基于肝硬化数据库研究症状性细菌性腹水的病原学特征与快速筛查模型及其临床应用

发布时间：2018-06-09 07:03

本文选题：肝硬化 + 腹膜炎　；参考：《南昌大学》2017年博士论文

【摘要】：背景与目的:症状性细菌性腹水(symptomatic bacterascites,SB)是一种变异的自发性细菌性腹膜炎(spontaneous bacterial peritonitis,SBP),有感染表现,但腹水多形核细胞(polymorphonuclear,PMN)计数250个/mm3,仅能通过腹水细菌培养阳性确诊。由于腹水培养常耗时数日,故临床难以及时发现SB。目前,缺乏早期诊断SB的有效方法。本研究目的是通过建立肝硬化数据库来总结SB病原学特点和探索SB快速筛查模型及其临床应用。方法:1.创建肝硬化数据库参照国内外肝硬化相关诊疗指南,经专家讨论确定数据库拟收集的内容,然后运用Epi InfoTM 7软件创建肝硬化数据库并录入临床数据。2.症状性细菌性腹水的病原菌及耐药特征从创建的数据库中筛选腹水培养阳性的SBP病例,分为SB组与常规SBP组(腹水PMN计数≥250个/mm3)。比较两组病原菌构成、耐药特征及预后。3.血降钙素原与C反应蛋白筛查症状性细菌性腹水从创建的数据库中选取符合要求的肝硬化病例,分为SB组与非感染性腹水组。利用受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under curve,AUC)来评价PCT及CRP筛查SB的价值。4.症状性细菌性腹水快速筛查模型的构建与临床应用从创建的数据库中选取符合条件的肝硬化病例,分为SB组与非感染性腹水组,通过单因素分析及二分类Logistic回归分析筛选自变量并构建SB快速筛查模型。将SB快速筛查模型判为阳性的新入院肝硬化腹水患者随机分为试验组(给予抗菌素治疗)与对照组(不予抗菌素治疗)。前瞻性比较两组的治疗应答率、病情变化及预后。结果:1.创建肝硬化数据库建立的肝硬化数据库包括患者基本信息界面、病史资料界面、主诉体检界面、实验室检查界面、影像学及内镜检查界面、内科治疗界面、干预治疗界面、病情评估及预后界面、并发症界面及出院随访界面,具有数据录入、数据自检、自动计算等功能,目前已录入568例住院肝硬化患者的临床数据。2.症状性细菌性腹水的病原菌及耐药特征共收集SB组103例和常规SBP组110例。SB组病原以革兰阳性菌为主(占55.3%,53/103),常规SBP组以革兰阴性菌为主(占71.8%,79/110),差异有统计学意义(c2=16.18,P0.01)。革兰阳性菌和革兰阴性菌对未加酶抑制剂的第三代头孢菌素的耐药分级,在SB组为安全和预警,在常规SBP组为安全和不宜经验用药。患者30天内死亡率在SB组为36.9%(38/103),在常规SBP组为43.6%(48/110),差异无统计学意义(c2=1.005,P=0.316)。3.血降钙素原与C反应蛋白筛查症状性细菌性腹水共纳入SB组30例与非感染性腹水组51例。PCT、CRP、两者串联、两者并联诊断SB的AUC依次为0.725、0.848、0.737、0.706,两两差异无统计学意义(P0.05)。诊断SB的最佳界值,PCT为≥0.43ng/m L,CRP为≥12.76mg/L。依据上述界值,PCT、CRP、两者串联、两者并联诊断SB的敏感性依次为70.0%、70.0%、53.3%、86.7%,特异性依次为76.5%、88.2%、94.10%、74.5%。4.症状性细菌性腹水快速筛查模型的构建与临床应用回顾性纳入SB组103例与非感染性腹水组204例,利用其临床数据构建了以体温、腹部压痛、血中性粒细胞百分比、血总胆红素、凝血酶原时间及腹水有核细胞计数为参数的SB快速筛查模型。模型诊断SB的AUC为0.939,最佳诊断界值为筛查评分≥0.328,其敏感性为86.4%,特异性为92.2%。前瞻性纳入试验组12例与对照组12例。试验组有6例全程使用头孢他啶,另6例因治疗反应不佳而调整抗菌素。至研究终点时,两组治疗应答率分别为66.7%(8/12)与16.7%(2/12),差异有统计学意义(Fisher精确概率法,P=0.036)。试验组中腹水PMN计数≥250个/mm3、腹水有核细胞计数≥500个/mm3、腹水培养阳性、确诊SB、发热且腹痛、腹部压痛或/和反跳痛以及肝性脑病的事件发生率均低于对照组,但差异无统计学意义(P0.05);两组均无死亡事件。结论:1.Epi Info~(TM7)软件构建肝硬化数据库切实可行,能有效协助临床医生进行临床科研。2.SB与常规SBP的病原菌构成及耐药特点不同,而预后相当。经验性初治,未加酶抑制剂的第三代头孢菌素可用于SB,但不宜用于常规SBP。3.血PCT、CRP及两者联合对SB均有良好的诊断价值且各有优势,临床可根据实际需要合理选用。4.SB快速筛查模型可有效识别SB病例,对该模型判为阳性的患者使用抗菌素可使其获益。
[Abstract]:Background and purpose: symptomatic bacterascites (SB) is a variant of the spontaneous bacterial peritonitis (spontaneous bacterial peritonitis, SBP), with infection, but the ascites multiform nucleus cells (polymorphonuclear, PMN) count 250 / mm3, only through the ascites bacteria culture positive diagnosis. Because of ascites culture often It takes time for several days, so it is difficult to find SB. in time and lack an effective method for early diagnosis of SB. The purpose of this study is to summarize the characteristics of SB etiology and explore the SB rapid screening model and its clinical application by establishing the database of liver cirrhosis. 1. To determine the contents of the database to be collected, then the Epi InfoTM 7 software was used to create the liver cirrhosis database and to enter the clinical data of the pathogenic bacteria and drug resistance of.2. symptomatic bacterial ascites. The positive SBP cases were screened from the created database and divided into the SB group and the routine SBP group (the ascites PMN count > 250 /mm3). The two groups of pathogens were compared. Bacterial composition, resistance characteristics and prognosis.3. serum calcitonin and C reactive protein screening symptomatic bacterial ascites from the established database to select the cases of liver cirrhosis in accordance with the requirements, divided into SB group and non infectious ascites group. The area under the curve of receiver operating characteristic, ROC (ROC) curve (area under curve). AUC) to evaluate the value of the value of PCT and CRP screening SB for the rapid screening of symptomatic bacterial ascites, the construction and clinical application of the cirrhosis cases were selected from the created database, divided into SB group and non infectious ascites group. The independent variable and two classified Logistic regression analysis were used to screen the independent variables and establish the rapid SB screening. Model. The new hospitalized cirrhotic ascites patients with SB rapid screening model were randomly divided into experimental group (antibiotic treatment) and control group (no antibiotic treatment). The treatment response rate, disease change and prognosis in two groups were compared prospectively. 1. the liver cirrhosis database established by the database of liver cirrhosis included the basic letter of the patient Interest interface, medical history data interface, medical examination interface, laboratory examination interface, imaging and endoscopic examination interface, internal medical treatment interface, intervention therapy interface, disease assessment and prognosis interface, complication interface and discharge follow-up interface, data entry, data self-examination, automatic calculation and so on, 568 cases of hospitalized liver cirrhosis have been recorded at present. The pathogenic bacteria and drug resistance of symptomatic bacterial ascites in.2. group were collected in 103 cases and 110 cases of group.SB were mainly Gram-positive bacteria (55.3%, 53/103) and Gram-negative bacteria (71.8%, 79/110) in the routine SBP group. The difference was statistically significant (c2=16.18, P0.01). Gram-positive and Gram-negative bacteria were found in the SBP group. The drug resistance classification of third generation cephalosporins without enzyme inhibitor was safe and early warning in group SB, safe and inexperienced in routine SBP group. The mortality rate in group SB was 36.9% (38/103) in 30 days and 43.6% (48/110) in routine SBP group (c2=1.005, P=0.316),.3. blood calcitonin and C reactive protein screening symptoms The sexual bacterial ascites were included in group SB 30 cases and non infectious ascites group 51 cases.PCT, CRP, both in series, the AUC in parallel diagnosis of SB was 0.725,0.848,0.737,0.706, 22 difference was not statistically significant (P0.05). The best boundary value of SB was diagnosed, and PCT was equal to 0.43ng/m L. The sensitivity of combined diagnostic SB was 70%, 70%, 53.3%, 86.7%, and the specificity of the specificity was 76.5%, 88.2%, 94.10%. The construction and clinical application of 74.5%.4. symptomatic bacterial ascites rapid screening model were included in 103 cases of SB group and non infectious ascites group in 204 cases, and the clinical data were used to construct body temperature, abdominal tenderness, and blood neutrophil. The SB rapid screening model of the ratio, blood total bilirubin, prothrombin time and the count of nuclear cells in ascites was a rapid screening model. The AUC of the model diagnosed SB was 0.939, the best diagnostic value was that the screening score was more than 0.328, the sensitivity was 86.4%, the specificity was 12 cases in the prospective trial group and 12 cases in the control group. 6 cases in the experimental group were used ceftazidime in the test group. At the end of the study, the response rates of the two groups were 66.7% (8/12) and 16.7% (2/12) at the end of the study. The difference was statistically significant (Fisher precision probability, P=0.036). The PMN count of ascites in the experimental group was more than 250 /mm3, the ascites had a nuclear cell count more than 500 /mm3, the ascites was cultured positive, confirmed SB, fever and abdomen. The incidence of pain, abdominal pressure pain, and / and anti jump pain and hepatic encephalopathy were all lower than those in the control group, but the difference was not statistically significant (P0.05). There was no death event in the two groups. Conclusion: the 1.Epi Info~ (TM7) software for the construction of liver cirrhosis database is feasible and can help clinical medical students to carry out the clinical scientific research of.2.SB and the common SBP pathogens and tolerance. The drug characteristics are different, and the prognosis is equal. Third generation cephalosporins without enzyme inhibitors can be used for SB, but it is not suitable for routine SBP.3. blood PCT. CRP and both have good diagnostic value and each have advantages. The rapid screening model of.4.SB can be used to identify SB cases effectively according to the actual needs, and the model can be effectively identified for this model. Patients with positive type can benefit from antibiotics.
【学位授予单位】：南昌大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：R572.2;R575.2

【相似文献】