【作者】 林渺；

【导师】 闵苏；

【作者基本信息】 重庆医科大学 ， 麻醉学， 2010， 硕士

【摘要】 目的通过对肝移植患者的围术期临床资料进行统计分析,探讨影响肝移植患者术后肾损害发生率的主要危险因素,及控制性低中心静脉压(controlled low central venous pressure,CLCVP)的作用。方法本研究的治疗方案经重庆医科大学附属第一医院伦理学委员会通过。回顾2002年2月至2009年5月期间肝移植患者的围术期临床资料,入选术前无肾脏疾病、高血压、糖尿病的肝癌或肝硬化患者共54例。记录患者年龄、肝病诊断、Child-Pugh分级、肝移植方式、是否采用CLCVP、手术时间、无肝期时间、术后肺部感染情况、术后肾功能损害情况、术中MAP低于60 mmHg时间、术中出血量、术中输血量、术中输液量、术中尿量和患者术后肾功能恢复时间共15项指标用于统计分析。患者按术后肾功能的情况分为术后肾功能正常组(N组),急性肾功能不全组(acute renal dysfunction,ARD)和急性肾功能衰竭组(acute renal failure,ARF)。将上述除肾功能恢复时间及术后肾功能损害情况外的13项指标作为协变量,赋值为等级变量后绘制患者在不同协变量影响下发生术后肾功能损害的频数表并进行秩和检验。用Kaplan-Meier法绘制协变量的术后肾功能恢复率曲线并进行log-rank检验。多分类有序应变量的Logistic回归分析:先将各协变量逐个代入Logistic回归模型进行单因素Logistic回归分析作为初步筛选,以P<0.05为协变量对术后肾功能损害发生率的影响有统计学意义。再将单因素筛选中对术后肾损害发生率影响有统计学意义的协变量同时引入Logistic回归模型进行多因素回归分析,计算各协变量的B值及OR值,并以似然比检验进行模型拟合优度检验,最终得出影响术后肾功能损害发生率的主要危险因素并根据OR值判断其影响力的大小。将CLCVP分别与单因素单因素Logistic回归分析中筛选出的危险因素同时代入Logistic回归模型,以探讨其相互作用。结果(1)秩和检验表明:术中低血压、术中出血量、术中输血量、术中输液量和术后肺部感染在N组、ARD组、ARF组组间差异有统计学意义(P<0.05),年龄、肝病诊断、Child-Pugh分级、肝移植方式、CLCVP、手术时间、无肝期时间、术中尿量在三组间差异无统计学意义(P>0.05)。(2)肾功能恢复率曲线:术中合并低血压的患者术后肾功能恢复速度明显慢于术中未合并低血压的患者(P<0.05),术中输血量大于4000 ml的患者术后肾功能恢复速度明显慢于输血量小于4000 ml的患者(P<0.05),术后合并肺部感染的患者肾功能恢复速度明显慢于术后未合并肺部感染的患者(P<0.05)。(3)多分类有序应变量的Logistic回归分析:①单因素Logistic回归分析表明,术中低血压、术中出血量、术中输血量、术中输液量、术后肺部感染对患者术后肾功能损害发生率变化的影响有统计学意义(P<0.05),年龄、肝病诊断、Child-Pugh分级、肝移植方式、CLCVP、手术时间、无肝期时间、术中尿量对患者术后肾功能损害发生率变化的影响无统计学意义(P>0.05)。②多因素Logistic回归分析表明,术中低血压(B=-2.998,OR=0.050)、术中输血量(B=-2.025,OR=0.132)及术后肺部感染(B=-2.201,OR=0.111)对术后ARF发生率的影响有统计学意义(P <0.05),似然比检验显示Logistic模型拟合良好(P<0.05)。③协变量的相互作用:将CLCVP分别与单因素Logistic回归分析中筛选出的术中低血压、术中出血量、术中输血量、术中输液量、术后肺部感染同时代入Logistic回归模型,结果显示在分别引入CLCVP后,术中低血压、术中输血量、术后肺部感染对患者术后肾功能损害发生率的影响仍有统计学意义(P<0.05),在单因素Logistic回归分析中对患者术后肾功能损害发生率影响有统计学意义的术中出血量、术中输液量无统计学意义(P>0.05)。结论(1)影响患者术后肾功能损害发生率因素的强弱依次为:①术中低血压,②术后肺部感染,③术中大量输血。(2)CLCVP可能减轻大量出血、输液对肾功能的损害,对肝移植患者术后肾功能有保护作用。但不能抵消术中大量出血、术中低血压、术后肺部感染对肾功能的损害。更多还原

【Abstract】 Objectives The purpose of this research is to discuss the factors which affect liver transplantation patient’s renal function impairment incidence rate by analyzing patients’perioperative data.Methods The medical treatment in this research is permitted by Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. The data collection process lasted from 2002.2 to 2009.5 in The First Affiliated Hospital of Chongqing Medical University. 54 samples were collected excluding patients who got kidney disease and hypertension, diabetes. The 15 indicators:age, cancer or cirrhosis, Child-Pugh degree, Liver transplantation method, CLCVP(controlled low central venous pressure), the renal function recovery time, the renal function after surgery, Hypotension, blood loss, blood transfusion volume, transfusion volume, urine volume, operation time, Anhepatic period and Pulmonary infection were recorded for statistical analysis. The Kaplan-Meier approach would be taken to plot the chart, and log-rank check was necessary. These samples were categorized into three groups: N Group, which represented post-op patients with normal renal function; ARD Group, which represented patients with ARD; and ARF Group, which represented patients with ARF. All the factors affecting renal functions would be recorded in frequency table, and followed be ranking and checking. All 13 factors excluding the renal function recovery time and the renal function after surgery as covariates will be given a number as equivalent variable, and be taken into the Logistic regression model. P<0.05 means the factors has statistical significance on patients renal recovery, and it was the condition of the screening process which all the factors will go through. Then, we put the risk factors that had statistically significant in Single factor screening into Logistic regression model at the same time,get the risk factors’B and OR value, and do likelihood ratio test of the model as well. Finally, the factors affecting renal function impairment incidence rate have been discovered. Also, the value of B, OR and the number given to covariates have been used to judge the property of covariates. The protective covariates and destructive covariate have been taken into Logistic regression model to see the interaction.Results (1)Ranking and checking shows that Hypotension, blood loss, blood transfusion volume, transfusion volume, and pulmonary infection are the cause of statistical significance among N group, ARD group, ARF group, while age, cancer or cirrhosis, Child-Pugh degree, Liver transplantation method, CLCVP, operation time, Anhepatic period, urine volume are not. (2)The renal function recovery curve:The curve shows that renal function recovery is faster on patients who has no hypotension than the one who has hypotension. The same situation happens on patients who’volume of bleeding more than 4000 ml or get pulmonary infection. ( 3 ) Logistic regression :①Single factor Logistic regression : Hypotension, blood loss, blood transfusion volume, transfusion volume, and pulmonary infection has statistical significance on patients’renal function impairment incidence rate, but age, cancer or cirrhosis, Child-Pugh degree, Liver transplantation method, CLCVP,operation time, Anhepatic period, urine volume are not.②Multivariate Logistic regression: Hypotension(B=-2.998,OR=0.050), blood transfusion volume,(B=-2.025,OR=0.132)and pulmonary infection(B=-2.201,OR=0.111)has statistical significance affect on renal function impairment incidence rate. Likelihood ratio test shows that the logistic regression model is fitting(P<0.05).③The interaction of covariate: Hypotension, blood transfusion volume, and pulmonary infection still remains the same after introducing CLCVP in Logistic regression model, but blood loss and transfusion volume had no statistics significant. In single Logistic regression, the 5 factors both have statistics significant.Conclusions (1) Factors which affect liver transplantation patient’s renal function impairment incidence rate sort by intensity are:① hypotension②pulmonary infection③massive blood transfusion (2) CLCVP could protect renal function by reducing blood loss and massive infusion,but can not offset the rennal impairment of massive blood transfusion , hypotension, pulmonary infection.更多还原