(SLC-027) Development and Validation of a Nomogram for Predicting the Risk of Refractory Mycoplasma Pneumoniae Pneumonia in Children

Background: Macrolide antibiotics are the first-line treatment of Mycoplasma pneumoniae pneumonia (MPP) in children. However, a subset of patients do not exhibit significant improvement after receiving standard macrolide therapy for seven days or longer, a condition defined as refractory MPP (RMPP). Due to the lack of timely and effective interventions, these patients may develop severe complications or long-term sequelae, leading to a higher disease burden and medical costs.

Objectives: This study aimed to develop and validate a predictive model for the early identification of RMPP in children, providing a reference for guiding treatment decisions.

Methods: We conducted a retrospective cohort study of pediatric MPP patients admitted to Beijing Children's Hospital between January 2019 and December 2023. Patients were classified into the RMPP group and the general MPP group based on their clinical manifestations. Data on demographic characteristics, medical history, clinical features, laboratory indicators, and radiological information were collected for both groups. The data was randomly divided into a training cohort and a validation cohort in a 7:3 ratio for model development and validation. Key predictors were selected using LASSO regression. The model was constructed using the logistic regression and visualized using a nomogram. The performance was evaluated using the area under the receiver operating characteristic curve (AUC-ROC) and calibration curves. Decision curve analysis (DCA) was further applied to assess the clinical utility of the model.

Results: A total of 2207 pediatric MPP patients were included in the study, with 333 classified as RMPP. After LASSO regression, seven key predictors were selected for model construction, including the fever duration, peak body temperature, CRP, ALT, AST, LDH and D-dimer. The logistic regression model demonstrated good discrimination, with an AUC-ROC of 0.794 (95% CI 0.763–0.824) in the training cohort and 0.796 (95% CI 0.752–0.841) in the validation cohort. Calibration curves closely aligned with the 45-degree reference line, indicating good agreement between predicted and observed probabilities. DCA demonstrated that the model provided a net clinical benefit across a threshold probability range of 10% to 60%, supporting its potential clinical utility for early identification of RMPP.

Conclusions: The predictive model developed in this study exhibited good discrimination, calibration, and clinical utility for the early identification of RMPP. The model helps clinicians to identify high-risk patients early and provides a basis for individualized treatment strategies in pediatric MPP patients.