Background: This study aims to construct a risk prediction model based on clinical manifestations, laboratory blood indicators and radiological fndings to help clinicians identify patients who are at high risk of refractory mycoplasma pneumoniae pneumonia.
Materials and methods: We retrospectively analyzed the medical records of 369 children with MPP.The data collected included demographics, clinical data ,laboratory findings and imaging data. Descriptive statistical analysis, involving numerous variables, was followed by univariate and multivariate logistic regression analysis.Subsequently,the clinical prediction model was constructed and underwent internal validation.
Results: The clinical prediction model was constructed from these eight variables included fever duration,Pleural effussion,WBC,NEP,CRP, LDH,NLR and SUA.The developed nomogram, which has a satisfactory level of accuracy and good calibration, can be utilized to predict RMPP patients.
Conclusion: Fever duration more than 10.5 days,Pleural effussion,WBC>10.13×109/L,NEP>6.43×109/L,CRP>29.45mg/L,LDH>370.50U/L,NLR>3.47 and SUA<170.5 umol/ml was early predictive model of RMPP.The developed nomogram, which has a satisfactory level of accuracy and good calibration, can be utilized to predict RMPP patients.