Clinical Features and Biomarkers for Early Prediction of Refractory Mycoplasma Pneumoniae Pneumonia in Children

Summary of Background Data.With the increasing incidence and mortality of refractory mycoplasma pneumonia (RMPP), the early diagnosis and treatment of RMPP have attracted more and more attention and have become a major concern in pediatrics. Objective. The study aimed to analyze the clinical characteristics of children with RMPP and to explore the biomarkers for the early prediction of RMPP, thus providing references for the clinical diagnosis and treatment of RMPP in children. Methods. Baseline clinical characteristics, clinical symptoms, physical examination, chest imaging, and laboratory indicators between children with RMPP and general refractory mycoplasma pneumoniae pneumonia (GMPP) were compared. Multiple logistic regression analysis was used to determine independent risk factors for RMPP. ROC curves were adopted to analyze the predictive values of biomarkers. Results. The RMPP group had more severe clinical symptoms and manifestations on imaging (including pleural effusion, pulmonary consolidation, and pulmonary atelectasis), a higher incidence of extrapulmonary complications, and a longer duration of hospital stays. Results of multiple logistic regression analysis showed that serum D-dimer (OR = 8.169, ), C-reactive protein (CRP) (OR = 1.146, ), and lactate dehydrogenase (LDH) (OR = 1.025, ) levels were independent risk factors for RMPP. The area under the receiver operating characteristic curve (AUROC) in RMPP prediction was 0.841, 0.870, and 0.893 for serum levels of D-dimer, CRP, and LDH, respectively (), with a cutoff value of 1.47 ng/ml, 39.34 mg/L, and 379 IU/L, respectively. Conclusions. Serum D-dimer, CRP, and LDH levels were related to the severity of mycoplasma pneumoniae pneumonia in children and had potential as biomarkers for the early prediction of RMPP, suggesting great applicative values for the early diagnosis and timely intervention of children with RMPP in clinical practice.