COMPARISON OF END-EXPIRATORY LUNG VOLUME MEASUREMENT BY ELECTRICAL IMPEDANCE TOMOGRAPHY AND NITROGEN WASHOUT METHOD IN PIGS

Abstract

End-expiratory lung volume (EELV) can be determined using several methods that allow clinically accurate measurements, but it is difficult to apply these methods to the patient's bedside. Electrical impedance tomography (EIT) is offered as another method for measuring EELV. The aim of the study is to compare changes in EELV measured by nitrogen washout method with changes of EELV calculated from the change in end-expiratory lung impedance (EELI) measured by EIT and to determine whether changes in EELV calculated from changes in chest impedance can be used as one of the parameters for EIT data analysis and description. The prospective interventional animal study was performed on ten pigs. The animals received total intravenous anesthesia with muscle relaxation. Mechanical lung ventilation was conducted in the volume-controlled mode. 16-electrode EIT system was used for data acquisition. End-expiratory lung volume was measured by a modified nitrogen wash-in/wash-out technique developed by Olegard et al. The study protocol consisted of the baseline phase, two incremental PEEP steps, two decremental PEEP steps and from normal saline i. v. administration. For each animal, a reference frame (baseline frame) was selected from the initial baseline phase and was used for the reconstruction of EIT images and impedance waveforms. For each breath cycle, tidal variation image was calculated as a difference between the end-inspiratory and the previous end-expiratory EIT image. An equivalent end-expiratory volume change (ΔEELVequiv) was calculated from EELI. The values of ΔEELVequiv were compared with reference EELV data measured by a modified nitrogen wash-in/wash-out technique (ΔEELVmeas). The measured and the estimated changes in EELV were statistically compared and correlation between ΔEELVequiv and ΔEELVmeas was calculated. Statistically significant difference between ΔEELVequiv and ΔEELVmeas was observed only in administration of normal saline bolus. Pearson’s correlation coefficients were 0.29 for increase in PEEP, 0.45 for decrease in PEEP and -0.1 during administration of normal saline bolus. The study showed that during changes in PEEP in the porcine model, there was no linear relationship between ΔEELVequiv and ΔEELVmeas. Although there was no linear relationship between ΔEELVequiv and ΔEELVmeas with changes in PEEP, no statistically significant difference was demonstrated between these two methods, which justifies the use of ΔEELVequiv as a parameter suitable for description and evaluation of EIT data.