Monitoring lung contusion in a porcine polytrauma model using EIT: an application study. (26th July 2017)

• Record Type: Journal Article
• Authors: Santos, Susana Aguiar; Wembers, Carlos Castelar; Horst, Klemens; Pfeifer, Roman; Simon, Tim-Philipp; Pape, Hans-Christoph; Hildebrand, Frank; Czaplik, Michael; Leonhardt, Steffen; Teichmann, Daniel
• Is Part Of: Physiological measurement. Volume 38:Number 8(2017:Aug.)
• Page Start: 1542
• Page End: 1560
• Publication Date: 2017-07-26
• Subjects: electrical impedance tomography -- lung contusion -- EIT indices -- side-specific analysis; Physiology -- Measurement -- Periodicals; Patient monitoring -- Periodicals; 612
• DOI: 10.1088/1361-6579/aa7985
• Languages: English
• ISSNs: 0967-3334

Abstract

Abstract: Lung contusion is the most common lung injury following blunt chest trauma which, in turn, is associated with high mortality rates (Gavelli et al 2002 Eur. Radiol .12 1273–94). Lung contusion is characterized by hemorrhage and edema with consecutively reduced compliance. Objective and Approach : In this study, unilateral lung contusion and other traumata were induced in 12 pigs by using a bolt gun machine. To investigate the pathophysiological consequences of lung contusion, information on clinical parameters was collected and monitored regularly while animals were additionally monitored with electrical impedance tomography (EIT) before trauma, and at 4, 24, 48 and 72 h after polytrauma. Main results : Statistical analyses showed significant differences between the measurement time points in terms of lung compliance ( p < 0.0001 ) and in global EIT parameters, such as absolute global impedance (aGlobImp) ( p < 0.0001 ), tidal impedance variation (TIV) ( p < 0.0001 ) and the center of ventilation (CoV) ( p = 0.005 ). Additionally, distinct analyses for the left (non-injured) and right (injured) lung were also performed. In this context, during the progress of lung contusion, significant changes were found for the injured lung in TIV ( p = 0.033 ), global inhomogeneity ( p = 0.002 ), regional ventilation delay ( p = 0.033 ), CoV ( p = 0.049 ) and in regions of non-ventilation (rNoVent) ( p = 0.001 ). Furthermore, TIV and rNoVent were capable to differentiate the injured and the contralateral healthy lung at 4 and 24 h after injury (TIV: p = 0.001 and p = 0.007 ; rNoVent: p = 0.004 and p = 0.027 ). TIV reached a sensitivity of 82% (specificity of 100%) at 4 h and sensitivity of 82% (specificity of 82%) at 24 h after injury, in detecting lung contusion specific consequences. Significance : The results indicate that EIT might be a valuable tool to detect and to monitor lung injuries including lung contusion. Most probably, EIT-derived indices could also be used to adapt ventilator settings to optimize individual lung protection.