| Background:The severe community-acquired pneumonia causes heavy social burden due to its high morbidities and mortalities.One of the main reason for the poor prognosis of this disease is the difficulty of early target anti-pathogen treatment due to uncertain etiology diagnosis.What's more,the etiology changes with the time going and differes in various countries and districts.Therefore,it is of enomous significance to establish a real-time microbes screening platfonn to find the microorganism and guide the medicine treatment in clinical practice.Ho we ver.few medical institutions have conditions to conduct the pathogen screening detection.So it is greatly necessary to establish a etiology predict model to differentiate virus from bacteria so as to guide initial effective target anti-pathogen treatment.Finally,since the clinical manifestation of severe community-acquired pneumonia was extremely complex,the establishment of prognosis analysis model will help to predict the prognosis of the patients.Objective:To establish a real-time etiology screening platform for severe community-acquired pneumonia and thereafter to explore the distribution and the law of various pathogens prevalent;to establish a model to differentiate viral infection from bacteria infection by comparing the clinical characteristic of each microorganism-induced respiratory infection;to establish a model for prognosis analysis in order to find the independent risk factors for mortality.Methods:1.Research objects:88 patients were recruited from 533 cases with community-acquired pneumonia in Fujian Provincial Hospital from April 2012 to December 2015,based on the diagnostic criteria of severe community-aquired pneumonis.Thereafter,the etiology and clinical data of these patients were studied prospectively.2.Laboratory detection:Loop-mediated isothermal amplification?LAMP?was used to dectect the samples of lower repiratory tract of the patients for Streptococcus pneumonia?SP?,Staphylococcus aureus?SA?,Methicillin-resistant Staphylococcus aureus?MRSA?,Escherichia coli?E.coli?,Klebsiella pneumonia?KP?,Pseudomonas aeruginosa?PA?Acinetobacter baumannii?Ab?,Stenotrophomonas maltophilia?SM?,Haemophilus influenza?HA?,Legionella pneumophila?LP?,Mycoplasma pneumonia?MP?,Chlamydia pneumonia?CP?,Mycobacterium tuberculosis complex?MTB?,the DNA of positive ones were sequenced or validated by microorganism culture for bacteria,by real-time polymerase chain reaction?PCR?for Mycoplasma pneumonia or Chlamydia pneumonia,by positive urinary antigen or four fold raise of serum antibodies in recovery stage than in acute stage for Legionella pneumophila.Luminex Liquid chip x-tag method was used to dectect the samples of lower repiratory tract of the patients for Influenza A,HIN1,Seasonal influenza H1N1 and H3N2,influenza B virus,Respiratory syscytial virus A and B?RSV?,Human metapneumovirus?HMPV?,Adenovirus?ADV?,Rhinovirus/enterovirus,Parainfluenza virus?PIV 1,2,3,4?,Coronavirus 229E,Coronavirus OC43,Coronavirus HKU1,Coronavirus NL63;real-time polymerase chain reaction?PCR?was also used to detect human cytomegalovirus?HCMV?,Adenovirus?ADV?,parainfluenza virus?PIVI,II,III?Respiratory syscytial virus?RSV?,SARS coronavirus,A/H1N1,Human high pathogenic avian influenza virus?H5N 1,H7N9?,Human metapneumovirus?HMPV?,Human Bocavirus?HBV?.Human high pathogenic avian influenza virus H5N1 and H7N9 were detected by Fujian Provincial Provincial Center for DiseaseControl.According to the etiology,the patients were devided into 5 groups:virus group?22 cases?,bacteria group?17 cases?,combination of virus and bacteria group?11 cases?,negative group?30 cases?,and others?3 fungi and 5 atypical pathogens?.3.Clinical data of the patients:?1?Basical data of the population:age,sex,smoking,comorbidity such as chronic obstructive pulmonary disease?COPD?,coronary artery disease,cardia dysfunction,cerebrovascular disease;diabetes,renal dysfunction,liver diseases,tumors except lung cacer,aspiration risky diseases such as cerebral apoplexy and nasopharyngeal carcinoma having received radiotherapy treatment,immunosuppressed situation which means having 3 months immunosuppressors treatment or 10 mg daily prednisone treatment?or other steroids treatment?for over 3 months,antibiotics treatment before admittion.?2?Clinical symptoms including fever,fear of cold and chills,cough,expectoration,fatigue,muscular soreness,chest pain,altered mental status;headache,dyspnea,gastrointestinal symptom?vomiting,nausea,diarrhea?,other symptoms?pharyngalgia,arthrodynia,myosalgia and edema?;?3?Clinical signs including temperature,repiratory rates,heart rates,consciousness,moise rales and dry rales.?4?Laboratory examination:routine blood test including white blood cell?WBC?,neutrophils?N?,the percentage of neutrophile granulocytes?n%?,leukomonocyte,the percentage of leukomonocyte,hemoglobin?HGB?,hemoglobin and hematocrit?HCT?and platelet count?PLT?;inflation index including procalcitonin?PCT?,erythrocyte sedimentation rate?ESR?,C reation protein?CRP?and neutrophil alkaline phosphatase?NAP?;biochemistry results including albumin?ALB?,glutamic-oxalacetic transaminase?AST?,glutamic-pyruvic transaminase?ALT?,creatine kinase?CK?,usea nitrogen?BUN?,creatinine?Cr?,blood potassium concentration and blood sodium concentration,coagulation function including prothrombin time?PT?,partial prothrombin time?APTT?,fibrinogen?FDP?,D-dimer?D-D?;immunologic function including the percent or counts of CD3,CD4,NK,CD 19 and serum IgG,IgM,IgA,complements C3/C4;blood gas analysis?pH,arterial partial pressure of oxygen?PaO2?,arterial partial pressure of carbon dioxide?PaCO2?,bicarbonate ion?HCO3-?,buffuer excess?BE?,oxygenation index?PaO2/FiO2?;Chest image?x-ray or chest CT scan?.?5?Treatment including respiratory support model?oxygen therapy,noninvasive mechanical ventilation,invasitive mechanical ventilation?and medicine use?antibiotics,antifungi drugs,antivirus drugs?,?6?Hospital day and treatment outcome?survival or death?.?7?Evaluation of the severity including PSI score,CURB-65,APACHE-II and Chalison score and index.4.Statistical methods:Factor analysis of variance was used to compare measurement data with normal distribution of more than two groups;a non-parametric test called Kruskal-Wallis H test was used to compare measurement data without normal distribution of more than two groups Mann-Whitney U was used to compare those data between every two groups;Pearson Chi-Square ? 2 test was used to compare the enumeration data.Then discriminant analysis was applied and a predict model was established to differentiate the viral and bacterial pneumonia.T-test was used to screen risk factors of death for the patient with SCAP before logistic multivariate regression was used to explore independent risk factors.Results:1.The etiology of 88 patients with SCAP in Fujian Provincial Hospital.In this study,the microbes of 58?58/88,65.90%?cases of were found,among which 25%?22/88?were virus,19%?17/88?were bacteria,13%?11/88?were combination of virus and bacteria,3,375%?3/88?were fungi,?1 penicillium marneffei,1 a shahi spores,1 aspergillus?,5.725%?5/88?were atypical pathogens.In terms of bacteria,the etiology of 32.95%?28/88?of the cases were found,among which 41 bacteria were identified,21/41 were single pathogen,7/41 were at least two pathogens.Of all the pathogenic bacteria,klebsiella pneumoniae accouting for 29%?12/41?,staphylococcus aureus ranked the second?17%,7/41?and then pseudomonas aeruginosa?6/41,14.6%?,and a few acinetobacter baumannii,streptococcus pneumoniae,haemophilus influenzae,stenotrophomonas maltophilia as well.In terms of virus,the etiology of37.5%?33/88?of the cases were found,among which 43 virus were identified,26/43 were single pathogens,7/43 were at least two kinds of virus.Of all the virus,HN virus accouting for 49%?21/43?,including 14 seasonal influenza A virus,1 influenza B virus,1 parainfluenza virus II,2 parainfluenza virus III virus and 3 human high pathogenic avian influenza virus H7N9;Human cytomegalovirus?HCMV?ranked the second?11/43,25%?,and then rhinovirus?3/43?,adenovirus?ADV??2/43?,respiratory syscytial virus?RSV??2/43?,human metapneumovirus?HMPV??2/43?and human bocavirus?HBV??2/43?.2.Baseline data of the population:The average age of the patients was?58.95±16.98?year,ranging from 21?96 years old.69.3%?61/88?of them were male,30.7%?27/88?were female.The average oxygen index of the patients was?200.60±69.23?mmHg,including 40 mild Acute respiratory distress syndrome?ARDS?,of which PaO2/FiO2 was 200?300 mmHg,45 moderate ARDS,of which PaO2/FiO2 was 100?200 mmHg and 3 severe ARDS,of which PaO2/FiO2 was lower than 100mmHg.The average score of CURB-65,Pneumonia severity index?PSI?,APACHE-II,Chalison's score,Chalison's Index were?1.17±1.02?score,?85.52±37.39?score,?11.12±5.09?score,?4.20±2.687?score and?0.485±0.405?respectively.The average hospital stay was?15.20±9.14?days.29.5%?26/88?of the patient smoked,60.2%?53/88?encountered repeated fever,70.5%?62/88?received antibiotics treatment before admittion.34.1%?30/88?of them suffered from hypertention,14.8%?13/88?received immunesupreesor treatment,12.5%?11/88?had diabetes,12.5%?11/88?had tumors,12.5%?11/88?had chronic lung diseases like COPD,only a few had brain,heat,liver and renal diseases or aspiration risky diseases.3.The comparision of clinical features between viral and bacterial pneumonia and the establishment of etiology predict modelThe clinical features between viral and bacterial pneumonia were compared,as a result,no differences were found in age,sex,smoking,basical diseases,clinical symptoms and signs and the severity of pneumonia.In the prospective of laboratory examination,the counts of lymphocyte,PCT,PT and APTT level showed different in four groups?Kruskal-Wallis H=10.645,P=0.014;Kruskal-Wallis H=8.961,P=0.03;Kruskal-Wallis H=13.359,P=0.004;Kruskal-Wallis H=10.191,P=0.017?,the counts of lymphocyte,PT and APTT level were much higher in bacteria group than those in virus group?Mann-Whitney U=74,P=0.003;Mann-Whitney U=83.5,P=0.003;Mann-Whitney U=90,P=0.005?;PCT levels were much lower in virus group than bacteria group?Mann-Whitney U=69,P=0.007?and negative group?Mann-Whitney U=139.5,P=0.027?when every two groups comparision were done,while no differences were found in other blood tests.Chest images were compared between viral groups?including single virus group and combination group?and nonviral group?including bacteria and negative group?by peason ?2 test,interlobular septal thickening or subpleural reticular was found more popular in viral group than in nonviral group??2=4.129,P=0.042,OR=2.726,95%CI:1.021?7.279?,while no differences of other changes were found.In addition,stepwise discriminant analysis was used to screen variable and a canonical discriminant function F=0.527X3+0.101X4-1.10X8-0.848 was established.This meaned that the viral or bacterial pneumonia could be differentiated according to the counts of lymphocyte?X3?,PCT?X4?level and interlobular septal thickening or subpleural reticular?X8?,the average of discriminant score of bacteria was 1.067,while the score of virus was-0.762,so general average score?z?was 0.305.If z>0.305,it may be bacterial infection.If z<0.305,it may be viral infection.Finally,data generation was done using the previous material to confirm its validity,the total accuracy rate of was 77.8%,with 90.5%coincidence rate for virus and 60%coincidence rate for bacteria.4.The treatment of these populationFirstly,the anti-pathogen treatment of 88 patients with SCAP could be described as follows:4.5%of which received extensive treatment including antibiotics,antifungal,and antiviral treatment.Of all the cases,99%received antibiotics treatment,18.18%?16/88?of which received 1 antibiotics treatment,42.05%?37/88?received two kinds,39.78%?35/88?no less than 3 categories.Of all the used antibiotics,P-lactams were mostly widely used?77/88,87.5%?,including 59%cephalosporin,56.8%carbon penicillin;quinolones were also commonly used?47/88,53.41%?,antibiotics against methicillin-resistant staphylococcus aureus?MRSA?such as vancomycin,teicoplanin and linezolid were not as often used as these two antibiotics?33/88,37.5%?.On the other hand,26.1%?23/88?of the patients received antiviral treatment,mainly using oseltamivir?17/88?and nucleoside antiviral drugs?7/88?,seldom using ribavirin treatment?3/88?,only 4 receiving conbine antiviral treatment.Secondly,about 40%of cases used more than 3 kinds of antibitics while no different numbers of kinds of antibitics were found in four groups.Antibiotics against MRSA were likely to be applied more in combinated infection of bacteria and virus?P<0.05?,while no other differences were found in other medicine treatment.Finally,in forms of respiratory support models,63.6%of cases used noninvasive positivepressure ventilation,27.3%used oxygen treatment,only 9.09%received invasive mechanical ventilation.5.Prognosis analysis of these populationFirsly,The mortality showed no difference in four groups with different etiologies?P>0.05?.Antibiotics treatment before adimittion increased death risk?P=0.029,OR=4.909,95%CI:1.049?22.973?.Hospital stay was defined as the duration from admittion to discharge or death,the hospital stay of the death group was significantly shorter than the survival[?7.20±7.46?vs?17.62±8.20??t=5.078,P<0.05?].Secondly,the oxygen index and CURB-65 showed no differences between two groups,while the APACH-II,PSI and chalison's score was higher in the death group than those in the survival group[?15.20±4.87?vs?9.84±4.70?,t=4.41,P=0.00?];[?110.60±40.27?vs?77.90±33.21?,t=3.305,P=0.003)];[?5.85±2.68?vs?3.80±2.50?,t=3.15,P=0.002)].In terms of clinical symptoms and signs,fatigue turned to more often in the death than the survival?5 5%vs3 0.9%,?2=3.885,P=0.049?,other symptoms like myalgia,arthralgia,edema were also more polular in death group?70%vs 38.2%,?2=6.289,P=0.012?.The morbiditie of septicopyemia was much higher in the death group?45%vs7.4%,?2=13.68,P=0.00?.In terms of laboratory examination,the PCT,and lactate level were higher in the death group?Mann-Whitney U=434.5,P=0.025?;?Mann-Whitney U=152,P=0.03?,so were the serum BUN,blood glucose[?9.15±6.61?vs5.86±2.83),t=2.534,P=0.019);?7.77±3.31?vs?6.15±2.30?,r=2.052,P=0.051)].As far as the treatment were concerned,antiviral treatment was much frequently used in the survival than the death?5%vs32.35%,?2=5.989,P=0.01?and no differences were found in other medicine including antifungal drugs,antibiotics and steroids.In addition,the respiratory support models differed in two group,invasive mechanision ventilation was more common than oxygen treatment??2=12.765,P=0.000?and noninvasive positivepressure ventilation??2=14.892,P=0.000?in the death group.Finally,8 indexes including the risk factors screened by the above analysis and the factors regarded as much significant in clinical practice were recruited into logistics regression equations by forward stepwise variable selection method,and three factors including higher lactate and the serum BUN and higher heart rates were found to be independent death risk factors.[OR value:4.704?95%CI:0.966?22.907?,1.264?95%CI:0.994?1.606?,1.081?95%CI:1.003?1.165?].Meanwhile,no invasive mechanical ventilation was determined as a protective factor.?OR=0033,95%CI:0.001?0.764??Conclusion:This study established a real-time pathogen-detection platform,screened all the possible microoriganisms for the 88 patients with SCAP.Then the etiology of SCAP in Fujian Province was firstly described and a predict model for the differentiation of virus and bacteria was established,which showed the lower counts of lymphocyte,lower PCT level and interlobular septal thickening or subpleural reticular may be effective preditors of virus infection.Lastly,the associate risk factors of prognosis were analyzed,which suggested higher lactate and the serum BUN,higher heart rates and invasive mechanical ventilation were independent risk factors of death. |
PDF Full Text Request