行为量表联合电生理检查对颅脑损伤意识障碍患者预后的评估

Combination of hehavioral scale and neuroelectrophysiological examination in prognosis of patients with disorders of consciousness

目的探讨行为量表联合电生理检查对颅脑损伤后意识障碍(DOC)患者预后评估的价值。方法回顾性分析49例脑损伤后意识障碍患者的临床资料。给患者进行昏迷恢复量表(CRS-R)评分、常规神经电生理及脑机接口(BCI)检测、格拉斯哥预后量表(GOS)评分,将结果划分不同等级。采用二元Logistic回归分析法,建立回归模型,并绘制模型对应的ROC曲线判断模型评估预后的价值。结果单因素分析发现,两组患者脑电图(EEG)分级、CRS-R评分、体感诱发电位(SEP)分级及BCI检测在线准确率比较,差异均有统计学意义(均P<0.05)。Logistic回归分析发现,BCI在线准确率与患者预后的相关性最强(OR=9.483,P=0.036),EEG等级(OR=7.765,P=0.043)及SEP等级(OR=7.075,P=0.047)次之。绘制ROC曲线发现,联合4种方法综合所对应的AUC值最高,为0.920(P=0.039)。结论 CRS-R评分、EEG、SEP及BCI均能较好地评估DOC患者的预后;而4种方法联合检测的准确率更高,值得在临床推广。

Objective To explore the value of combination of behavioral scale and neuroelectrophysiological examination in prognosis of patients with disorders of consciousness（ DOC）. Methods The clinical data of 49 patients with disorders of consciousness were analyzed retrospectively. The scores of behavioral scale and results of neuroelectrophysiological examination and brain computer interface（ BCI）were collected and graded. Glasgow outcome scale（ GOS） was used to assess the prognosis of DOC patients. Bilateral logistic regression analysis was used to determine whether those methods can predict outcome of DOC accurately or not and its extent. ROC curve was drawn to detect the value of the model.Results Univariate analysis showed that EEG,CRS-R score,SEP and online accuracy of BCI were different significantly（ P 0. 05）. Logistic multivariate analysis showed that BCI（ OR = 9. 483,P =0. 036） correlated with the prognosis most significantly,then next were EEG（ OR = 7. 765,P =0. 043） and SEP（ OR = 7. 075,P = 0. 047）. AUC of the four methods integration was better than any single method. Conclusion CRS-R score,EEG,SEP and BCI can be used to evaluate the prognosis of DOC patients,but integration of these four methods is much better and should be applied widely.