The evaluation of regional pulmonary perfusion is of great interest for cardiopulmonary management in the intensive care unit (ICU). For many years, researchers have been searching for imaging techniques suitable for the assessment of pulmonary blood flow. Recent years have witnessed substantial progress in this area, and several techniques for pulmonary perfusion assessment, including MRI, CT, PET, and single-photon emission CT (SPECT), have become available (Hopkins et al., 2012; McCollough et al., 2015; Bondesson et al., 2019; Kohli et al., 2019). However, these methods are rarely used in critically ill patients because of the non-bedside implementation, high risk of transfer, inconvenience, etc.

Electrical impedance tomography uses a single-ring external electrode to image changes in impedance within a transversal section of the lungs; it is a non-invasive, radiation-free imaging modality that enables the continuous bedside monitoring of ventilation. A multitude of studies has confirmed the feasibility and reliability of ventilation images obtained by electrical impedance tomography (EIT) to inform mechanical ventilation settings (PEEP titration, lung recruitment, etc.) in critically ill patients (Frerichs et al., 2017; Yoshida et al., 2019). Electrical impedance tomography is also able to assess lung perfusion distribution. There are two EIT methods used for assessing lung perfusion: the pulsatility and conductivity contrasting (saline) bolus methods. Recently, increasing attention has been given to the application of the saline bolus EIT method in the visual evaluation of regional pulmonary perfusion in severe patients. This study seeks to provide an overview of experimental and clinical studies with the aim of clarifying the progress made in the use of the saline bolus EIT method.

The pulsatility-based EIT method of measuring cardiac-related impedance variation, which has been widely investigated in both unwell and healthy subjects (Frerichs et al., 2009; Grant et al., 2011; Li et al., 2014), features non-invasive and real-time perfusion monitoring. In the last two decades, developments in EIT signal separating techniques involving frequency-domain filtering (Frerichs et al., 2009), ECG gating (Vonk Noordegraaf et al., 1998), respiratory pause (Fagerberg et al., 2009), and principal component analysis (PCA) (Deibele et al., 2008) have optimized pulsatile signal processing. We summarized the clinical studies using pulsatility-based EIT assessments for pulmonary perfusion in healthy volunteers and patients with pulmonary hypertension (PH) (Table 1).

The basic principle of the pulsatility method for estimating pulmonary perfusion is based on the measurement of pulsatile changes in pulmonary blood volume instead of real forward lung blood flow. Hence, many potential factors (such as changes in vascular tone, synchronous changes in air content, airway pressure, and distensibility of the small pulmonary vessels) could affect the accuracy of using cyclic pulsatile changes in regional impedance to reflect the proportion of lung blood flow in those regions. The validity of the pulsatility method for assessing perfusion has been questioned (Borges et al., 2012; Putensen et al., 2019). Studies have found that downstream pulmonary vascular resistance and the distensibility of small pulmonary vessels can impact EIT pulsatility data (Smit et al., 2004; Borges et al., 2012). Borges JB et al. found that pulsatile impedance increased within the collapsed lung region, which had reduced lung blood flow based on both the SPECT and saline bolus EIT methods (Borges et al., 2012). Although the pulsatility method could easily provide continuous information on lung perfusion, the conductivity contrast bolus injection method showed superiority in terms of feasibility and accuracy (Frerichs et al., 2002; Deibele et al., 2008; Borges et al., 2012).

The conductivity contrast bolus method, also named contrast-enhanced EIT or the saline bolus EIT method, was performed by introducing a bolus of hypertonic saline into the pulmonary circulation during a respiratory hold. During the breath-hold, tidal impedance was lacking; however, chest impedance was relatively constant. Hence, the change in impedance caused by the saline bolus reflects forward lung blood flow. The high variation in the regional impedance indicates more saline passing through the corresponding region, which also indicates high local lung perfusion. According to the typical first-pass kinetics manifested in indicator-enhanced CT and MRI (Konstas et al., 2009; Ohno et al., 2016), hypertonic saline following the blood flow from the right atrium through the pulmonary circulation is indicative of a significant decline in electrical impedance due to its high conductivity. The impedance–time curves produced by saline boluses were reconstructed to fit a gamma variate model and were quantitively analyzed based on the Fick principle and the accumulated mass of the indicator in every region of interest (Mullani and Gould, 1983).

In general, there were two methods used to analyze the indicator dilution, namely, the transfer model and the slope analysis. The slope analysis was validated against PET in animal studies (Borges et al., 2012; Bluth et al., 2019). Several recent publications on lung perfusion using EIT and saline bolus injections in human subjects adopted the slope analysis (He et al., 2020a,b; Mauri et al., 2020; Spinelli et al., 2021). A high slope in the lung regional time–impedance curves obtained following a saline injection indicates a high accumulation of saline, which also reflects a high lung blood perfusion.

More recently, Kircher et al. optimized an EIT-gamma function fitting based on a linear piecewise approximation of the drift model in a porcine acute respiratory distress syndrome (ARDS) model, simplifying the quantization algorithm for potential clinical applications (Kircher et al., 2021). Simultaneously, following the administration of a saline bolus, by calculating the SD of each pixel and plotting them at the relevant image position, a new type of EIT image presenting the exact position, into which the indicator is carried by the bloodstream and visualizing the temporal variation in perfusion distribution can be generated (Frerichs et al., 2002).

The experimental and clinical studies using the saline bolus EIT method to visually assess lung perfusion are summarized in Tables 2–4 (Frerichs et al., 2002; Borges et al., 2012; Nguyen et al., 2013, 2015; Reinius et al., 2015; Hentze et al., 2018; Bluth et al., 2019; Grassi et al., 2020; He et al., 2020a,b,c, 2021b; Mauri et al., 2020; Safaee Fakhr et al., 2020; Kircher et al., 2021; Spinelli et al., 2021; Yuan et al., 2021). The administration of saline boluses, data analyses, precision, and cutoff values differ among studies, and a consensus must be reached regarding the clinical application of the saline bolus EIT method.

The availability of contrast EIT presented in multiple experimental and clinical studies allowed the instant measurement of lung regional perfusion changes under various pathological conditions, such as PE, atelectasis, and ARDS (Frerichs et al., 2002; Borges et al., 2012; He et al., 2020a,c; Mauri et al., 2020; Kircher et al., 2021). A remarkable agreement between EIT and SPECT further proved the correctness and practicability of this emerging perfusion imaging method.

Saline bolus EIT can determine differences in regional perfusion resulting from perfusion distributions reflecting the diseases, therapeutic maneuvers (e.g., administration of vasoactive agents, and mechanical ventilation), and ventilation/perfusion maps of multiple pathophysiological phenomena. Therefore, it is possible for the contrast EIT method to have clinical applications in the future. First, saline bolus-based EIT might have the potential to identify various etiologies of acute respiratory failure (ARF). Recently, He et al. demonstrated that contrast EIT-derived regional ventilation and perfusion measurements were able to characterize three broad ARF etiologies, including pulmonary embolism-related disease (PED), diffuse lung involvement disease (DLD), and focal lung involvement disease (FLD) (He et al., 2021b). Second, for mechanically ventilated patients, increasing clinical evidence has shown that perfusion EIT could provide bedside guidance for tidal setting, PEEP titration, and prone ventilation and help physicians evaluate the prognosis of patients with ARDS. Third, this saline bolus-based EIT method featured with convenience and promptness might be able to satisfy the clinical use for monitoring therapeutic effects for severe patients, for example, to assess the effects of inhaled medicine to PH and the therapeutic effect of anticoagulation/thrombolytic therapy on PE.

For the extensive application of the saline bolus EIT lung perfusion method in the ICU, more clinical researches and experimental studies are required to confirm the accuracy and validity of saline contrast EIT. Importantly, there is a strong need for a consensus on the cutoff value to determine pulmonary perfusion parameters, like the one for ventilated/perfused units in assessing V/Q mismatch (He et al., 2020b,c; Mauri et al., 2020; Spinelli et al., 2021). Although there has been little evidence showing the feasibility of EIT perfusion imaging in patients with spontaneous breathing, there is a great chance that breath-holding maneuvers adjustive to awake patients like inspiratory holds using typical entry standards are next in line. Furthermore, to achieve the continuous monitoring of pulmonary blood flow, it is foreseeable that the combination of contrast EIT with pulsatility-based EIT for dynamic assessments will be explored in the future.

However, the EIT analysis has drawbacks in providing detailed morphological information to accurately identify the vertical location of perfusion deficits attributed to restrained spatial resolution. Hence, the reconstruction algorithm warrants further optimization to provide robust evaluations in clinical applications. Additionally, using a single-ring external electrode, the EIT method is only allowed to monitor one cylindrical lung part at one time, instead of providing an overall lung image like CT and MRI. Moreover, auxiliary evaluations or data processing procedures to address the conductivity variations are necessary to accommodate background conductivity changes when a conductive contrast agent is used (Bluth et al., 2019). Finally, despite the validation in intubated subjects, the feasibility of perfusion EIT in spontaneous respiratory patients is yet to be fully understood. Since the effectiveness of the breath-holding maneuver may vary due to the individual variations in conscious patients, contrast EIT focusing on applications in non-mechanically ventilated patients demands future discussion before being used in clinical practice.

Using a hypertonic saline solution based on first-pass kinetics, contrast EIT imaging generating dilution impedance waveforms is reliable for determining lung perfusion changes reflecting atelectasis, spatial heterogeneity, and artery embolism in a regional manner. With the ability to monitor regional lung perfusion and produce V/Q mapping, saline bolus EIT is promising and meaningful for patients in ICUs. Further study is required to validate the impact of the described saline bolus EIT method on decision-making, therapeutic management, and outcomes in critically ill patients.

MX, HH, and YL contributed to conception and design of the review. MX and HH searched and organized the database. MX wrote the first draft of the manuscript. HH and YL wrote and refined several important sections of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

This study was supported by the CAMS Innovation Fund for Medical Sciences (No. CAMS Innovation Fund for Medical Sciences (No. 2020-I2M-C&T-B-042), Capital's Funds for Health Improvement and Research (No. 2020-2-40111), Excellence Program of Beijing Clinical Key Specialty for Critical Care Medicine (2020), and Beijing Municipal Science and Technology Commission (Grant No. Z201100005520051).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.