The Predictive Value Of Quantitative CT Lung Density In The Timing Of Tracheal Intubation In Acute Exacerbation Of Chronic Obstructive Pulmonary Disease

BackgroundChronic obstructive pulmonary disease（COPD）is a common chronic respiratory system disease,and acute exacerbation of chronic obstructive pulmonary disease（AECOPD）is a frequent clinical event in the process of COPD,with severe cases even leading to respiratory failure and life-threatening.As the first station for treating AECOPD patients,the emergency department needs to make timely judgments and rapid treatment of the severity of the condition.Currently,the commonly used respiratory support methods for such patients are non-invasive ventilator assisted ventilation and tracheal intubation assisted ventilation.But the timing of tracheal intubation depends on the empirical judgment of clinical doctors,often influenced by medical conditions and family wishes.How to quickly identify critically ill patients who require tracheal intubation has become the key to delaying disease progression and improving prognosis.Currently,chest CT is the most accurate imaging technique for detecting and evaluating chronic obstructive pulmonary disease.With the development of medical imaging technology,quantitative CT lung density parameters are closely related to the lung function,disease progression,and clinical outcomes of patients with chronic obstructive pulmonary disease.In the outbreak of COVID-19 in 2020,some researchers predicted the outcome of noninvasive ventilation and tracheal intubation in severe patients with COVID-19 through quantitative CT lung density parameters,reflecting the value of quantitative CT in the diagnosis and assessment of lung diseases.PurposeExploring the predictive value of quantitative parameters on tracheal intubation outcomes in AECOPD patients by analyzing the chest CT lung density.MethodsFrom January 1,2019 to August 31,2022,58 patients with AECOPD admitted to the Emergency Department,Intensive Care Unit,and General Ward of the First Affiliated Hospital of Guangzhou Medical University were selected.They were divided into endotracheal intubation group and non-invasive ventilation group based on different types of mechanical ventilation,including 26 patients with endotracheal intubation assisted ventilation and 32 patients with non-invasive assisted ventilation.We collected general information,laboratory examinations,and chest CT images of the included patients,and imported the chest CT images into the medical image quantitative analysis software 3DSlicer for lung density analysis.We obtained the percentage of lung tissue in each density range to the overall lung tissue,which is represented as non ventilated lung tissue（%）,low ventilated lung tissue（%）,normal ventilated lung tissue（%）,over ventilated lung tissue（%）,and damaged lung tissue（%）,respectively,The damaged lung tissue（%）is the sum of non ventilated lung tissue（%）and hypoventilated lung tissue（%）.Then,the general information,laboratory examinations,and quantitative CT lung density analysis parameters between the two groups were further compared.The influencing factors on tracheal intubation outcomes were explored through multivariate binary logistic regression analysis.Finally,the predictive value of quantitative CT lung density analysis parameters on tracheal intubation outcomes was analyzed using the subject operating characteristic curve（ROC）.Results（1）By establishing a multivariate binary logistic regression analysis between gender,age,laboratory examination,past/concomitant diseases,and quantitative CT lung density analysis parameters,it was found that damaged lung tissue（%）can accurately predict tracheal intubation outcomes in AECOPD patients,and damaged lung tissue（%）is the only significantly correlated variable（P<0.05）.（2）The strongest correlation was found between Pa O₂/Fi O₂ and damaged lung tissue（%）,showing a negative correlation（rs=-0.530,P<0.05）.（3）Through the ROC curve,it was found that the optimal diagnostic cutoff point for predicting tracheal intubation in AECOPD patients with damaged lung tissue（%）is 10.5%,AUC is 0.680（P<0.05）,sensitivity is 0.577,and specificity is 0.844.ConclusionsDamaged lung tissue（%）can predict the outcome of tracheal intubation in AECOPD patients.When the damaged lung tissue（%）is greater than 10.5%,the risk of tracheal intubation in AECOPD patients will significantly increase,providing a reference basis for selecting the timing of tracheal intubation.