Establishment And Validation Of A Nomogram For Predicting The Occurrence Of Mycoplasma Pneumoniae Pneumoniae In Adults

Objective:The predictors of mycoplasma pneumoniae pneumonia(MPP)in adults were explored,and a nomogram model was established to predict the occurrence of MPP in adults,so as to quickly identify high-risk groups of MPP,carry out clinical intervention as soon as possible,and improve the prognosis of patients.Methods:We retrospectively analyzed the clinical data of 337 adult patients with communityacquired pneumonia(CAP)admitted to Shanxi Bethune Hospital from January 1,2021 to December 31,2021,and divided them into MPP groups and non-MPP groups according to whether they were infected with mycoplasma pneumoniae(MP).General information,clinical manifestations,laboratory tests and imaging results of the two groups were compared.Univariate analysis and multivariate logistic regression analysis were used to screen the predictors of MPP in adults.The nomogram model for predicting the occurrence of MPP in adults was established by R software,and the model was verified internally by Bootstrap self-sampling method.Finally,the prediction performance of the model was evaluated.Results:1.In the end,337 patients were enrolled in this study,and all patients were randomly divided into training set(236 patients)and validation set(101 patients)in a 7:3ratio.Comparing the baseline characteristics of the training set and the validation set study population,there was no significant difference in 35 clinical variables between the two groups.2.Univariate analysis results showed that the gender,age,season of onset,fever,duration of fever body temperature,dry cough,pharyngeal malaise,chest tightness and shortness of breath,aversion to cold,chills,fatigue,dizziness,CRP,D-dimer,tree-in-bud sign and pleural effusion were statistically significant between the MPP group and the non-MPP group(P<0.05).3.The results of multivariate logistic regression analysis showed that age(OR=0.950,95%CI: 0.932～0.969,P<0.001),body temperature(OR=1.658,95%CI:1.250～2.199,P<0.001),dry cough(OR=2.420,95%CI: 1.144～5.117,P=0.021),dizziness(OR=0.254,95%CI: 1.358～11.442,P=0.012),CRP(OR=0.987,95%CI:0.979～0.995,P=0.001)and tree-in-bud sign(OR=2.161,95%CI: 1.004～4.654,P=0.049)is a predictor of MPP in adults(P<0.05).4.Based on the above predictors,the nomogram was successfully established.The Bootstrap self-sampling method was used to verify the nomogram model internally,the C-index was 0.836,the correction curves of the training set and the validation set were well fitted,the area under the working characteristic curve(AUC)of the subjects in the training set was 0.829(95%CI: 0.774～0.883),and the AUC of the verification set was0.847(95%CI: 0.768～0.925).The decision curve analysis(DCA)of the nomogram is always above the None line and the AII line within the threshold range,indicating that the nomogram model has good clinical practicability.Conclusions:Based on six clinical indicators,including age,body temperature,dry cough,dizziness,CRP and tree-in-bud sign,we successfully established a nomogram model to predict the occurrence of MPP in adults,which has good predictive efficacy and helps clinicians quickly and accurately identify high-risk patients with MPP and adopt targeted treatment strategies.