摘要目的 明确联合中心静脉-动脉二氧化碳分压差与动脉-中心静脉氧含量差的比值(Pcv-aCO2/Ca-cvO2)和乳酸清除率(LCR)对脓毒症休克患者病情评估及预后判断的价值.方法 纳入自2013年3月至2017年5月兰州大学第二医院重症医学2科收治的145例脓毒症休克患者.对所有脓毒症休克患者根据早期目标导向治疗方案进行复苏治疗,分别于复苏前(T0)、复苏后6 h(T6)同时行桡动脉和上腔静脉血气分析,计算Pcv-aCO2/Ca-cvO2、LCR.根据复苏后6 h Pcv-aCO2/Ca-cvO2和LCR将患者分为4组:A组为Pcv-aCO2/Ca-cvO2＞1.8和LCR＜ 30％;B组为Pcv-aCO2/Ca-cvO2＞ 1.8和LCR≥30％;C组为Pcv-aCO2/Ca-cvO2≤1.8和LCR＜ 30％;D组为Pcv-aCO2/Ca-cvO2≤1.8和LCR≥30％.比较4组患者复苏前后一般临床指标、血流动力学及氧代谢指标、急性生理与慢性健康评分(APACHEⅡ)、序贯器官衰竭评分(SOFA)、重症监护病房(ICU)住院时间及28 d病死率.Kaplan-Meier预测28 d生存曲线,Log-rank检验组间差别,受试者工作特征曲线(ROC)评估指标对28 d病死率的预测价值.结果 D组复苏后3 d APACHEⅡ评分及SOFA评分均较A组低(t=-2.909、-3.630,均P＜0.05),ICU住院时间较A组短(t=-2.575,P=0.011),28 d病死率较A组低(9.1％比40.0％,x2=3.124,P=0.011).Kaplan-Meier生存分析,A组中位生存时间较其他三组短(x2=10.332,P=0.016);Log-rank两两比较,A组与D组的生存分布的差异具有统计学意义(x2=8.304,P=0.004).Cox回归分析显示,复苏6h的Pcv-aCO2/Ca-cvO2(RR=3.888,95％ CI:2.443～6.189,P＜0.001)和LCR(RR=0.073,95％ CI:0.008～0.640,P=0.018)是28 d病死率的预测因子.ROC分析显示,Pcv-aCO2/Ca-cvO2联合LCR曲线下面积(0.919,95％ CI:0.862～0.958)高于Pcv-aCO2/Ca-cvO2 (0.862,95％ CI:0.795～0.914)或LCR(0.820,95％CI:0.748～0.879),差异有统计学意义(Z=2.032、2.364,均P＜0.05).结论 Pcv-aCO2/Ca-cvO2联合LCR可更好地判断脓毒症休克患者病情及评估预后.

abstractsObjective To investigate the prognostic significance of central venous-arterial carbon dioxide tension to arterial-venous oxygen content ratio (Pcv-aCO2/Ca-cvO2) combined with lactate clearance rate (LCR) as early resuscitation goals of septic shock.Methods One hundred and forty-five septic shock patients admitted to Second Department of Critical Care Medicine of Lanzhou University Second Hospital from March 2013 to May 2017 were enrolled in this study.All septic shock patients received an initial resuscitation therapy according to early goal-directed therapy.The arterial and central venous blood gases were measured simultaneously at baseline (T0) and 6 hours after resuscitation (T6).Pcv-aCO2/Ca-cvO2 and LCR were calculated.Patients were classified into four groups according to Pcv-aCO2/Ca-cvO2 and LCR at T6:group A,Pcv-aCO2/Ca-cvO2 ＞ 1.8 and LCR ＜30％;group B,Pcv-aCO2/Ca-cvO2 ＞ 1.8 and LCR≥30％;group C,Pcv-aCO2/Ca-cvO2 ≤ 1.8 and LCR ＜ 30％;group D,Pcv-aCO2/Ca-cvO2 ≤ 1.8 and LCR ≥ 30％.General demographics,hemodynamic parameters,oxygen metabolism parameters,acute physiology and chronic health evaluation (APACHE Ⅱ) scores,sequential organ failure assessment (SOFA) scores,length of intensive care unit (ICU) stay,and 28-day mortality rate were compared among the 4 groups.A KaplanMeier curve showed the survival probabilities at day 28 using a log-rank test for multiple comparisons.Parameters were introduced into a Cox's proportional hazards regression model to analyze the prediction of 28-day mortality.Receiver operating characteristics (ROC) curves were constructed to evaluate the ability of Pcv-aCO2/Ca-cvO2,LCR,Pcv-aCO2/Ca-cvO2 combined with LCR at T6 to predict 28-day mortality.Results Compared with patients in group A,patients from group D had the lower APACHE Ⅱand SOFA score at day 3 (t =-2.909,-3.630,both P ＜ 0.05),shorter ICU stay (t =-2.575,P =0.011),and lower mortality rate at day 28 (x2 =3.124,P =0.011).Survival curves up to day 28,illustrated by Kaplan-Meier method,showed that group A had the shortest median survival time (x2 =10.332,P =0.016),difference between group A and group D was statistically significant (x2 =8.304,P =0.004).The Cox regression analysis revealed that Pcv-aCO2/Ca-cvO2 (RR =3.888,95％ Cl:2.443-6.189,P ＜ 0.001) and LCR (RR =0.073,95 ％ CI:0.008-0.640,P =0.018) at T6 were independent predictors of 28-day mortality.The area under ROC curve for Pcv-aCO2/Ca-cvO2 combined with LCR (0.919,95％ CI:0.862-0.958) was significantly greater than whether Pcv-aCO2/Ca-cvO2 (0.862,95％CI:0.795-0.914) or LCR (0.820,95％ CI:0.748-0.879) alone (Z =2.032,2.364,both P ＜0.05).Conclusion Combination of Pcv-aCO2/Ca-cvO2 and LCR is better than single parameter to predict the risk of adverse outcomes of septic shock patients,and may provide useful information for assessing the adequacy of resuscitation at early stage of septic shock.