儿童重症监护室脓毒症患儿死亡相关影响因素分析

Analysis of death-related factors in children with sepsis among Pediatric Intensive Care Unit

目的探讨儿童重症监护病房(PICU)脓毒症患儿死亡相关影响因素。方法法选择2019年6月至2021年1月,于四川大学华西第二医院PICU收治的234例脓毒症患儿为研究对象。采用回顾性分析方法,按照PICU治疗结局,将其分别纳人死亡组(n=44)和存活组(n=190)。对2组患儿临床病例资料、入院24h内实验室检查结果进行统计学分析。结合已有研究结果及临床经验及单因素分析中差异有统计学意义(P<0.05)因素,对脓毒症患儿死亡相关影响因素进行多因素非条件logistics回归分析,并建立logistic回归方程(预测模型)。绘制不同指标预测脓毒症患儿死亡相关影响因素的受试者工作特征(ROC)曲线,并计算曲线下面积(AUC)。本研究遵循的程序符合2013年修订的《世界医学协会赫尔辛基宣言》的要求。结果①2组患儿年龄、脓毒症不同严重程度构成比、使用血管活性药物者占比比较,差异均有统计学意义(P<0.05);年龄<5岁和≥5岁患儿的脓毒症严重程度占比比较,差异均有统计学意义(P<0.05)。②2组患儿的血清血小板计数(PLT)、丙氨酸转移酶(ALT)、尿素氮、血清肌酐(SCr)、凝血酶原时间(PT)、血清HCO:一、血清乳酸浓度比较,差异均有统计学意义(P<0.05)。③2组患儿的Glasgow评分、儿童序贯器官衰竭评分(pSOFA)和第3代小儿死亡危险(PRISMIⅢ)评分比较,差异均有统计学意义(P<0.05),而2组患儿小儿危重病例评分(PCIS)比较,差异无统计学意义(P>0.05)。④多因素非条件logistics回归分析显示,年龄(OR=1.011,95%CI:1.005~1.018,P<0.001)、血清乳酸浓度(OR=1.132,95%CI:1.005~1.275,P=0.041)是导致脓毒症患儿死亡的独立危险因素。建立的预测脓毒症患儿死亡相关影响因素的logistic回归方程为:y=0.918-0.312x_(1)+0.124x_(2)+0.011x_(3)(x_(1)为Glasgow评分,x_(2)为血清乳酸浓度,x_(3)为年龄),对脓毒症患儿死亡具有良好预测价值,其ROC-AUC为0.84(95%CI:0.775~0.904,P<0.05),预测脓毒症患儿死亡的灵敏度为86.4%,特异度为68.4%。结论该logistic回归方程对预测脓毒症患儿预后相关影响因素的灵敏度及综合价值,均优于pSOFA和PRISMⅢ,适用于对脓毒症患儿预后不良的早期筛查。

Objective To investigate the death-related factors in children with sepsis among Pediatric Intensive Care Unit(PICU).Methods From June 2019 to January 2021,234 children with sepsis admitted to PICU of West China Second University Hospital,Sichuan University were selected in this study.They were divided into death group(n=44)and survival group(n=190)based on whether they died during PICU stay.Clinical data and laboratory test results within 24 hours of admission were retrospectively analyzed.Based on factors with statistically significant differences(P<0.05)in the univariate analysis and clinical experience,multivariate unconditional logistic regression analysis was conducted to identify the death-related factors of children with sepsis,and a logistic regression equation(predictive model)was established.Receiver operating characteristic(ROC)curves were used to evaluate the predictive value of different indicators for mortality risk in children with sepsis,and the area under the curve(AUC)was calculated.The procedures followed in this study was in line with the World Medical Association Declaration of Helsinki revised in 2013.①Results There were statistically significant differences in the age,severity of sepsis,and proportion of patients using vasoactive drugs between two groups(P<0.05).There were also statistically significant differences in the proportion of patients with different levels of sepsis severity between patients aged<5 years and those aged≥5 years(P<0.05).②There were statistically significant differences in the levels of serum platelet count(PLT),alanine transaminase(ALT),blood urea nitrogen,serum creatinine(SCr),prothrombin time(PT),serum HCO_(3)^(-),and lactate between two groups(P<0.05).③There were statistically significant differences in the Glasgow score,pediatric sequential organ failure assessment(pSOFA),and pediatric risk of mortality Ⅲ(PRISM Ⅲ)scores between two groups(P<0.05),but no statistically significant difference in the pediatric critical illness score((PCIS)score(P>0.05).④Multivariate unconditional logistic regression analysis showed that age(OR=1.011,95%CI:1.005-1.018,P<0.001)and lactate level(OR=1.132,95%CI:1.005-1.275,P=0.041)were independent risk factors for prognosis of children with sepsis.The logistic regression equation(y=0.918-0.312x_(1)+0.124x_(2)+0.011x_(3),where x_(1) is the Glasgow score,x_(2) is the lactate level,and x_(3) is age)had good predictive value for mortality risk in children with sepsis,with the ROC-AUC of 0.84(95%CI:0.775-0.904,P<0.05).Its sensitivity was 86.4%,and its specificity was 68.4%.Conclusions Thelogistic regression equation established in this study has higher sensitivity and comprehensive value in predicting mortality risk in children with sepsis than the pSOFA score and PRISMII score,and are suitable for early screening of poor prognosis in children with sepsis.