Evaluation of an electrical impedance tomography-based global inhomogeneity index for pulmonary ventilation distribution

To evaluate the plausibility, stability, and interindividual comparability of the global inhomogeneity index (GI) based on electrical impedance tomography (EIT). The lung area in an EIT image was identified by using the lung area estimation method, which mirrors the lung regions in the functional EIT image and subsequently subtracts the cardiac-related areas. The tidal EIT image, showing the difference in impedances between end-inspiration and end-expiration, was calculated and the variations in its pixel values within the predefined lung area were then used as an indicator of inhomogeneous ventilation (the GI index). Fifty patients were investigated including 40 patients tracheally intubated with double-lumen tubes (test group) and 10 patients under anesthesia without pulmonary disease (control group). Positive end-expiratory pressure (PEEP) of 5 mbar was applied in the test group during both two-lung ventilation (TLV) and subsequent one-lung ventilation (OLV). The patients of the control group were ventilated without PEEP. EIT data were recorded in both groups. A significantly lower GI value was found in the control group (0.40 +/- A 0.05, P = 0.025 vs. TLV 0.74 +/- A 0.47 and P < 0.002 vs. OLV 1.51 +/- A 1.45). A significant difference was also found in the test group between TLV and OLV (P < 0.002). If GI was calculated only in the ventilated lung area during OLV (0.71 +/- A 0.32), it did not significantly differ from the test group during TLV. The GI index quantifies the gas distribution in the lung with a single number and reveals good interpatient comparability.