Development and Evaluation of a Nomogram for Predicting Pulmonary Embolism in Children With Severe Mycoplasma pneumoniae Pneumonia

To construct a nomogram utilizing pediatric severe Mycoplasma pneumoniae pneumonia (SMPP) risk factors for pulmonary embolism (PE), facilitating the clinical identification and management of high-risk patients and reducing the excessive use of CT pulmonary angiography (CTPA).

This was a retrospective analysis conducted between August 2021 and March 2024. We identified 35 children with SMPP complicated by PE, forming the PE group. A control group of 70 age- and sex-matched children with SMPP without PE was randomly selected at a 1:2 ratio. Clinical, laboratory, and CT findings were compared between the groups. Least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to develop a scoring model using a nomogram. The model's performance was assessed via the receiver operating characteristic curve (ROC), fivefold cross-validation, calibration curve, and clinical decision curve analysis.

LASSO regression and multivariate logistic regression analyses revealed that D-dimer, neutrophil ratio, time to admission, pleural effusion, and necrotizing pneumonia were independent risk factors for PE in patients with SMPP. A nomogram prediction model was established based on the aforementioned independent risk factors. The area under ROC curve was 0.900. Fivefold cross-validation results further confirmed the model's stability. The calibration curve revealed good agreement between the predicted and actual probabilities of PE caused by SMPP, and the decision curve demonstrated that the nomogram model had a higher clinical net benefit.

The nomogram serves as a predictive tool to aid in early intervention for pediatric patients with SMPP at high risk for PE, while minimizing unnecessary CTPA and overtreatment in low-risk patients.