Evaluation of Regional Pulmonary Ventilation in Spontaneously Breathing Patients with Idiopathic Pulmonary Fibrosis (IPF) Employing Electrical Impedance Tomography (EIT): A Pilot Study from the European IPF Registry (eurIPFreg)

Objectives: In idiopathic pulmonary fibrosis (IPF), alterations in the pulmonary surfactant system result in an increased alveolar surface tension and favor repetitive alveolar collapse. This study aimed to assess the usefulness of electrical impedance tomography (EIT) in characterization of regional ventilation in IPF. Materials and methods: We investigated 17 patients with IPF and 15 healthy controls from the University of Giessen and Marburg Lung Center (UGMLC), Germany, for differences in the following EIT parameters: distribution of ventilation (TID), global inhomogeneity index (GI), regional impedance differences through the delta of end-expiratory lung impedance (dEELI), differences in surface of ventilated area (SURF), as well as center of ventilation (CG) and intratidal gas distribution (ITV). These parameters were assessed under spontaneous breathing and following a predefined escalation protocol of the positive end-expiratory pressure (PEEP), applied through a face mask by an intensive care respirator (EVITA, Draeger, Germany). Results: Individual slopes of dEELI over the PEEP increment protocol were found to be highly significantly increased in both groups (p < 0.001) but were not found to be significantly different between groups. Similarly, dTID slopes were increasing in response to PEEP, but this did not reach statistical significance within or between groups. Individual breathing patterns were very heterogeneous. There were no relevant differences of SURF, GI or CGVD over the PEEP escalation range. A correlation of dEELI to FVC, BMI, age, or weight did not forward significant results. Conclusions: In this study, we did see a significant increase in dEELI and a non-significant increase in dTID in IPF patients as well as in healthy controls in response to an increase of PEEP under spontaneous breathing. We propose the combined measurements of EIT and lung function to assess regional lung ventilation in spontaneously breathing subjects.

Supplementary Material S1: Measured EIT variables in our study.
1. Intratidal gas distribution (ITV): ITV, originally described by Loewhagen et al., exhibits the way the inspired gas distributes in the lungs during the tidal breath at defined ROIs. It therefore delineates the changes in regional ventilation with time course during one breath in different ROIs. When the lung region is divided into two ROIs, i.e. the ventral and dorsal regions, impedance changes from the beginning of inspiration to a certain time point in each region are expressed as a fraction of the contribution to global impedance change [30]. In our study, we have been analyzing 4 different ROIs, namely a ventral, a dorsal, a medioventral and a mediodorsal ROI.
Tidal Impedance Distribution (TID): This parameter reflects the average ventilation for a defined period of breath, as reflected by the difference in impedance values at the beginning and the end of an inspiration. For the purpose of our study, we set the TID, as measured in IPF and controls at a PEEP level of zero, as 100%. For illustration of regional changes in TID, impedance changes above 10% of the maximum regional impedance change are displayed in dark blue. As values increase, the dark blue turns into a lighter blue. A white color indicates the regions of maximum regional impedance change 2. Difference of TID over time (dTID): This parameter reflects a global change of TID between two points of time, in our study the change in EELI between baseline (PEEP =0cm H2O) and the various PEEP increment time points. dTID will be measured as % and as deviation to the reference section (dTID= TID-100) of the baseline value. For illustration of regional changes in dTID, the differential image indicates increases in dTID (vs. the reference) in turquoise and decreases in orange color.

Surface of ventilated areas (SURF):
This parameter describes the surface of ventilated areas of the tomogram, as defined by a regional impedance chance between inspiration and expiration. A pixel whose regional impedance change exceeds 10 % of the maximum impedance change within the section is interpreted as ventilated. For global analysis, SURF is given as number of ventilated pixels for one section. For illustration of regional changes in the tomogram, ventilated pixels are displayed in white color; non-aerated pixels are given in a dark grey color.

Global inhomogenity index (GI):
This parameter represents the spatial extent and dispersion in the distribution of tidal breath, i.e., global inhomogeneity in tidal ventilation [31]. For the global GI, this value is calculated as the sum of absolute differences between the median impedance value of all pixels and each single pixel value, divided by the sum of all impedance values in order to normalize the calculated values. The smaller the GI, the more homogeneous the tidal volume is distributed within the ventilated area. A GI with a value 0 represents a perfectly homogeneous distribution. For illustration of regional changes in the tomogram, the absolute differences between the median impedance value of all pixels and each single pixel value is depicted. No difference between the single pixel and the median value is depicted in white color, differences in are depicted in darker color in dependency of the magnitude of difference. 6. End-expiratory Lung Impedance (EELI): This parameter reflects the impedance at endexpiration. The higher the EELI, the more lung volume is present at end-expiration. However, as the absolute lung volume is rarely known, EELI cannot be directly related to the end-expiratory lung volume. For the global analysis, EELI is set 100% for the baseline condition.

Difference of EELI over time (dEELI):
This parameter reflects a change of EELI between two points of time; in our study the change in EELI between baseline (PEEP = 0 cm H2O) and the various PEEP increment time points. For the global analysis, the differences of all sections are scaled to the reference section and are given in percentage of the baseline value. For illustration of regional changes, the dEELI value of each pixel is indicated by a blue to white color and the more color the higher the percental increase.