【作者】 吴兴华；

【导师】 仇小强；

【作者基本信息】 广西医科大学 ， 流行病与卫生统计学， 2010， 博士

【摘要】 急性脑炎、脑膜炎是严重威胁人类的一类疾病,其中乙脑、流脑、人感染猪链菌病和脑炎型手足口病是近年来我国备受关注的病种。我国至今尚无该类疾病的整体发病资料,当前相关法定传染病的监测也存在一定缺陷。2007年WHO发布《乙脑监测手册》和《细菌性脑膜炎监测手册》,需要收集现场应用资料来支持手册的持续开发。鉴于此,本研究选择广西乙脑高发地区贵港市,应用WHO新版监测手册,首次将病毒性脑炎与细菌性脑膜炎合并开展症候群监测。研究共评估疑似病例1400余例,采集血液、脑脊液标本1400和1100余份,开展细菌学培养、乳胶凝集试验、RT-PCR和10种病毒ELISA试验等多种病原确诊检测,是迄今为止我国开展最大规模的急性脑炎脑膜炎症候群监测。本研究依照WHO新版监测手册制定各类病例定义；创新性的借助队列研究中资料收集方法与计算指标来评估《乙脑监测手册》中的关键病例定义;采用具有明确时间顺序的两段式流行病学调查,分别在病人入院和出院时完成；应用logistics回归、条件logistics回归和判别分析方法来分析细菌性脑膜炎与病毒性脑炎之间临床表现与实验室常规检查结果的特异性差别,以探讨临床上进行倾向性分类的可能性。通过本研究,可以描述当前各类急性脑炎、脑膜炎的发病强度和流行病学特征,为疾病的系统防控提供动态依据；现场应用急性脑炎脑膜炎症候群监测,为更大范围监测的推广提供参考；评估WHO新版《乙脑监测手册》,为手册提供现场依据和修订建议；本监测还可以探索急性脑炎脑膜炎症候群的病原谱,捕获新发传染病。目的：为了解各类急性脑炎、脑膜炎的发病强度、流行病学特征与临床特点,分析其病原谱构成、疾病负担和高危人群,评估当前相关疾病的临床诊断与传染病报告情况,为疾病系统防控提供依据。方法：在广西贵港市建立急性脑炎脑膜炎症候群监测系统,以临床症状作为评估病例的纳入标准,采集血液和脑脊液标本,完成出入院两次流行病学调查,开展细菌学培养、乳胶凝集试验、实时荧光PCR和ELISA等多种病原确诊试验,依照WHO《乙脑监测手册》和《细菌性脑膜炎监测手册》中病例定义将评估病例分为乙脑、细菌性脑炎、可能的病毒性脑炎等6个分类。应用logistics回归分析病毒性脑炎与细菌性脑膜炎病例在临床症状、体征和实验室常规检测结果等临床指征之间的特异性差异；应用Bayes判别模型拟合乙脑和细菌性脑膜炎的临床判别函数以探讨两者在临床上进行分类的可能。结果：监测期内共评估急性脑炎脑膜炎症候群病例1424例,采集血液、脑脊液标本1418和1100份。其中98份血液和85份脑脊液标本JE IgM ELISA检测阳性,239份血液标本为乙脑以外其它病毒IgM ELISA检测阳性,36份血液和34份脑脊液标本细菌学培养阳性,1份血液和8份脑脊液RT-PCR检测阳性。1424例症候群病例诊断分类为：乙脑103例、细菌性脑膜炎51例、排除病例127例、可能的病毒性脑炎1 88例、临床诊断细菌性脑膜炎病例189例和疑似脑炎、脑膜炎病例766例。估算相应的年发病率如下：乙脑0.8209／10万、细菌性脑膜炎年0.4065／10万、可能的病毒性脑炎3.7461／10万、临床诊断细菌性脑膜炎年1.5064／10万。细菌性脑膜炎中病原菌构成前三位为：新型隐球菌(17.65%)、肺炎链球菌(15.69%)和猪链球菌(13.73%)；病毒性病原检测的阳性标本构成前三位为：肠道病毒(30.89%)、腮腺炎病毒(27.39%)和乙脑病毒(23.89%)。80.52%的病原学确诊病例(其中乙脑：96.10%,细菌性脑膜炎：49.00%)发生在10岁以下儿童中。95.15%的乙脑病例发生在5、6、7月,细菌性脑膜炎病例全年散在分布。监测点在乙脑疫苗纳入计划免疫后发病率下降明显。常见脑膜炎致病菌对萘啶酸、磺胺甲基异噁唑和环丙沙星等抗生素耐药。logistics回归分析中,乙脑与细菌性脑膜炎病例间共有头痛、血常规白细胞计数和脑脊液葡萄糖等5个临床指征的差别具有统计学意义,发热、头痛、惊厥等7个变量最终进入模型。Bayes判别分析中“脑脊液葡萄糖含量”、“头痛”、“前囱膨隆”等7个变量进入判别分析模型,交互验证符合率为78.57%。监测点中乙脑与细菌性脑膜炎的临床误诊率分别为19.42%和15.69%,按照目前法定传染病报告程序,监测期间65.05%乙脑病例可能在法定传染病报告系统中漏报。本监测系统的阳性预测值为10.81%。结论：本研究以临床症状而不是临床诊断为病例筛检依据,配备快速诊断的实验室网络,结合主动监测为主的质量控制方法,初步具备了症候群监测系统的基本特征。监测采用严格的病例定义,确诊试验多级复核,确保病例诊断的准确性。急性脑炎、脑膜炎可造成较高的疾病负担。流脑已不再是细菌性脑膜炎的主要构成,新型隐球菌等条件致病菌成为细菌性脑膜炎的重要病原,肺炎链球菌、人感染猪链球菌病所致脑膜炎相对多发。乙脑是发病最多的病毒性脑炎之一,肠道病毒和腮腺炎病毒也是重要的病毒性脑炎构成。急性脑炎、脑膜炎主要危害10岁以下散居儿童,病毒性脑炎具有明显的季节性,细菌性脑膜炎的季节性则不明显。脑膜刺激征、血常规白细胞升高不再是细菌性脑膜炎的特征性临床表现。脑脊液标本诊断价值较高,细菌学培养是目前大规模监测的首选确诊方法,抗生素滥用是影响细菌学检出率的重要原因,常见脑膜炎致病菌存在耐药现象。疫苗是重要的保护因素,乙脑疫苗计划免疫工作在项目点显示了巨大的防控效果。乙脑、流脑等法定传染病的临床诊断并不可靠,存在一定的误诊率和漏报率,我国法定传染病报告系统有进一步提升的空间。症候群监测系统捕获病例的效力高于国家法定传染病报告系统,并能发现新发传染病,是我国传统疾病监测系统的必要补充。第二部分世界卫生组织乙脑监测手册的现场评估目的：评估WHO新版《乙脑监测手册》中关键病例定义和分类方法,分析乙脑的特异性临床指征和临床诊断可靠性,为手册提供现场依据和修订建议。方法：依托急性脑炎脑膜炎症候群监测项目,捕获AES病例和乙脑确诊病例,借助队列研究中RR和AR%等指标评估两者间是否存在统计学意义上的关联及关联的密切程度大小,估算AES定义对人群中乙脑病例的覆盖范围；计算乙脑相关各项临床指征的敏感度、特异度、约登指数和阳性预测值,以发现有筛检意义的临床表现；运用条件logistics回归分析分析乙脑特异性临床指征和临床诊断的可靠性。结果：研究期间共评估病例1424例,当中确诊乙脑103例。AES病例定义评估中,AES症候群对人群中乙脑病例的RR值为4.62,95%可信区间为2.80—7.63；AR%为78.35%,95%可信区间为64.25%886.89%。AES定义在人群中对乙脑病例的筛检敏感度为81%(95%可信区间：79%-83%),特异度为53%(95%可信区间：50%-55%),约登指数为0.33,kappa值为0.10,阳性、阴性预测值分别为13%和97%。全部评估病例中共有706例符合AES病例定义,按照《手册》分类方法分为乙脑确诊83例、未知病因的急性脑炎425例和其它病原体引起的急性脑炎198例。在多个临床表现的筛检价值综合评估中,“抽搐”得分最高,为17.62。假设的“AES+抽搐”病例对人群中乙脑病例的RR值为9.84,AR%值为89.83%。条件logistics回归分析中,有“囟门隆起”、“惊厥”、“脑膜刺激征”等7个变量最终进入模型,其似然比统计量G=144.07(x2=61.79,P=0.00)。判别分析中有“发热”、“意识障碍”、“抽搐”等6个变量最终进入模型,模型交互验证预测符合率为75.18%。结论：本研究中JE-IgM ELISA检测结果可靠。AES病例定义对人群中乙脑病例有统计学意义上的筛检作用, AES症候群与乙脑病例间存在较强程度的关联,开展AES症候群监测可以捕获了人群中大部分乙脑病例。“抽搐”具有良好的乙脑病例筛检价值,建议在新版《乙脑监测手册》中予以采用。手册中AES病例的分类方法在实践应用显得不够严谨,建议在新版手册中增加“细菌性脑膜炎”和“确诊其它疾病”等两个分类。“惊厥”是重要的临床筛检指征,“脑膜刺激征”、“发热”和“脑脊液外观”等3个临床指征也具有一定筛检价值,提示我们在现场工作中注意以上几种临床指征的识别。判别分析显示从临床表现和实验室常规检查结果来诊断乙脑病例并不可靠,建议对于乙脑的法定传染病报告应在报告制度上强调病原学依据。更多还原

【Abstract】 Acute encephalitis, meningitis are serious threat to human. Including encephalitis, meningitis, human infection with swine streptococcus disease and encephalitis-type hand foot and mouth disease is a major concern in recent years. China has no information on the incidence of such diseases as a whole; the current statutory surveillance of communicable diseases is also related to some defects. WHO published "JE Surveillance Manual" and "Bacterial Meningitis Surveillance Manual" in 2007, to be collected on-site application of information to support continued development of the manual.A bacterial meningitis and virus Encephalitis Joint surveillance was conduct in china firstly,1400 suspected cases were assessed, and more than 2000 CSF and serum samples were collected.In our study, mutli-pathogen diagnosis tests were carry out, Discriminant analysis and logistic regression were apply to data analysis. The Japanese Encephalitis Surveillance Standards was assessed by cohort study, some revisions were submitted to WHO expert committee.ChapterⅠStudy on Acute Meningitis, Encephalitis Syndrome Surveillance in Guigang City, GuangxiObjectives Apply WHO new surveillance standards to conduct acute meningitis, encephalitis syndrome surveillance for exploratory practice, describe disease burden and epidemiological characteristics of bacterial meningitis, Japanese encephalitis and other viral encephalitis, and provide the basis for the disease control strategy development.Methods Acute meningitis, encephalitis syndrome surveillance system was established in Guigang city. For the suspected cases, serum and CSF were collected,2-stage epidemiological investigation was conducted, and bacterial culture, latex agglutination test, real-time PCR and ELISA tests were carry out. All involved cases were identifying to 6 categories according to WHO case definition. Epidemiological characteristics and disease burden were described, for the purpose of risk factors analysis. Discriminant analysis and logistic regression were applied to find the Characteristic difference between bacterial meningitis and virus Encephalitis.Results 1424 suspected cases were evaluated in surveillance, Serum and/ or cerebrospinal fluid specimens in 1422 cases were collected. Laboratory testing results,98 serum and 85 CSF were positive for JE IgM ELISA test,239 serum were positive for other virus IgM ELISA test. Bacteria were culture out in 36 bloods and 34 CSF.1 copy of blood and 8 copy of CSF were positive for real-time PCR detection. For the 3 common meningitis pathogens, antibiotic resistance was observed in Nalidixic acid, Sulfamethoxazole and Ciprofloxacin. According WHO standards, there are 103,51,127,188,189 and 766 cases were confirmed for JE, bacterial meningitis, excluded cases, possible encephalitis, Clinical diagnosis of bacterial meningitis and suspected encephalitis, meningitis cases respectively, yield the Estimated annual incidence of 0.8209/100,000, 0.4065/100,000,3.7461/100,000,1.5064/100,000 for JE, bacterial meningitis, possible encephalitis, Clinical diagnosis of bacterial meningitis respectively. 95.15% of the JE and 50.00% of possible encephalitis cases was observed occurred in May, June and July, and 48.15% of the clinical diagnosis of bacterial meningitis cases was observed occurred in May, June, July and August. The highest incidence on 4 kinds of AMES was observed in Guiping city.96.10% JE cases were found in children under the age of 10, of which 73.80% of children under age of 5, bacterial meningitis cases have a distribution of ages. For JE cases, high frequencies of clinical manifestations are following:fever (98.10%), meningeal irritation sign (97.09%), convulsion (82.52%), consciousness change (77.67%) and fever (65.05%). In multi-factor analysis, "headache", "diarrhea", "bulging anterior fontanel", "blood white blood cell count" and "CSF glucose" etc 5 clinical indications have statistically significant difference between viral encephalitis and bacterial meningitis. "CSF glucose content," "headache," "Bulging anterior fontanel" and other six variables into the discriminant analysis model, "convulsions" has the biggest difference between the coefficients of discriminant function. According to the current legal infectious disease reporting system,65.05% Japanese encephalitis cases would be omitted in project site. A positive predictive value of 10.81% was observed in this syndrome surveillance system.Conclusions Our surveillance base on clinical manifestations rather than clinical diagnosis to capture suspected cases, equipped rapid mutli laboratory-confirmed test, combined with active surveillance-based as quality control measures, initially with a basic feature of syndrome surveillance system. Antibiotic resistance is Common for 3 major meningitis, vaccine is an important protective factor, JE vaccine immunization program shows the enormous effect in project site. This study show, the CSF has the high diagnostic value for pathogen confirm, bacteriological culture is still the preferred diagnostic method, antibiotic abuse is affecting factor of bacteriological detection. JE is one of the largest numbers of viral encephalitis; Coxsackie virus, ECHO virus and mumps virus are also important composition of viral encephalitis pathogen. Epidemic cerebrospinal meningitis is no longer the main component of bacterial meningitis, Cryptococcus neoformans and other conditions of bacterial meningitis pathogenic bacteria to become an important pathogen, Streptococcus Pneumonia, People infected with meningitis caused by Streptococcus Suis disease are relative multiple. Clinical diagnosis is not reliable, legal infectious diseases report system in China have further room for improvement. Syndrome surveillance is far higher than the national infectious disease reporting system in case capture ability, and captured the emerging infectious diseases in project site, is a necessary complement of traditional monitoring systems.Chapter II On-site Assessment of WHO Japanese Encephalitis Surveillance StandardsObjectives Assess the new edition of WHO Japanese Encephalitis Surveillance Standards base on syndrome surveillance data, to provide on-site evidence and suggestion for the improvement.Methods Base on syndrome surveillance data, categorize the AES case according to the new WHO standards. A cohort study was applied in study to estimate the AES definition in standard. The sensitivity, specificity, Youden index and the positive predictive value of AES components were calculated for the purpose of finding clinical significance with screening value. Discriminant analysis and logistic regression were applied to data analysis.Results 1424 suspected cases were reported in syndrome surveillance, and 1396 case with ELISA result, of which 109 positive cases were detected. According to "standards" classification, a total of 706 cases in line with AES case definition, was categorized into 83 cases of JE,425 cases of AES unknown and 198 cases of AES other agent. In cohort study, a relative risk of 4.62 (95% CI:2.80-7.63) and percentage of attributable risk of 78.35%(95%CI:64.25%-86.89%) were observed. The screening parameters are following:sensitivity 81%(95% confidence interval:79%-83%), specificity 53%(95% confidence interval:50%-55%), Youden index 0.33, kappa value of 0.10, positive and negative predictive values were 13% and 97%. In multi-factor analysis, the "fontanel uplift," "convulsions," "meningeal irritation" and so on seven variables entered the final model, the likelihood ratio statistic G= 144.07 (χ2=61.79, P=0.00). In discriminant analysis, the "fever", "disturbance of consciousness," "convulsions" and other six variables and eventually enter the model, with cross-validation rate of 75.18%.Conclusions AES definition for JE was statistically significant effects on the screening, there is strong correlation strength was observed in study, AES syndrome can cover most of the Japanese encephalitis cases. "Convulsions" with quite screening value, was recommended involved in the new version of the WHO standards. Classification in JE surveillance standards is not stringent enough in practical application, it is recommended to add classification of "bacterial meningitis" and "diagnosed with other diseases," two categories. The "seizure" is an important clinical screening indicator in field, "meningeal irritation," "Fever" and "appearance of CSF" and three clinical indications also has certain screening value. Discriminant analysis shows that, it’s not reliable for diagnosis to JE base on routine clinical and laboratory examination, and only laboratory testing is sufficient basis for JE case report.更多还原