APACHE Ⅱ与GCS评分预测神经重症监护患者病死率的价值

Value of APACHE Ⅱ and GCS scoring in predicting mortality of patients in neurological intensive care unit

目的比较急性生理学和慢性健康状况评估Ⅱ(APACHEⅡ)和格拉斯哥昏迷评分(GCS)对神经重症监护室(NICU)患者病死率的预测价值。方法回顾性分析568例NICU患者的临床资料。根据预后分为生存组(421例)和死亡组(147例),比较两组患者的APACHEⅡ和GCS评分。采用受试者工作特征曲线(ROC曲线)评价APACHEⅡ和GCS评分对NICU患者病病死率的预测价值。经多因素Logistic回归分析影响病死率的独立因素。结果死亡组APACHEⅡ评分高于生存组[(21.01±5.59)分vs.(14.12±4.26)分](P<0.05);GCS评分低于生存组[(8.29±2.72)分vs.(10.32±2.08)分](P<0.05)。APACHEⅡ和GCS评分预测病死率的ROC曲线下面积分别为0.836和0.720(Z=3.921,P<0.01)。APACHE II和GCS评分均为NICU患者病死率的独立预测因素。结论 APACHEⅡ和GCS评分均能有效预测NICU患者的病死率,APACHEⅡ的预测价值优于GCS。

Objective To compare the value of acute physiology and chronic health evaluation Ⅱ （APACHE Ⅱ ） and Glasgow coma scale（GCS） scores in predicting the mortality of patients in neurological intensive care unit（NICU）. Methods Data of 568 patients hospitalized in（NICU） were retrospectively studied,of whom 147 cases were died（group D） and 421 cases survival（group S）. The scores of APACHE Ⅱ and GCS were compared between two groups. Receiver operating characteristic （ROC） curve was adopted for evaluating the predictive value of mortality. Multivariate logistic regression analysis was used to investigate the independent predictors for the mortality. Results The APACHE Ⅱ score of group D was （21.01i 5.59） points, which was higher than （14. 12± 4. 26） points of group S（P〈0. 05）. The GCS score of group D was （8.29±2.72） points,which was lower than （10. 32±2.08） points of group S（P〈0. 05）. The predictive values of APACHE Ⅱ and GCS were moderate. The values of the area under ROC curve were 0. 836 and 0. 720, respectively for APACHE Ⅱ and GCS （Z=3. 921,P〈0. 01）. APACHE [I and GCS scores were the independent predictors for the mortality. Conclusion APACHE Ⅱ and GCS scoring can be used as reliable predictors for the mortality of patients in neurological intensive care unit. APACHE Ⅱ scoring has a higher predictive value for the mortality than GCS.