Optimization strategy for the early timing of bronchoalveolar lavage treatment for children with severe mycoplasma pneumoniae pneumonia

BACKGROUND: Early evaluation of severe mycoplasma pneumoniae pneumonia (SMPP) and the prompt utilization of fiberoptic bronchoscopic manipulation can effectively alleviate complications and restrict the progression of sequelae. This study aim to establish a nomogram forecasting model for SMPP in children and explore an optimal early therapeutic bronchoalveolar lavage (TBAL) treatment strategy. METHODS: This retrospective study included children with mycoplasma pneumoniae pneumonia (MPP) from January 2019 to December 2021. Multivariate logistic regression analysis was used to screen independent risk factors for SMPP and establish a nomogram model. The bootstrap method was employed and a receiver operator characteristic (ROC) curve was drawn to evaluate the accuracy and robustness of the model. Kaplan–Meier analysis was used to assess the effect of lavage and hospitalization times. RESULTS: A total of 244 cases were enrolled in the study, among whom 68 with SMPP and 176 with non-SMPP (NSMPP). A prediction model with five independent risk factors: left upper lobe computed tomography (CT) score, sequential organ failure assessment (SOFA) score, acute physiology and chronic health assessment (APACHE) II score, bronchitis score (BS), and c-reactive protein (CRP) was established based on the multivariate logistic regression analysis. The ROC curve of the prediction model showed the area under ROC curve (AUC) was 0.985 (95% confidence interval (CI) 0.972–0.997). The Hosmer–Lemeshow goodness-of-fit test results showed that the nomogram model predicted the risk of SMPP well (χ2 = 2.127, P = 0.977). The log-rank result suggested that an early BAL treatment could shorten MPP hospitalization time (P = 0.0057). CONCLUSION: This nomogram model, based on the left upper lobe CT score, SOFA score, APACHE II score, BS, and CRP level, represents a valuable tool to predict the risk of SMPP in children and optimize the timing of TBAL.