论文部分内容阅读
目的:探索脓毒性休克患者初始复苏后外周灌注指数（PI）联合中心静脉-动脉二氧化碳分压差/动脉-中心静脉氧含量差（Pv-aCOn 2/Ca-vOn 2）对预后的预测价值。n 方法:河北医科大学脓毒哈励逊国际和平医院急诊重症监护病房（EICU）2019年1月至2021年1月76例脓毒性休克患者，均按照2016年严重脓毒症与脓毒性休克治疗国际指南（SSC 2016）进行液体复苏，并监测PI，完成3 h初始液体复苏时行中心静脉及动脉血气分析，计算Pv-aCOn 2/Ca-vOn 2比值。记录初始复苏3 h PI、Pv-aCOn 2/ Ca-vOn 2、血流动力学参数、氧代谢参数、APACHEⅡ、SOFA。根据28 d生存情况分为生存组和死亡组，比较两组患者临床指标，绘制根据PI、Pv-aCOn 2/Ca-vOn 2临界值分组患者的Kaplan-Meier生存曲线，Log-rank法检验组间差别，多因素Cox回归分析筛选28 d病死率的预测因素，受试者工作特征（ROC）曲线评估两指标对预后的预测价值。n 结果:存活组37例、死亡组39例，存活组PI[（1.77±0.63）n vs.（0.89±0.69）]高于死亡组，Pv-aCOn 2/Ca-vOn 2[（1.52±0.52）n vs.（2.57±0.86）]低于死亡组，（均n P<0.05）。Kaplan- Meier生存曲线显示高PI组中位生存时间[20.09 d（95%n CI:16.95-23.24) n vs.11.00 d（95%n CI:7.14-14.86)]长于低PI组（n χ2=12.424，n P=0.000），低Pv-aCOn 2/Ca-vOn 2组中位生存时间[23.74 d（95%n CI:20.35~27.13)n vs.12.85 d（95%n CI:9.75~15.95)]长于高Pv-aCOn 2/Ca-vOn 2组（n χ2=12.200，n P=0.000）。Cox回归分析显示PI（n RR=0.397，95%n CI：0.230~0.687，n P=0.001),Pv-aCOn 2/Ca-vOn 2（n RR=1.878，95%n CI：1.169~3.019，n P=0.009)均是脓毒性休克28 d死亡的预测因素。经ROC分析显示，PI、Pv-aCOn 2/ Ca-vOn 2预测脓毒性休克患者28 d死亡的AUC分别为0.828（95%n CI：0.732~0.923）、0.785（95%n CI：0.677~0.893），分别以0.52（灵敏度58.3%，特异度94.4%）及2.35（灵敏度88.9%，特异度63.9%）为最佳临界值，二者联合预测的AUC为0.903（95%n CI：0.835~0.971）。n 结论:PI联合Pv-aCOn 2/ Ca-vOn 2可更好地预测脓毒性休克患者的预后，是复苏阶段有价值的指标。n “,”Objective:To explore the predictive value of peripheral perfusion index (PI) combined with central venous-arterial carbon dioxide tension to arterial-venous oxygen content ratio(Pv-aCOn 2/Ca-vOn 2)for prognosis after initial resuscitation of septic shock.n Methods:A total of 76 cases of patients with septic shock from January 2019 to January 2021 in emergency intensive care unit (EICU) of Harrson international peace hospital affiliated to Hebei Medical University were enrolled. All recovered according to 2016 Severe Sepsis and Septic Shock Treatment International Guidelines 2016 (SSC 2016) , and PI was monitored, central vein and arterial blood gas analysis was performed, and the ratio of Pv-aCOn 2/Ca-vOn 2 was calculated.The PI and Pv-aCOn 2/Ca-vOn 2 at 3 h,hemodynamic variables,oxygen metabolism indexes,APACHEⅡ and SOFA score were recorded.Patients were divided into survival group and death group according to 28 d survival condition, the dfferences in demographics and clinical data were compared between two groups.The Kaplan-Meier urviving curve was created and the survival of the patients was analyzed by the Log-rank test. Risk factors associated with the prognosis were analyzed using the Cox regression analysis. The role of PI and Pv-aCOn 2/Ca-vOn 2 in prediting death was evaluated by receiver operating characteristic curves(ROC).n Results:There were 37 cases in survival group and 39 cases in death group.Compared with death group, PI in survival group [(1.77±0.63) n vs. (0.89±0.69)]was significantly higher,and Pv-aCOn 2/Ca-vOn 2[(1.52±0.52) n vs. (2.57±0.86)] was significantly lower ( n P<0.05). Kaplan-Meier survival curve showed that the median survival time in the high PI group [20.09 d (95%n CI:16.95-23.24) n vs.11.00d (95%n CI:7.14-14.86)] was longer than that in the low PI group(χn 2=12.424, n P=0.000),and that in low Pv-aCOn 2/Ca-vOn 2 group [23.74 d (95%n CI:20.35-27.13) n vs.12.85d (95%n CI:9.75-15.95)] was longer than that in the high Pv-aCOn 2/Ca-vOn 2 group (χn 2=12.200, n P=0.000) .Cox regression analysis showed that both PI (n RR=0.397, 95%n CI: 0.230-0.687, n P =0.001) and Pv-aCOn 2/Ca-vOn 2 (n RR=1.878, 95%n CI: 1.169-3.019, n P =0.009) were predictors of 28 d mortality.The area under the ROC curve of PI and Pv-aCOn 2/Ca-vOn 2 for predicting 28 d death in patients with septic shock were 0.828 (95%n CI: 0.732-0.923) and 0.785 (95%n CI: 0.6777-0.893)respectively. The optimal cutoff values were 0.52 (sensitivity 58.3% and specificity 94.4%) and 0.35 (sensitivity 88.9% and specificity 63.9%)respectively, and the AUC of the combined prediction of the two indicators was 0.903 (95%n CI: 0.835-0.971).n Conclusions:Combination of PI and Pv-aCOn 2/Ca-vOn 2 is better to predict the risk of adverse outcomes of septie shock patients,and may provide useful information for the resuscitation at early stage.n