The measurement of regional lung ventilation by electrical impedance tomography (EIT) has been evaluated in many experimental studies. However, EIT is not routinely used in a clinical setting, which is attributable to the fact that a convenient concept for how to quantify the EIT data is missing. The definition of region of interest (ROI) is an essential point in the data analysis. To date, there are only limited data available on the different approaches to ROI definition to evaluate regional lung ventilation by EIT. For this survey we examined ten patients (mean age +/- SD: 60 +/- 10 years) under controlled ventilation. Regional tidal volumes were quantified as pixel values of inspiratory-to-expiratory impedance differences and four types of ROIs were subsequently applied. The definition of ROI contours was based on the calculation of the pixel values of (1) standard deviation from each pixel set of impedance data and (2) the regression coefficient from linear regression equations between the individual local (pixel) and average (whole scan) impedance signals. Additionally, arbitrary ROIs (four quadrants and four anteroposterior segments of equal height) were used. Our results indicate that both approaches to ROI definition using statistical parameters are suitable when impedance signals with high sensitivity to ventilation-related phenomena are to be analyzed. The definition of the ROI contour as 20-35% of the maximum standard deviation or regression coefficient is recommended. Simple segmental ROIs are less convenient because of the low ventilation-related signal component in the dorsal region.

EIT and automated curve fitting provide information on regional lung inflation and deflation which may be of clinical use for optimizing ventilator settings.

Considerable regional heterogeneity was present during ventilation and a derecruitment maneuver. Significantly lower change in relative DeltaZ in the most dependent lung regions suggests alveolar collapse during ventilation before suctioning.