机构地区:[1]浙江大学医学院附属第一医院重症医学科,杭州310006 [2]浙江省立同德医院重症医学科,杭州310012
出 处:《中华危重病急救医学》2020年第11期1320-1323,共4页Chinese Critical Care Medicine
基 金:浙江省基础公益研究计划(LGF18H150010);浙江省中医药科研项目(2018ZB034)。
摘 要:目的分析血中微量元素铜和锌水平对脓毒症患者预后的预测价值。方法采用回顾性病例对照研究方法,收集2016年1月至2019年12月浙江省立同德医院收治的脓毒症患者的临床资料,按28 d预后将患者分为死亡组和存活组,分析并比较不同预后两组脓毒症患者的一般人口学资料、急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ)、序贯器官衰竭评分(SOFA)、全血铜、全血锌、铜/锌比值的差异;将差异有统计学意义的指标进行受试者工作特征曲线(ROC)分析和判别分析,评价各危险因素的检验效能;根据ROC分析结果构建判别函数,分析预测值和真实值结果的一致性。结果共纳入脓毒症患者86例,其中死亡组30例,存活组56例。与存活组比较,死亡组患者年龄更大(岁:70.8±12.9比65.4±11.1),APACHEⅡ评分(分)、SOFA评分(分)、全血铜(μmol/L)、铜/锌比值水平更高(分别为26.8±5.9比19.5±6.7、14.0±3.6比7.4±4.2、17.2±2.2比15.0±1.5、0.24±0.05比0.18±0.03),全血锌水平更低(μmol/L:71.2±12.0比87.7±11.3),差异均有统计学意义(均P<0.001)。ROC曲线分析显示,APACHEⅡ评分、SOFA评分、全血铜、全血锌、铜/锌比值对脓毒症患者预后均有预测价值(均P<0.001),ROC曲线下面积(AUC)分别为0.788、0.879、0.802、0.869、0.873,95%可信区间(95%CI)分别为0.691~0.886、0.806~0.951、0.686~0.918、0.780~0.957、0.780~0.966,决策截断值分别为23.500、11.500、16.500、79.400、0.195,敏感度分别为70.0%、76.7%、73.3%、83.9%、90.0%,特异度分别为73.2%、78.6%、85.7%、83.3%、76.8%,阳性预测值分别为58.3%、65.7%、72.4%、73.5%、67.5%,阴性预测值分别为82.0%、86.3%、84.2%、90.4%、93.5%。对5个有意义的预后因素进行联合预测分析,AUC提高到0.941,决策截断值为2.500,敏感度为83.3%,特异度为91.1%,P<0.001。建立的判别方程对预后预测的结果与实际转归的符合率为88.4%(76/86),一致性评价Kappa值为0.744(P<0.001)。�Objective To analyze the prognostic value of serum trace elements Copper and Zinc in sepsis patients.Methods A retrospective case-control study was conducted to collect the clinical data of sepsis patients in Tongde Hospital of Zhejiang Province from January 2016 to December 2019.The patients were divided into death group and survival group according to 28-day outcome.The differences of general demographic data,acute physiology and chronic health evaluationⅡ(APACHEⅡ),sequential organ failure assessment(SOFA),whole blood Copper,whole blood Zinc and Copper/Zinc ratio were analyzed and compared between the two groups.Receiver operating characteristic curve(ROC)and discriminant analysis were used to analyze the date with statistical significance,and the test efficiency of each risk factor was evaluated;according to the results of ROC analysis,a discriminant function was constructed to analyze the consistency between the predicted value and the real value.Results A total of 86 patients were enrolled,including 30 in the death group and 56 in the survival group.Compared with the survival group,the patients in death group were older(years old:70.8±12.9 vs.65.4±11.1),APACHEⅡscore,SOFA score,whole blood Copper(μmol/L),Copper/Zinc ratio were higher(26.8±5.9 vs.19.5±6.7,14.0±3.6 vs.7.4±4.2,17.2±2.2 vs.15.0±1.5,0.24±0.05 vs.0.18±0.03),whole blood Zinc level was lower(μmol/L:71.2±12.0 vs.87.7±11.3),and the differences were statistically significant(all P<0.001).ROC curve analysis showed that APACHEⅡscore,SOFA score,whole blood Copper,whole blood Zinc and Copper/Zinc ratio had predictive values for the prognosis of sepsis patients(all P<0.001).The area under ROC curve(AUC)was 0.788,0.879,0.802,0.869 and 0.873,95%confidence interval(95%CI)was 0.691-0.886,0.806-0.951,0.686-0.918,0.780-0.957 and 0.780-0.966,the cut-off values were 23.500,11.500,16.500,79.400 and 0.195,the sensitivity was 70.0%,76.7%,73.3%,83.9%and 90.0%,the specificity was 73.2%,78.6%,85.7%,83.3%and 76.8%,the positive predictive values were 5
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...