Risk Factors And Predictive Scoring System For Multiple Bronchoscope Adjuvant Therapy In Children With Refractory Mycoplasma Pneumoniae Pneumonia

Objectives: To investigate the risk factors of multiple bronchoscopic adjuvant therapy in children with refractory Mycoplasma pneumoniae pneumonia(RMPP)and generate a predicted score using the clinical records of RMPP patients Methods :Retrospective analysis of the clinical data of 332 RMPP children treated by Department of Respiratory Diseases in Children’s Hospital of Soochow University from January 2011 to December 2016.Based on the number of bronchoscope,the patients were divided into the single group and the multiple group(≥2 times).Comparison was made between the above two groups in age,sex,clinical,laboratory characteristics.Multivariate logistic regression was performed to identify the risk factors of multiple bronchoscope adjuvant therapy by the comparisons of single factor analysis showed P<0.05.Based on the β value of significant risk factors,we create a predictied score.The predictive scoring system for multiple bronchoscope adjuvant therapy in children with RMPP was evaluated by ROC curve,paired chi-square test and consistency test.Results: 1.Conforming to the selected standard was a total of 332 cases,According to the number of bronchoscopy,the patients were divided into single group(n = 223)and multiple group(n = 109).The male / female were 0.94: 1 and 0.85: 1,The average age was 69.7 ± 36.7 months,72.0 ± 33.7 months,the difference was not statistically significant(P = 0.66,0.58,respectively).2.Children undergoing multiple bronchoscopy had the fever durantion before the bronchoscopy≥10 days and course of disease before the bronchoscopy≥10 days higher than in the single bronchoscopy group(118(52.9%)vs 71(65.1%),69(29.6%)VS45(41.3%)),the difference was statistically significant(P <0.05).At the same time,the use of glucocorticoid,macrolide,glucocorticoid combined with macrolide antibiotics in single group and multiple groups in the first week of illness was significantly lower in multiple groups than in single group,the differences were statistically significant(P <0.05).In multiple groups,the percentage of neutrophils(N),CRP,CRP> 44 mg / L,LDH,LDH> 480 U / L,DD and Fib were higher than those in the single group and the levels of CD3 +,CD4 + and CD4 / CD8 were lower than those of the single group,with statistical significance(P <0.05).In addition,the multiple group of mixed infection and pleural effusion were higher than that of the single group.The proportion of in multiple groups was higher than that of the single group in bronchoscopy.In bronchoscopy,the mucus plug blocking and mucosal erosion were more than that of the single group(P <0.05).3.Logistic regression analysis was performed on the possible risk factors(P <0.05)in the multiple group and single group observation,showed that fever durantion before the bronchoscopy≥10 days(P=0.000,OR=19.504),the unused of macrolide antibiotics in the first week of illness(P=0.000,OR=5.072),the unused of glucocorticoid in the first week of illness(P=0.000,OR=14.051),C-reactive Protein(CRP)>44mg/L(P=0.004,OR=2.638),lactate dehydrogenase(LDH)> 480U/L(P=0.004,OR=2.326)and mucosal erosion(P=0.002,OR=11.15)for multiple bronchoscopic procedures in children with RMPP.The area under the receiver operator characteristic curve of the regression model is 0.817,P<0.001,with middle-grade diagnostic accuracy,73.8% sensitivity and 75% specificity.4.Based on the β value of significant risk factors,we assigned 1 point for C-reactive Protein(CRP)>44mg/L and lactate dehydrogenase(LDH)>480U/L,2 points for the unused of macrolide antibiotics in the first week of illness,,3 points for fever durantion before the bronchoscopy≥10 days,the unused of glucocorticoid in the first week of illness and mucosal erosion.The predictive scoring system is 13 points and the prediction value is 5 points.ROC curve area is 0.803,and the evaluation results are as follows: sensitivity(0.927),specificity of(0.614),the total coincidence rate(0.717),the positive likelihood ratio(2.4),negative likelihood ratio(0.059),Kappa value(0.457).Conclusions: 1.Severe inflammation,immune imbalance,imaging changes,mixed infections and whether or not to actively resist infection and inflammation in the early stage,were important risk factors for multiple bronchoscopic procedures.However,there are still no specific and sensitive indicators to determine the timing of bronchoscopy,which still needs further observation and research.2.When the children with RMPP with the factors:fever durantion before the bronchoscopy≥10 days,the unused of macrolide antibiotics in the first week of illness,the unused of glucocorticoid in the first week of illness,C-reactive Protein(CRP)>44mg/L,lactate dehydrogenase(LDH)>480U/L.,it may indicate the multiple bronchoscope adjuvant therapy,but its exact application value still needs to be further studied.3.The predictive scoring system for multiple bronchoscope adjuvant therapy in children with RMPP,Which fever durantion before the bronchoscopy≥10 days,the unused of glucocorticoid in the first week of illness,mucosal erosion heralded a higher risk of multiple bronchoscopy.The evaluation system combined with the relevant evaluation of the scoring system is better,with a certain degree of discrimination and accuracy,there is a certain clinical relevance.However,due to the single-center,retrospective study and the limited inclusion factors,there is still some discrepancy between the evaluation results and the actual results,and no prediction is made in the external population.The scoring system needs to be further improved in future work.