危重症患者预后的危险因素及相关指标的预后预测价值分析

目的:探讨危重症患者预后的危险因素,并分析相关指标对患者预后的预测价值。方法:对2016年4月至2018年4月上海交通大学附属第六人民医院救治的5585例ICU危重症患者病例进行回顾性分析,收集患者一般资料、初始乳酸水平、24 h乳酸水平、24 h乳酸清除率以及APACHEII评分等,采用多因素logistic回归分析危重症患者预后的影响因素,ROC曲线评估多项指标预测患者预后的价值。结果:共纳入1465例危重症患者,多因素logistic回归分析显示,住ICU平均时间长、APACHEII评分高、初始乳酸水平大于4 mmol/L、24 h乳酸清除率低、术后肾功能异常是危重症患者住院期间死亡的危险因素(P0.05),24 h乳酸清除率、APACHEII评分及初始乳酸水平三者联合评估预测患者住院死亡率的ROC曲线下面积(ROC-AUC)为0.900,高于单独检测。结论:住ICU时间、APACHEII评分、初始乳酸水平、24 h乳酸清除率、术后肾功能是危重症患者住院期间死亡的影响因素,初始乳酸水平、24 h乳酸清除率和APACHEII评分联合使用可以更好地预测危重症患者的短期预后。

Analysis of the Predictive Value of Prognosis Prediction of Risk Factors and Related Indexes of Prognosis in Critically Ill Patients

ABSTRACT Objective: To explore the risk factors of the prognosis of critically ill patients and analyze the predictive value of related indexes on the prognosis of patients. Methods: From April 2016 to April 2018, a retrospective analysis was made of 5585 cases critically ill patients in ICU treated by the Sixth People''s Hospital Affiliated to Shanghai Jiaotong University. General datas, initial lactate level, 24 h lactate level, 24 h lactate clearance rate, APACHE II score and other datas of patients were collected. Multivariate logistic regression was used to analyze the influencing factors in the prognosis of critically ill patients. ROC curve was used to evaluate the value of multiple indexes in predict the prognosis of patients. Results: A total of 1465 critically ill patients were included. Multivariate logistic regression analysis showed that long average stay in ICU, high APACHE II score, initial lactate level greater than 4 mmol/L, low 24 h lactate clearance rate, postoperative renal dysfunction were risk factors for death of critically ill patients during hospitalization(P<0.05). The ROC area under curve(ROC-AUC) of assessed and predicted the death rate of critically ill patients during hospitalization by combining 24 h lactate clearance rate, APACHE II score and initial lactate level was 0.900, which higher than single tested. Conclusion: The time of stay in ICU, APACHE II score, initial lactate level, 24 h lactate clearance rate, postoperative renal function were influencing factors of critically ill patients death during hospitalization. Combining use of initial lactate level, 24 h lactate clearance rate and APACHE II score can predict the short-term prognosis of critically ill patients better.