Oral Presentation - 55
Development of a Risk Stratification Model for Complications Associated with Bronchial Blockers versus Endobronchial Intubation in One-Lung Ventilation for Infants and Young Children
Yongsheng Qiu
Department of Anesthesiology, Children’s Hospital of Henan Province, Zhengzhou, China
Objective
This study aimed to establish a risk stratification model for complications related to bronchial blockers (BB) and endobronchial intubation (EBI) during one-lung ventilation (OLV) in infants and young children (2 months to 3 years), using multicenter retrospective data to inform evidence-based individualized airway management.
Methods
A total of 104 pediatric patients undergoing thoracic surgery (BB group: n=60; EBI group: n=44) were included. Propensity score matching was applied to balance baseline characteristics. Moderate and severe airway complications (e.g., tube obstruction, isolation failure, hypoxemia requiring intervention) were defined. A dynamic risk stratification system was developed using machine learning algorithms incorporating multimodal parameters, including age, preoperative SpO₂, OLV duration, surgical laterality, and intraoperative tidal volume.
Results
The EBI group exhibited significantly higher rates of severe complications compared to the BB group (5.8% vs. 2.1%, P=0.035), primarily due to complete endotracheal tube obstruction (4.3% vs. 0.4%) and intraoperative reintubation (3.4% vs. 1.1%). Core risk factors included age <1 year, preoperative SpO₂ <98%, right-sided surgery (risk increase: 2.33-fold), and OLV duration >60 minutes. Model validation demonstrated an AUC of 0.82 (95% CI: 0.76–0.89) for predicting severe complications, with 84.6% sensitivity.
Conclusion
BB significantly reduces severe complications such as airway obstruction in young children undergoing OLV, particularly for left-sided or shorter-duration surgeries, while EBI remains preferable for older subgroups (>2 years) requiring right-sided procedures. The proposed risk stratification model guides anesthesiologists in selecting optimal lung isolation strategies based on patient profiles and surgical demands, though its clinical utility warrants further validation through prospective multicenter studies.