An Introduction to the Clinical Application and Interpretation of Electrical Impedance Tomography

Tidal Impedance Variation

TIV is simply the change in impedance from end exhalation to end inspiration; it represents the change in V_T being distributed through the lung. It is visually represented as an impedance waveform on the EIT device, like that of a V_T waveform on a ventilator (see Fig. 3).

Dorsal Fraction of Ventilation

A simple concept of measuring the predominant area of ventilation distribution is observing the dorsal fraction of ventilation (DFV), which describes the percentage of ventilation reaching the dorsal lung regions and can be used to determine if an increase in PEEP should be attempted and also as an indication to decrease PEEP³ (described later).

Regional Compliance

When EIT devices calculate compliance of various regions, TIV is used as the numerator (like V_T) for a compliance measurement with the change in pressure as the denominator. The compliance can either be measured globally (all areas of the lung) or regionally by setting the desired ROI to be assessed. The compliance measurements are also used in an analysis to compare the changes in compliance that occur in different lung regions during a decremental PEEP titration, indicating areas of potential overdistension and collapse (described later).

Intratidal Gas Distribution

The intratidal gas distribution (ITV) separates a single breath into 8 different time segments as a percentage of an entire breath. ITV compares the distribution of ventilation in the ventral and dorsal ROI at each time segment.⁴ This allows a more in-depth analysis of changes that occur throughout inspiration. The utility of this single breath analysis is to visually assess the presence of tidal recruitment (opening from a state of atelectasis) as well as overdistension. It has been used as a method for determining optimal PEEP.⁵ It should be noted that commercially available devices do not analyze ITV at the bedside; it requires an off-line analysis.

Additionally, the ITV of the ventral regions can be divided by the ITV of the dorsal lung regions to provide the ITV index.⁵

Regional Ventilation Delay

Regional ventilation delay describes regions of the lungs where impedance changes later than the global impedance change (late opening of lung units). The late opening may be due to regions of atelectasis opening during tidal ventilation; it can also be present when reverse triggering is present (patient effort occurs after the onset of a ventilator-delivered breath, causing a change in the direction of ventilation distribution), and it can also appear when there is excessive effort resulting in the pendelluft effect occurring between different regions of the lungs.⁶

(i = each pixel; Δt_i^40% = the time from the start of global impedance change to reach 40% of the maximum impedance change; t_max − t_min = inspiration time.)

Global Inhomogeneity Index

The global inhomogeneity index (GI) is one of the more complex analyses performed (off line) by EIT devices. It is a measure of how homogenous ventilation distribution is throughout the lung. Lower values indicate a more homogenous distribution, whereas higher values indicate increased inhomogeneous distribution. The measurement is done after PEEP titration, and the value at each PEEP step is compared; the PEEP level with lowest GI index is considered the preferred or optimal setting.⁷

(DI = impedance variation; xy = a specific lung area pixel; lung = all pixels of the lung area)

Assessing the Response to PEEP

The ability of EIT to monitor both global and regional distribution of ventilation gives it a unique advantage in the assessment of PEEP responsiveness and alveolar recruitment. When distribution of ventilation occurs predominantly in the ventral regions, the goal of increasing PEEP would be to improve distribution of ventilation to the dorsal lung regions, resulting in a more homogenous distribution of ventilation. This can be visually assessed at the bedside using simple assessments such as the DFV described earlier, where ∼45–50% implies a more balanced distribution of ventilation between the ventral and dorsal lung regions.³ A number of the measurements described earlier can be used to assess responsiveness to PEEP and will be described further.

Optimal PEEP Using Changes in Regional Compliance

Whereas improvements to regional ventilation or increased homogeneity can indicate recruitment, it does not directly assess the possibility of overdistension resulting from recruitment. All EIT devices have software capable of comparing overdistension resulting from high PEEP to collapse resulting from low PEEP. This analysis is made following a decremental PEEP titration. The decremental PEEP trial begins from a PEEP level considered to be higher than what would be required for the patient, and PEEP is decreased in increments of 2 cm H₂O to a minimally acceptable PEEP level. The resulting analysis will plot the percent of compliance lost at the highest PEEP (risk of overdistension) compared to the lowest PEEP level, and on the same graph it will plot the percent of compliance lost at the lowest PEEP (risk of collapse) compared to the highest PEEP level (Figure 4).⁸ Different approaches have been used to set PEEP using this analysis. These include (1) setting PEEP where there is < 10% collapse; (2) setting PEEP where the 2 lines intersect; or (3) if the 2 lines intersect between 2 different PEEP levels, the PEEP where the GI index is the lowest is used. In Figure 4, these approaches would result in a PEEP selected between 22–24 cm H₂O in a patient who was very responsive to higher levels of PEEP. The currently available data do not provide evidence of superiority for one technique over another. However, the ability to assess risk and benefits of PEEP through regional compliance changes is a method of individualizing PEEP gaining popularity and study. The challenge for clinical trials using EIT-selected PEEP will be how the device is used and how often. Considering how dynamic patients can be early in the course of mechanical ventilation, if EIT is only used once to set PEEP without follow-up use it does not answer the question of whether using EIT to guide PEEP is useful.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Example of the overdistention/collapse (OD/CL) analysis done after a decremental PEEP trial and the global inhomogeneity index (GI) analysis for the same decremental PEEP trial.

Optimal PEEP Using ΔEELI

When CPAP or PEEP is applied or increased, there is always an increase in end-expiratory lung volume (EELV). The additional pressure might increase EELV in the areas currently being ventilated or recruit areas that were previously collapsed, something referred to as recruited volume. Bedside methods for assessing recruitment volume are multiple pressure volume curves (flow-derived method), the helium dilution technique, nitrogen washin/washout, and the recruitment to inflation ratio.^9-11 EIT uses the change in end-expiratory lung impedance (ΔEELI) to estimate the changes in EELV. Recruited volume estimated by EIT is significantly correlated with the nitrogen washin/washout method and with the flow-derived method; however, the limits of agreement are highly variable.^12-14 Regardless, the additional information provided by EIT in the impact of this recruited volume is the improvements in regional compliance or any loss of regional compliance that could indicate hyperinflation. This allows the assessment of the potential risk and benefit of increasing PEEP as described in the previous section.

EIT as an Incentive to Decrease PEEP

Whereas many studies have assessed methods of individualizing PEEP, there is a lack of studies that has focused on optimal methods to determine when to decrease PEEP. Methods to individualize PEEP may involve setting PEEP, weaning F_IO2 as tolerated, then a slow reduction in PEEP at various time intervals (ie, every 6 h). When EIT is used to assess recruitment, or during a decremental PEEP trial, it is often noted that distribution to the nondependent (ventral regions) while supine is diminished when PEEP is high, resulting in a DFV to be increased > 50%. When applying EIT at the bedside, this increased DFV is easily recognizable and can indicate the need to perform a stepwise decrease in PEEP to achieve a more even distribution of ventilation between the ventral and dorsal regions (ie, DFV ∼45–50%).³

Distribution Changes During Spontaneous Breathing Trials

The ability for EIT to monitor distribution of ventilation changes that occur during mode changes makes it an interesting tool for the assessment of weaning readiness or as a predictor of spontaneous breath trial (SBT) success. Whereas EIT has been used in studies assessing its potential role for this, the studies include varying methods of performing SBTs and different types of patient populations and, therefore, may be difficult to generalize the findings to all patients at this time.^15-18 Additionally, the measurement that provides the best information remains unclear and likely may depend on which method of SBT is used.

Pendelluft Resulting From Excessive Effort

The idea that mechanically ventilated patients with excessive effort can worsen or even cause lung injury is a concept referred to as patient self-inflicted lung injury. Whereas there is a strong physiological rationale for its existence, it is uncertain at this time if the efforts to prevent it (sedation, paralysis) would be more injurious than allowing excessive effort. Additionally, the bedside methods for determining excessive effort may not correctly identify those with the highest risk of patient self-inflicted lung injury. The use of EIT to determine the potential negative effects of excessive effort has been published.^6,19 The concept that localized strain can occur as patients draw volume not only from the ventilator but also from other areas of the lungs through the pendelluft effect. It's possible that excessive effort in a lung with homogeneous distribution of ventilation is less injurious than excessive effort resulting in pendelluft and localized strain within certain areas of a heterogeneous lung.

Perfusion

Similar to measuring the impedance of air across the thorax, EIT devices have the ability to measure the impedance of fluid to determine distribution of perfusion. This is normally done through a brief interruption in ventilation (apnea period is required) while fluid (typically hypertonic saline) is injected, and the impedance changes during this time are captured, and distribution of these impedance changes due to fluid is displayed on the screen. Additional software can provide a V̇/Q̇ analysis by overlapping distribution of ventilation and perfusion.^20,21 The ability to monitor perfusion is not universally available on all EIT devices at this time but likely will be. The titration of PEEP always has potential to impact perfusion through the lung, and the ability to assess changes due to changes in PEEP is a welcome addition.

Measurement Artifacts

Whereas many devices measure the impedance change from a baseline value, the baseline can be affected by movement, external pressure applied to the leads, and rapid changes in the patient’s fluid status. An active patient who is moving or repositioning themselves will greatly impact the baseline trend in EELI. However, patients who are completely passive can also have artifacts from pressure applied to electrodes. An example of this artifact is shown in Figure 5²² where the ICU bed is providing percussion that has an impact on the baseline (EELI) trend. Rapid changes in fluid status can impact the tissue reference that can cause a shift in the EELI baseline downward when fluid boluses are given or upward during aggressive diuresis (Fig. 6).²² It is worth noting that EIT measurements that assess regional compliance or distribution of ventilation are not impacted by these artifacts; only the assessments using the trends of baseline (ie, ΔEELI) are affected by these artifacts.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

The effects of automated bed mattress pulsation on the end-expiratory lung impedance measurements. From Reference 22.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

End-expiratory lung impedance (EELI) measurements following a bolus of furosemide. EELI continues to increase with no change in distribution of ventilation. From Reference 22.