【作者】 刘庆华；

【导师】 何礼贤； 胡必杰； 王葆青； 李华茵；

【作者基本信息】 复旦大学 ， 内科学， 2005， 博士

【摘要】 研究背景:呼吸机相关肺炎(ventilator-associated pneumonia,VAP)是最为常见的院内获得性肺炎(noscomial pneumonia,NP),发病率和死亡率高,预后差,在老年人发病率和死亡率更高,预后更差,预防极其关键。目前相对缺少大规模、多中心、前瞻性、随机对照临床研究,针对老年人VAP预防这一特殊群体的研究也较少,相关动物实验研究也不甚成熟。 目的:评价综合性干预措施对老年人VAP预防价值,初步探讨VAP发病机制。 方法:临床部分,采用前瞻性、多中心、随机对照设计,对>60y接受人工气道机械通气>48h患者给予综合性干预措施(半卧位+持续声门下分泌物引流+促胃动力药物莫沙必利),以普通气管导管组作对照,记录基线资料,监测胃液、声门下分泌物和下呼吸道菌群,比较两组住院时间、ICU内住院时间、机械通气时间、VAP发病率、死亡率、归因死亡率、相对危险度和VAP患者不同部位菌群与VAP致病菌一致率,对部分患者胃液内注入美蓝,观察比较两组下呼吸道美蓝发现率以反映胃液反流情况。对中山医院两组VAP患者按照时间先后顺序在胃液和/或声门下分泌物以及下呼吸道监测到的相同菌株,应用全基因组DNA稀有位点限制性内切酶酶切脉冲电场电泳(genome macrorestriction—pulsed field gel electrophoresis,GM—PFGE)方法,鉴定其同源性,并比较同源性发生率差异。动物试验,首先建立VAP动物模型,机械通气过程中监测胃液菌群和血气情况,处死动物测定肺脏湿重,对肺组织匀浆需氧培养、病理检查和血需氧培养,部分动物行胸部CT检查。VAP动物模型建立后,应用枸橼酸莫沙必利或易蒙停干预胃动力,监测不同时间点口咽部、大气道菌群以及血需氧培养,观察胃动力改变对消化道菌群移位变迁影响和对不同菌群所致VAP发生率的影响。 结果:(1)临床研究中入组86例,综合干预组(I)41例,普通气管导管对照组(C)45例;C组住院时间、ICU内住院时间、机械通气天数均长于I组,总体VAP发病率为51.2%,I组为34.1%,C组为66.7%,其中I组和C组早发性VAP发病率(early-onset VAP,EOP)分别为4.9%、20.0%,两组VAP发病率和EOP发病率有统计学差异,I组VAP患者死亡率、归因死亡率均高于C组,无统计学差异(p>0.05)。相对危险度(RR)为0.51(95%CI 0.164～0.634,p<0.05)。VAP致病菌和消化道菌群总体一致率为28.9%,I组和C组分别为14.6%和35.6%(p<0.05),一致性菌群以GNB主,主要发生于晚发性VAP(Late-onset VAP,LOP)更多还原

【Abstract】 Background: Ventilator-associated pneumonia (VAP) is one of the most common noscomial pneumonia (NP) and had serious influence on the outcome. Incidence and mortality of VAP are considerably high, especially in the elderly patients. Prevention becomes very critical. There are relative short of multi-center, prospective, randomized design, including large population participated clinical research of VAP as well as prevention measures in the elderly patients. The animal experiments are also not mature.Objective: To assess the value of comprehensive prevention measure to VAP of patients more than 60ys and investigate the pathogenesis of VAP.Methods: Patients more than 60ys with expected mechanical ventilation >48h were randomly assigned to control group receiving a standard endotracheal tube or comprehensive intervention group being given semi-recumbent, mosapride citrate, and endotracheal tube for continuous subglottic secretions drainage. The characteristics of cases were record when they enrolled the clinical trial and bacterial culture of gastric juice, material from subglottic drainage and lower respiratory tract (LRT) were taken. Compare the duration of mechanical ventilation, length of ICU/ hospital stay, the incidence, mortality, contributable mortality, relative risk as well as the bacterial concordance in gastric juice or LRT between two groups. Some cases were given methylene blue through the gastric tube and the time, rates of methylene blue found in LRT under microscope were observed to illustrate the aspiration. Pulsed-field gel electrophoresis of DNA macro-restriction fragments was used to identify the relatedness of the pathogen from different part and analyzed the incidence rates of homology in two groups. In the animal 1 experiment, the VAP model established first. The bacterium of gastric fluid and blood gas were monitored during ventilation. Wet weight of the lung, blood stream and lung homogenized were both cultured after the rabbits sacrificed. Some animals were given computerized tomography (CT) of thechest before death. Then mosapride citrate or loperamide were administered and the flora in oropharynx and airway were monitored in different time.Results: (1) Eighty-six cases enrolled the clinical trial, including 41 in I group and 45 in C group. The duration of mechanical ventilation, length of ICU/hospital stay in I group were all less than C group. The incidence of VAP among whole cases was 51. 2%,34.1% in I group and 66. 7 % inC(p<0. 05). Two (4. 9%)cases ingroup I get EOP while 9 cases (20. 0 %) in group C(p<0. 05). The mortality, attributable mortality in group C was higher (p>0. 05). The RR was 0. 51 (95% CI 0.164—0. 634). The same bacterium type (28. 9%) predominantly are G- rod, commonly happened in LOP. Methylene blue in LRT or oropharynx under microscope being detected later in group I than C. (2) Fifteen cases including 35 strains (G—29 strains, G+ 6 strains) enrolled in PFGE. Twenty-nine identical strains mainly G-rod, were found in 12 cases including 4 of cross-infection. The ratio of same strains among LRT, gastric juice or the drain from subglottic were 11.9%, and 19.4 in group C, 3.2% in group I (p<0. 05). (3) In VAP model established experiment, different degree of inflammation were found in both MV and BMV group and much more serious in BMV group. Ratio of positive lung culture (bacterium concentration more than 10’cfu/g) was 40.09^ 81.8% separately in MV and BMV, and lung infection caused by inoculated bacterium was 9. 1%. The WW/BW and score of the lung were different (p<0. 05). CT shows density in multi-lobe of two lungs patchy increased, mainly distributed in lower and middle lobe. Positive blood stream was 45.5% in MV. In 37 animals 94.4% have bacterial colonization in oropharynx and 83.3% in airway, predominately Bord. bronchi sept ica. Ratio of lung infection caused by inoculate bacteria was 10.8%. Four negative colonization in upper airway also negative culture of lung. No significant variance about colonization was seen between I and P group, while the semi-quantitative concentration of inoculated bacterium in I was higher than in P. Although no significance difference of the VAP whole incidence found, VAP caused by inoculated bacterium in I was much higher than in P (15. 8% vs. 5.5%, p<0. 05). About 45. 9% in 37 cases, 63. 2% in inoculated更多还原