AUTHOR=Ding Xiaoyan , Bai Jing , Liu Rui 

TITLE=Risk factors for airway clearance dysfunction in children with severe pneumonia: a retrospective study of LASSO model

JOURNAL=Frontiers in Pediatrics

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2025.1638103

DOI=10.3389/fped.2025.1638103

ISSN=2296-2360

ABSTRACT=BackgroundThis study aimed to investigate the types and risk factors of airway clearance dysfunction (ACD) in children with severe pneumonia based on LASSO regression and multivariate Logistic regression analysis.MethodsThis study was approved by the Hospital Ethics Committee. A retrospective study was conducted on 147 children with severe pneumonia admitted between January 2024 and October 2024. Demographic and clinical characteristics were collected, and the incidence and types of ACD were analyzed. Patients with ACD were assigned to the ACD group, while those without ACD were assigned to the non-ACD group. LASSO regression and multivariate Logistic regression were used to identify risk factors for ACD in these children, and ROC curves were constructed to evaluate the predictive value of these factors for ACD.ResultsA total of 63 cases of ACD were observed among children with severe pneumonia, including ineffective coughing, thick sputum, dyspnea, nasal flaring/ increased nasal congestion, tachypnea, and oxygen saturation <93%, with an incidence rate of 42.86%. Significant differences were found between the ACD group and non-ACD group in terms of age, mechanical ventilation, nutritional status, comorbid respiratory failure, comorbid heart failure, pulmonary rales, and pulmonary rhonchi (P < 0.05). Additionally, the ACD group had lower platelet count (PLT), immunoglobulin A (IgA), and immunoglobulin M (IgM) levels, while C-reactive protein (CRP), procalcitonin (PCT), and interleukin-6 (IL-6) levels were higher compared to the non-ACD group (P < 0.05). Variable screening was performed using the LASSO regression model, identifying three significant influencing factors. These were incorporated into a Logistic regression model, which revealed that mechanical ventilation, CRP, PCT, IL-6, and IgA were influencing factors for ACD in children with severe pneumonia. Based on these findings, an ROC curve was constructed, demonstrating that the combined prediction of mechanical ventilation, CRP, PCT, IL-6, and IgA for ACD in children with severe pneumonia achieved an AUC of 0.882, significantly higher than the AUC of any single indicator (P < 0.05).ConclusionChildren with severe pneumonia are at risk of developing ACD, which may be influenced by mechanical ventilation, CRP, PCT, IL-6, and IgA levels. These five factors can be used to assess the risk of ACD in children with severe pneumonia. Accordingly, clinical measures should be developed to improve airway clearance ability and promote recovery in these patients.