【作者】 张辉；

【导师】 易著文；

【作者基本信息】 中南大学 ， 临床医学， 2012， 博士

【摘要】 第一部分儿童重症监护室中脓毒症患儿发病情况调查目的为了了解儿童重症监护病房PICU脓毒症的发病特点。方法对2011年4月1日至2011年7月31日入住湖南省儿童医院重症监护病房二区的1月至18周岁之间的脓毒症患儿作为研究对象。按照脓毒症诊断标准筛选病例,符合诊断标准者均纳入研究,对其临床特点、各检查结果及预后的变化,进行分析。结果本实验共纳入109例脓毒症患儿,其中普通脓毒症占53.2%,严重脓毒症占30.6%,脓毒症休克者为16.2%。男性占64.9%,女性占35.1%,男女之比为1.85：1。所有患儿中,18%来自城市,82%来自农村。年龄最小者30天,最大者10岁,中位数年龄为294天,1月至3月龄者占17.1%,3月至6月龄者有15.3%,6月龄至1岁者占23.4%,1岁内患儿累计占55.9%,1岁至2岁患儿占31.2%。66例有病原学培养结果：血培养阳性者16例(14.7%),其中9例为革兰氏阴性菌感染,7例(45.9%)为革兰氏阳性菌感染。痰培养阳性者59例(54.1%),其中最常见的是肺炎克雷伯杆菌(12例)及肺炎链球菌(10例)。其原发病灶最常见的为呼吸系统疾病,占64.9%,其次为颅内感染及败血症。脓毒症的死亡率为21.6%,其死亡率及总费用随着脓毒症的严重程度的增加呈上升趋势。而死亡组与存活组比较,虽总费用并无显著差异,但前者住院时间较后者短,日平均费用明显增加(P=0.005)。经多因素回归分析发现,重症脓毒症不良预后的危险因素包括：PCIS评分、降钙素原、PF比值、胱抑素C。结论1.脓毒症在PICU中非常常见,以婴幼儿为主,主要病因为呼吸系统疾病。体液培养以革兰氏阴性菌为主。血培养阳性率低,痰培养阳性率较高。2.脓毒症患儿的死亡率及总费用均随着脓毒症的严重程度呈上升趋势,死亡患儿的日平均费用较存活组明显增高。3. PCIS评分、降钙素原、PF比值、胱抑素c是脓毒症死亡相关危险因素。第二部分儿童重症监护病房中脓毒症急性肾损伤的临床观察目的为了了解湖南省儿童医院重症监护病房中(PICU)脓毒症AKI患儿临床特点及其检查指标的变化,通过比较脓毒症AKI组及脓毒症非AKI组临床资料、预后等,探讨其异同点。方法观察自2011年4月1日至2011年7月31日期间入住PICU的30天至18岁的脓毒症患儿,建立统一的调查表,自患儿入ICU至其出院第28天或死亡当日作为观察时间。入PICU时,入住PICU第3天及第7天留取患儿血、尿标本送检。按照急性肾损伤网络(AKIN)制定的AKI诊断标准,将其分为脓毒症AKI组及脓毒症非AKI组。每日观察患儿病情,所有资料如实填入统一调查表。结果纳入的109例患儿中,其中17例因缺乏36-48小时血肌酐值而无法判断其肌酐变化。余下92例患者以肌酐变化作为诊断标准诊断AKI患儿32例(34.78%)。其中因尿少诊断为AKI者仅1例。而脓毒症、严重脓毒症及脓毒症休克患者其AKI的发病率分别为：19.15%(9/47),30%(9/30),93.3%(14/15)。随着脓毒症严重程度的增加,其发病率明显上升,其OR值为4.414(95%的可信区间为1.737-11.216),提示脓毒症是AKI发生的独立危险因素。各组的死亡率分别为2.12%(1/47),23.3%(7/30),46.7%(7/15)。所有AKI患儿中,AKI1期者8例,AKI2期者11例,AKI3期者13例。将脓毒症AKI组(32例)与脓毒症非AKI组(60例)进行比较,其性别、城乡、血培养、痰培养、其他培养及尿常规值之间无明显差异。而两组间预后、进行机械通气的人数及脓毒症严重程度之间有显著性差异。两组间比较,其PCIS评分、第3d血肌酐值、入院时尿素氮、心肌酶(肌酸激酶及其同工酶、乳酸脱氢酶)、血小板、降钙素原、碳酸氢根及二氧化碳分压之间均有显著性差异。脓毒症AKI组住PICU时间、总费用也较脓毒症非AKI组明显增加,均有显著差异。而住院时间、平均日费用及其他临床检查指标间均无明显差异。结论根据脓毒症急性肾损伤诊断标准,脓毒症急性肾损伤的发病率为34.78%。脓毒症AKI的发生率及死亡率随着病情严重程度及脓毒症的严重程度的增加呈上升趋势。脓毒症AKI与脓毒症非AKI组比较,其住PICU时间及总费用明显增加,所需的医疗资源投入明显增多。两者临床检查资料多个指标也有明显差异,肾损伤与其他脏器损伤之间有密切联系,提示多脏器功能损伤的机制可能相似。第三部分脓毒症肾损伤相关指标间的联系目的探讨早期诊断AKI生物标记物CysC, KIM-1, NGAL与脓毒症AKI诊断及其预后的关系。方法本研究选择2011年4月1日至2011年7月31日入住湖南省儿童医院重症监护病房二区的1月至18周岁之间的脓毒症患儿作为研究对象。按照脓毒症及急性肾损伤的诊断标准筛选病例,纳入研究者分为脓毒症AKI组和脓毒症非AKI者。通过观察比较Cys C、KIM-1及NGAL等指标在脓毒症AKI的诊断及其预后等方面的特点。分析其诊断能力及预测能力。结果经过分析比较发现在脓毒症AKI诊断方面,Cys C、KIM-1、血NGAL及尿NGAL水平均在入院时即有明显升高,较血肌酐更早的预测脓毒症AKI的发生,此外,以上指标在预测脓毒症AKI预后方面也是比较好的指标。Cys C诊断脓毒症AKI的ROC曲线下面积(AUC)分别为0.927,cutoff值为1.045,超过此值时诊断为脓毒症AKI的敏感度为86.5%,特异度为93.8%。在预测脓毒症AKI的预后方面,其AUC高达0.815,Cutoff值为2.805mg/l,超过此值时其敏感度及特异度分别为80%,83.1%。尿KIM-1诊断脓毒症AKI的AUC0.852,其Cutoff值为20.47ng/l,超过此值时诊断为脓毒症AKI的敏感度高达90.7%,特异度72.5%。其预测脓毒症AKI的ROC曲线下面积为0.799(95%CI为0.687-0.910,P<0.001)。Cutoff值为33.4ng/l,此时敏感度及特异度分别为76.9%、72.9%。血NGAL (pNGAL)诊断及预测脓毒症AKI的AUC分别为0.931、0.815,两者的Cutoff值分别为63.61pg/ml及134.61pg/ml,超过此值时诊断为脓毒症AKI的敏感度高达87.5%,特异度86.6%,预测其预后的敏感度及特异度分别为85.7%,66.2%。尿NGAL (uNGAL)诊断及预测脓毒症AKI的AUC分别为0.906、0.808,两者的Cutoff值分别为65.88pg/ml,80.37pg/ml,超过此值时诊断为脓毒症AKI的敏感度高达74.2%,特异度为96.8%,预测其预后的敏感度及特异度分别为83.3%,86%。各指标与第3天血肌酐值呈不同程度的正相关(相关系数自0.330至0.585,P<0.05),比较发现血检结果与血肌酐的相关性较尿检结果更高,前者多为肾小球功能,而后者多反映的是肾小管功能。综合分析比较后发现各指标在脓毒症的诊断及预测方面各有优点。结论脓毒症肾损伤在儿童重症监护室中的发病率高,肌酐及尿量作为其诊断指标敏感性低。胱抑素C、尿KIM-1、血尿NGAL均可作为AKI的早期诊断指标,较血肌酐可提前36h。且与脓毒症AKI的预后有关。各指标与第3天血肌酐均呈显著正相关。血检结果与尿检结果具有差异,AKI也应从肾小球功能、肾小管功能等多方面阐述。更多还原

【Abstract】 Part Ⅰ:The clinical observation of sepsis in pediatric intensive care unitObjectiveThe consistent data about the incidence and outcome of sepsis in pediatric intensive care(PICU) are lacking in China. This study was designed to understand the clinical characteristics of sepsis in Hunan Provincial Children’s Hospital PICU.MethodsThis is a Prospective cohort study performed in Hunan Provincial Children’s Hospital intensive care ward Ⅱ. We prospectively followed the children (from1month to18years old) diagnosis with sepsis from April1,2011to July31,2011, until their discharge,28th day of stay or death。 For all patients we collected the following data at PICU admission:age, gender, pediatric critical illness score (PCIS), associated underlying diseases and cost. Sepsis was diagnosed according to SIRS, sepsis, severe sepsis, and septic shock creteria publiced on Pediatric Critical Care Medicine (PCCM) in2005.ResultsThere were109patients,53.2%sepsis,30.6%severe sepsis and16.2%septic shock. Of these,64.9%were male, and82%from rural areas. The youngest patient was30days, and the oldest was10years old, the median age was294days. And55.9%of the children less than1-year-old. In64.9%, the infection was of respiratory origin followed by intracranial infection. There were66cases were detected with etiology culture:Blood culture was positive in16cases (14.7%), of which nine cases were Gram-negative bacteria, seven cases (45.9%) were Gram-positive bacteria. Sputum culture was positive in59cases (54.1%), Klebsiella pneumoniae (12cases) and Streptococcus pneumoniae (10cases) were the most common. Mortality rate was21.6%. The mortality and the total cost was rising with the severity of sepsis. Compared to the survival group, the average cost of the death group was significantly higher (P=0.005). Multivariate regression analysis showed that the poor prognosis (death) risk factors of severe sepsis include:the PCIS score, procalcitonin, the PO2/FiO2ratio and cystatin C.ConclusionSepsis was very common in yonger children in PICU,and the main cause was respiratory diseases. The mortality and the total cost was rising with the severity of sepsis. Compared to the Survival group, the average cost of the Death group was significantly higher. Multivariate regression analysis showed that poor prognosis (death) risk factors of severe sepsis include:the PCIS score, procalcitonin, the PF ratio and cystatin C. Part II The clinical observation of septic acute kidney injury in pediatric intensive care unitObjectivesWe conducted a study to evaluate the incidence, risk factors and outcomes associated with early acute kidney injury (AKI) in sepsis in PICU.MethodsThe study was a prospectively collected data from the Hunan Provincial Children’s Hospital intensive care ward Ⅱ. We prospectively followed the children (from lmonth to18years old) diagnosis with sepsis from April1,2011to July31,2011, until their discharge,28th day of stay or death. The blood and urine specimens were collected on the1d,3d,7d after the PICU admission. AKI was diagnosed according to Acute Kidney Injury Net(AKIN). We divided the patients into septic AKI group and septic non-AKI group. The main outcome measures were clinical and laboratory data and outcomes.ResultsOf all109patients,17cases could not determine the creatinine change due to the lack of second serum creatinine. Of the92patients, there were32(34.78%) septic AKI patients according to the AKI diagnostic criteria, of which8were AKI stage1,11were stage2and13were stage3. The incidence of AKI induced by sepsis, severe sepsis and septic shock was19.15%(9/47),30%(9/30),93.3%(14/15), respectively. And the mortality of each group were2.12%(1/47),23.3%(7/30),46.7%(7/15), respectively. Sepsis were independent risk factors for AKI occurred, with the severity of sepsis, the incidence increased significantly[OR=4.414(95%confidence interval:1.737-11.216)]. We found that there are significant differences in the prognosis, PICU time, the total cost, the mechanical ventilation, sepsis severity, the PCIS score, the3d serum creatinine, blood urea nitrogen, serum creatine kinase (creatine kinase and its isoenzymes, lactate dehydrogenase), platelets, procalcitonin, bicarbonate and carbon dioxide partial pressure between the two groups. ConclusionThe incidence rate of septic AKI was34.78%in PICU according to the diagnostic criteria of septic AKI. The mortality rate was significantly increased with the severity of sepsis. Compare to the sepsis non-AKI group, septic AKI group had a longer stay in PICU and the total costs increased significantly. There were significant differences in the clinical examination data. Sepsis AKI and other organ damage are closely linked, suggesting that the mechanism of multiple organ dysfunction may be similar. Part Ⅲ The clinical observation of septic acute kidney injury in pediatric intensive care unitObjectivesTo investigate the relationship between the early diagnosis of AKI biomarker (cystatin C, KIM-1, NGAL) and the diagnosis and prognosis of septic AKI.MethodsScreening cases in accordance with the diagnostic criteria for sepsis and acute kidney injury were divided to two group(septic AKI group and septic non-AKI group). Observing the characteristisc of cys C, KIM-1and NGAL in the diagnosis and the prognosis of septic AKI. Study the diagnostic capabilities and the prediction ability of each indicators.ResultWe found that the concentration of Cys C, KIM-1, NGAL was increased sinificantly at the time of PICU admission. Compare to serum crearinine, they were earlier diagnostic indicators and prognostic marker of septic AKI. The area under the curve(AUC) for reciver operationg characteristic(ROC) for Cys C-2h for prediction of septic AKI was0.927, the cutoff was1.045mg/1, The sensitivites and specificities were86.5%,93.8%, respectively. The AUC for prognosis of septic AKI was0.815. The cutoff was2.805mg/1, and the sensitivity and specificitiy were80%and83.1%, respectively. The AUC for uKIM-1for prediction of septic AKI was0.852, the cutoff was20.47ng/1, The sensitivity and specificitiy were90.7%and72.5%, respectively. The AUC for prognosis of septic AKI was0.799(95%CI:0.687-0.910, P<0.001). The cutoff was33.4ng/l, sensitivity and specificitiy were76.9%and72.9%, respectively. The AUC for pNGAL for prediction of septic AKI was0.931, the cutoff was63.61pg/ml, The sensitivity and specificitiy were87.5%and86.6%, respectively. The AUC for prognosis of septic AKI was0.815, the cutoff was134.61pg/ml, sensitivity and specificity were85.7%and66.2%, respectively. The AUC for uNGAL for prediction of septic AKI was0.906, the cutoff was65.88pg/ml. The sensitivity and specificity were74.2%and96.8%, respectively. The AUC for prognosis of septic AKI was0.808, the cutoff was80.37pg/ml, sensitivity and specificity were83.3%and86%, respectively. All of them were positively correlated with the peak serum creatinine concentration(r=0.330-0.585, P<0.05).ConclusionSeptic AKI in pediatric intensive care unit have higher incidence rate; cys C、uKIM-1、pNGAL、uNGAL were earlier marker for diagnose and predict AKI。All of them were positively correlated with the peak serum creatinine concentration。更多还原