Center of Ventilation—Methods of Calculation Using Electrical Impedance Tomography and the Influence of Image Segmentation

The aim of the study was to validate the ability of electrical impedance tomography (EIT) to detect local changes in air content, resulting from modified ventilator settings, by comparing EIT findings with electron beam computed tomography (EBCT) scans obtained under identical steady-state conditions. The experiments were carried out on six anesthetized supine pigs ventilated with five tidal volumes (VT) at three positive end-expiratory pressure (PEEP) levels. The lung air content changes were determined both by EIT (Goe-MF1 system) and EBCT (Imatron C-150XP scanner) in six regions of interest, located in the ventral, middle, and dorsal areas of each lung, with respect to the reference air content at the lowest VT and PEEP, as a change in either local electrical impedance or lung tissue density. An increase in local air content with VT and PEEP was identified by both methods at all regions studied. A good correlation between the changes in lung air content determined by EIT and EBCT was revealed. Mean correlation coefficients in the ventral, middle, and dorsal regions were 0.81, 0.87, and 0.93, respectively. The study confirms that EIT is a suitable, noninvasive method for detecting regional changes in air content and monitoring local effects of artificial ventilation.

Although surfactant replacement therapy is an established treatment in infant respiratory distress syndrome, the optimum strategy for ventilatory management before, during, and after surfactant instillation remains to be elucidated. To determine the effects of surfactant and lung volume recruitment on the distribution of regional lung ventilation. Acute lung injury was induced in 16 newborn piglets by endotracheal lavage. Optimum positive end-expiratory pressure was identified after lung recruitment and surfactant was administered either at this pressure in the "open" lung or after disconnection of the endotracheal tube in the "closed" lung. An additional recruitment maneuver with subsequent optimum end-expiratory pressure finding was executed in eight animals; in the remaining eight animals, end-expiratory pressure was set at the same level as before surfactant without further recruitment. ("Open" and "closed" lung surfactant administration was evenly distributed in the groups.) Regional ventilation was assessed by electrical impedance tomography. Impedance tomography data, airway pressure, flow, and arterial blood gases were acquired during baseline conditions, after induction of lung injury, after the first lung recruitment, and before as well as 10 and 60 min after surfactant administration. Significant shift in ventilation toward the dependent lung regions and less asymmetry in the right-to-left lung ventilation distribution occurred in the postsurfactant period when an additional recruitment maneuver was performed. Surfactant instillation in an "open" versus "closed" lung did not influence ventilation distribution in a major way. The spatial distribution of ventilation in the lavaged lung is modified by a recruitment maneuver performed after surfactant administration.

Electrical impedance tomography (EIT) is a noninvasive technique providing cross-sectional images of the thorax. We have tested an extended evaluation procedure, the functional EIT (f-EIT), to identify the local shifts of ventilation known to occur during the transition between spontaneous, controlled and assisted ventilation modes. Ten patients scheduled for elective laparotomy were studied in the surgical ward, operating theatre and ICU during spontaneous and different modes of mechanical ventilation. Sixteen ECG electrodes were placed on the circumference of the thorax and connected with an EIT device (APT System Mark I, IBEES, Sheffield, UK). Measurements lasting 180 s were performed and f-EIT images of regional ventilation computed. The geometrical centre of ventilation was determined to quantify the regional distribution of lung ventilation during individual modes of ventilation. F-EIT confirmed the differences in the distribution of ventilation associated with various modes of artificial ventilation. Accentuated ventilation of the dependent lung regions was observed during spontaneous breathing, whereas a shift of the centre of ventilation to the nondependent regions was found during controlled ventilation. In the course of assisted ventilation a continuous displacement of the centre of ventilation back towards the dependent lung regions, consistent with an increased proportion of spontaneous breathing, was detected. Unassisted spontaneous breathing after weaning from mechanical ventilation resulted in a similar ventilation distribution as during tidal breathing prior to surgery. F-EIT determined the redistribution of lung ventilation during different modes of mechanical ventilation. We expect that f-EIT will become a useful noninvasive bedside monitoring technique for imaging regional ventilation in pulmonary diseased patients during mechanical ventilation.