的AUC为0.897,BIS最佳截断点为64.1时,敏感度+特异度最大(敏感度=0.734,特异度=0.920)。结论: ABI患者的BIS和GOSe显著相关;BIS值早期评估ABI患者预后有效,连续、实时的BIS监测则更有意义。

Objective: To analyze the feasibility of using bispectral index (BIS) values to predict outcomes early in patients with acute brain injury (ABI) and study the correlation between extended Glasgow outcome scale (GOSe) scores and BIS. Methods: A prospective and double-blinded study collecting BIS values and GOSe scores was perforned in patients with ABI. Mean BIS value (BISmean) in non-survival, survival, unfavorable, and favorable outcome groups of ABI patients were calculated. Linear regression between BIS values and GOSe scores was constructed. Receiver operating characteristic (ROC) curves predicting non-survival (GOSe score of 2 or less) and unfavorable outcome (GOSe score of 5 or less) were plotted respectively. Results: There were 309 patients selected. BISmean values were significantly different between non-survival and survival groups (27.11±14.68 vs 67.43±15.71, P <0.01) and the same result was found between unfavorable and favorable outcome groups (46.63±15.36 vs 72.31±13.66, P <0.01). There was a significant positive correlation between BIS values and GOSe scores (R=0.759, R 2=0.576, P <0.01). Regression equation: GOSe=0.093×BIS-0.288. ROC curve predicting non-survival patiennts showed that area under the curve (AUC) was equal to 0.981 and the optimal cut-off point of BIS value corresponding with the maximum of sensitivity+specificity was 40.7 (sensitivity=0.952, specificity=0.944);ROC curve predicting unfavorable outcome showed that AUC was equal to 0.897 and the optimal cut-off point of BIS value corresponding with the maximum of sensitivity+specificity was 64.1 (sensitivity=0.734, specificity=0.920). Conclusions: BIS value significantly correlates with GOSe score in ABI patients. It is effective that using BIS value to predict outcome of patients with ABI, continuous and real-time BIS monitoring is more meaningful.