摘要
目的:探讨中心静脉-动脉血二氧化碳分压差( Pcv-aCO2)对血液净化治疗顽固性感染性休克的预测价值。方法采用前瞻性观察研究方法,选择2011-06~2014-06在葛洲坝中心医院重症医学科治疗的55例接受血液净化治疗的顽固性感染性休克患者,将血液净化治疗48 h后去甲肾上腺素(NE)剂量及血乳酸(Lac)下降≥30%的患者定义为治疗有反应,反之为治疗无反应,对应将患者分为有反应组及无反应组。记录两组患者在血液净化治疗前和治疗后48 h循环灌注指标,包括Pcv-aCO2、中心静脉血氧饱和度( ScvO2)、血Lac;观察器官功能指标,包括血肌酐(Cr)、N末端脑钠肽前体(NT-proBNP)、血清降钙素原(PCT)、氧合指数(PaO2/FiO2)及急性生理学与慢性健康状况评分系统Ⅱ( APACHEⅡ)评分、序贯器官衰竭评分系统( SOFA)评分。通过Logistic回归分析,评价各指标与治疗反应的关系,通过受试者工作特征( ROC)曲线评估各个参数对治疗反应的预测能力。结果与治疗有反应组比较,治疗无反应组Pcv-aCO2、PCT明显升高[Pcv-aCO2(mm Hg):9(8,12)比8(6,8),Z=-4.551,P=0.000;PCT(ng/mL):6.0(4.0,9.0)比4.5(3.0,6.0),Z=-2.442,P=0.015],7 d SOFA评分变化[ΔSOFA (D1~D7)评分]明显恶化[ΔSOFA (D1~D7)评分(分):-1(-2,0)比3(2,4),Z=-3.743,P=0.000],住院病死率比有反应组升高[60.87%(14/23)比25.00%(8/32),P=0.007]。相关分析显示,入组48 h后Pcv-aCO2与血Lac(r=0.494,P=0.000)呈明显正相关。对单因素分析中有统计学意义的 Pcv-aCO2和PCT进行多因素Logistic回归分析发现,Pcv-aCO2水平升高[ OR=3.198,95% CI 1.487~6.877,P=0.003]是血液净化治疗反应差的独立危险因素。 Pcv-aCO2预测血液净化治疗无反应的ROC曲线下面积( AUC)为0.855,最佳临界值为8 mm Hg 时,敏感度为69.57%,特异度为84.37%,高于血清PCT、血Lac及ScvO2的预测价值。联合Pcv-aCO2和血清PCT共同预测治疗反应,AUC为0.873,敏感度为73.91%,特异度为93.75%。结论 Pcv-aCO2可预测血液净化治疗顽固性感染性休克的治疗反应。
Objective To approach the predictive value of central venous - to - arterial carbon dioxide difference (Pcv -aC02 ) in refractory septic shock patients receiving blood purification therapy. Methods A prospective observational study was conducted. Fifty -five refractory septic shock patients admitted to Department of Critical Care Medicine of Gezhouba Central Hospital from January 2011 to March 2014 were enrolled. After all patients received blood purification therapy at least 48 h, they were divided into responders group and non - responders group. Patients were considered responders if they stabilized MAP with ≥ 30% decrease in both norepinephrine requirements and lactate, otherwise they were considered non - responders. The following data were collected at 0 and 48 hours ( To, T4s ) after enrolled: perfusion parameters [ Pcv - aCO2, ScvO2, serum lactate (Lac) ], organ function - related parameters [serum creatinine (Cr) , oxygenation index (PaOJFiO2) , N -terminal prohormone brain natriuretic peptide ( NT - proBNP) ], APACHE II score, SOFA score, and hospital mortality. The independent risk factors of the response to the blood purification therapy were analyzed by univariate and muhivariable Logistic regression, Receiver operating characteristic curve (ROC) was plotted to evaluate the value of Pcv - aCOz in predicting the response to HVHF. Results Compared with responders group, the non -responders group showed significantly elevated Pcv -aCO2 and PCT [ Pcv -aCO2 (mm Hg): 9 (8,12) vs 8 (6,8), Z= -4.551, P=0. O00; PCT (ng/mL) : 6.0 (4.0,9.0) vs4.5 (3.0,6.0) , Z = - 2. 442, P = 0.015 ] , significantly worse 7 d SOFA score changes [ ASOFA ( D1 - D7) score] [ASOFA (D1 -DT) score (points): -1(-2,0) vs 3 (2,4), Z= -3.743, P= 0. 000], and significantly higher hospital mortality [ 25.00% ( 8/32 ) vs 60. 87% ( 14/23 ), P = O. 007]. Correlation analysis showed that the Pcv - aCOz and Lac (r = 0. 494, P = 0. 000) were positively correlated at T4s. Pcv - aCOz and PCT were founded statistical significance in univariate analysis when analyzed by muhivariable logistic regression. High Pcv - aCO2 at admission [ odds ratio (OR) = 3. 198, 95% confidence interval (95% CI) = 1. 487 ~ 6. 877, P = 0. 003 ] was the independent prognostic factors for the poor response to HVHF. The area under the ROC curve of Pcv - aCOa was 0. 855, the optimal critical value of Pcv - aCO2 was 8 mm Hg, the sensitivity was 69.57%, and specificity was 84. 37%, the levels of Pcv - aCOz was better in predicting the response to blood purification therapy than PCT, Lac and ScvO2 in refactory septic shock patients. The AUC of Pcv - aCO2 combined PCT was O. 873, the sensitivity was 73.91% and specificity was 93.75%. Conclusion Central venous- arterial carbon dioxide difference (Pcv- aGO2 ) can predict the response to blood purification treatment of refractory septic shock.
出处
《中国急救医学》
CAS
CSCD
北大核心
2015年第9期787-793,共7页
Chinese Journal of Critical Care Medicine
基金
云南省应用基础研究项目(昆明医科大学联合专项,2014FA012)