Real-time monitoring of lung injury in a porcine model using electrical impedance tomography: correlation between early quantitative assessment and AIS grading

Abstract Background Lung injuries are a prevalent and potentially fatal condition in clinical emergency settings and on the battlefield, accounting for 30% to 40% of emergency department cases involving chest trauma. However, current diagnostic methods are inadequate for the urgent need for rapid triage. This study aimed to use electrical impedance tomography (EIT) technology to reflect pathophysiological changes from dynamic variations in lung impedance, thereby establishing a new method for the early assessment of lung injury. Methods The study on anaesthetised Landrace pigs established lung injury models (projectile impact, n = 24) at varying velocities. An intergroup analysis with a pre-post design was employed. Prior to and following injury, EIT technology monitored ventilation and perfusion impedance signals. The ventilation uneven index ( VUI ), perfusion uneven index ( PUI ), and ventilation/perfusion ( V/Q ) spatial matching index were extracted from the EIT images to analyse their relationship with impact velocity and their correlation with Abbreviated Injury Scale (AIS) grading. Results EIT imaging revealed a positive correlation between the degree of pulmonary ventilation and perfusion deficit and the impact velocity. Statistical analysis indicated that EIT indices exhibited a progressive increase in conjunction with rising impact velocity. Post-injury V/Q spatial matching index demonstrated a significant decrease, while VUI and PUI exhibited significant increases (all P < 0.01). The implementation of correlation analysis yielded a significant correlation between Δ V/Q spatial matching index and AIS grading ( r = 0.87, P < 0.001). Δ V/Q spatial matching index exhibited the highest diagnostic accuracy for identifying severe lung injury ( AUC = 0.96, 95%CI: 0.8739—1.00, P < 0.001), while Δ VUI ( AUC = 0.85, 95%CI: 0.6956—1.00, P < 0.01) and Δ PUI ( AUC = 0.79, 95%CI: 0.6067—0.9826, P < 0.05) also demonstrated adequate diagnostic performance. Conclusion EIT was applied in this study to quantify lung injury severity, which further confirmed its correlation with AIS classification. Capable of early evaluation, prompt triage and continuous dynamic monitoring, this method is expected to facilitate clinical application and translational research.