Establishment and validation of a predictive model for Lobar pneumonia caused by Mycoplasma pneumoniae infection in children.

To develop and validate a reliable and noninvasive prediction model for early identification of lobar pneumonia in children with mycoplasma pneumoniae (MP) infection. This retrospective study included 209 pediatric patients diagnosed with MP infection between January and December 2023 at the Children's Hospital of Kunming Medical University. Clinical symptoms, laboratory findings, and imaging data were collected. Univariate analysis and receiver-operating-characteristic (ROC) curve analysis were performed to assess the predictive value of variables. least absolute shrinkage and selection operator (LASSO) regression was used to identify key predictive variables, which were then integrated into a logistic regression model. A nomogram was developed based on the model, and its performance was validated using ROC curves, calibration plots, and decision curve analysis (DCA). The predictive nomogram identified albumin (ALB), lactate dehydrogenase (LDH), presence of rales, and co-bacterial infection as significant predictors of lobar pneumonia in children with MP infection. The model demonstrated strong predictive accuracy, with area under the curve (AUC) values of 0.8463 (95% CI: 0.7793-0.9134) in the training set and 0.810 (95% CI: 0.742-0.8773) in the validation set. Calibration plots showed good agreement between predicted and actual outcomes, and DCA confirmed the model's clinical utility, demonstrating greater net benefit compared to all-intervention or no-intervention strategies. This study presents a novel, accurate, and clinically useful nomogram for predicting lobar pneumonia in children with MP infection based on routine clinical and laboratory data. The model provides a non-invasive approach to early diagnosis, potentially reducing the need for repeated imaging and associated radiation exposure.