Introduction

Obesity is an increasing problem, both worldwide and in Hong Kong, where the proportion of adults who were obese increased by over 40%, from 21.2% in 2003–2004 to 29.9% in 2014–20151,2. Obesity is reported to be associated with reduced lung function in previous research, as the adipose tissue reduces the maximum volume for the lungs to expand into, as well as increasing the resistance to the movement of air in and out of the lungs3,4,5. Obesity also increases the likelihood of comorbidities such as asthma6,7. In the assessment of chronic respiratory diseases, the classical method for lung function assessment is spirometry. However, spirometry has the drawback of only being able to measure the global function of the lungs, with regional lung function assessment not being possible, and is uncomfortable for those with difficulty using a large degree of effort.

Electrical impedance tomography (EIT) can be an alternative non-invasive diagnostic measure that does not involve ionizing radiation8,9,10,11. EIT has been employed in clinical use for more than a decade to provide qualitative measurement of ventilation in patients. Using this technology, non-invasive and low-cost tomographic imaging of the body is employed to detect and monitor several key diseases, such as chronic obstructive pulmonary disease (COPD), asthma8 and fatty liver12. Using EIT to assess the lung function can detect where the inhalation and expiration of air results in large changes of conductivity.

It has been suggested that higher percentage of body fat to be associated with lower lung volumes with a greater effect in men than women13. Furthermore, the distribution of fat can also affect the lung function. Intrathoracic fat, the volume of visceral abdominal fat and subcutaneous thoracic fat were inversely associated with forced expiratory volume in 1 s (FEV1) percentage predicted in male patients14. Given the knowledge of how body fat percentage and obesity can affect lung function, with the potential limitations of spirometry in lung function assessment, we propose the use of EIT as an alternative way of lung function assessment in obese patients.

We conducted the current clinical study on obese and non-obese subjects, to establish the feasibility of using EIT devices in detecting lung functional changes. Specifically, we aimed to establish the relationship between EIT-derived functional indicators and standard pulmonary functional testing results (i.e., the spirometry results).

Methods

The study assessed the lung function of obese and non-obese subjects using spirometry and EIT. The Gense-EIT lung device to be used is a potentially self-administrable, portable and non-invasive imaging modality capable of capturing the cross-sectional bioelectrical impedance distribution of the lungs up to 50 frames per second using EIT. Simultaneous paired EIT/Spirometer measurements are recorded when the subject goes through a 40 s forced exhale paradigm, repeated three times. The Gense-EIT lung device employs a band consisting of 16 equally distributed electrodes around the thorax of a subject. The electrode band is connected to a control unit that switches the electrode stimulation and measurement pairs and subsequently sends the collected measurements to a computational unit for image reconstruction. The images are then processed to extract functional lung images and indicators. This device can be connected to a computer, a smartphone, and/or a large LED screen, allowing the lung functional images and indicators to be visualized. In this study, all the participants performed spirometry and EIT in dedicated research facility in the presence of trained technicians, who instructed patients to perform spirometry as well as performing EIT examination for the patients. The spirometry and EIT were performed by 2 independent technicians.

Adult Chinese obese subjects with a BMI ≥ 25 and non-obese subjects with BMI < 25, based on the BMI cut-off for Asian population15; age from 18 to 75 were recruited in this study16. The participants’ age (± 5 years), sex, smoking status were matched among obese and non-obese subjects. All the participants were ambulatory with good physical functional status and could travel to the research facility for the assessment. All the included subjects had a structured questionnaire, together with cross checking with electronic medical records to ensure they do not have any clinical features of respiratory diseases. The exclusion criteria include subjects with underlying respiratory diseases, subjects with implanted electronic devices which would interfere with EIT assessment, subjects with spinal diseases, subjects who had with recent heart or lung surgery in past 3 months, pregnant women, subjects who were unable to perform spirometry and subjects whose exhale curve in EIT did not pass the noise-based quality criterion.

Spirometry was conducted using MIR Spirobank II®. EIT was conducted using Gense-EIT lung device. The subjects included had spirometry and EIT assessments the same day. The recruitment period was from 2024-01-12 to 2024-08-23.

Statistical analysis

The primary outcome of the study was testing the correlation of EIT assessed lung function in obese individuals against that of spirometry.

The demographic and clinical data were described as actual frequency, mean ± standard deviation (SD) or median [inter-quartile range (25th-75th centile)]. Baseline demographic and clinical data were compared between the two groups (with or without obesity) with independent t-tests (if normally distributed) or non-parametric tests (if not normally distributed). Chi square test was employed to compare the categorical variables in the two groups (with or without obesity). EIT-derived voxel-wise and regional bioimpedance, FEV1 and FVC, coefficient of variation (CV) and heterogeneity index (HI) parameters were measured/calculated. The correlation FEV1 and FVC (in Litres) measured by spirometry and EIT were assessed by Pearson correlation.

Ethical approval

The study was approved by the Institutional Review Board (IRB) of the University of Hong Kong and Hospital Authority Hong Kong West Cluster (UW 23–509). All the patients recruited signed the informed consent. The study was conducted in compliance with the Declaration of Helsinki.

Results

A total of 42 obese subjects and 35 non-obese subjects were included. 14 subjects were excluded based on an a priori curve index quality classification, as they had suboptimal EIT assessment limiting the interpretation. The mean age was 60 ± 14 years, with 62% of the subjects being male. The mean FEV1 and FVC measured by spirometry were 1.86 ± 0.69 L (77 ± 27%) and 2.57 ± 0.97 L (82 ± 30%). The mean FEV1 and FVC measured by EIT were 1.81 ± 0.77 L (74 ± 30%) and 2.76 ± 0.98 L (88 ± 31%). There was no statistically significant difference in the background clinical features and demographics among the obese and non-obese groups, except for their body weight and BMI. The baseline demographics of the subjects were illustrated in Table 1.

Table 1 Baseline demographics of the included patients.
Full size table

Correlation of EIT and spirometry parameters in obese and non-obese subjects

The correlation of FEV1 and FVC (in Litres) between spirometry and EIT among non-obese subjects measured by Pearson correlation coefficient were 0.673 and 0.541 respectively with both p-values < 0.001. This suggested a significantly strong correlation for FEV1 and a significant moderate correlation for FVC among non-obese subjects. The correlation was demonstrated in Fig. 1.

Fig. 1
Fig. 1
Full size image

Correlation of EIT and spirometry parameters (FEV1 and FVC in Litres) in non-obese subjects.

The correlation of FEV1 and FVC (in Litres) between spirometry and EIT measured by Pearson correlation coefficient among obese subjects were 0.705 and 0.568 respectively, with both p-values < 0.001. This suggested a significant strong correlation for FEV1 and significant moderate correlation for FVC among obese subjects. The correlation was demonstrated in Fig. 2.

Fig. 2
Fig. 2
Full size image

Correlation of EIT and spirometry parameters (FEV1 and FVC in Litres) in obese subjects.

Images from EIT, representing the change in conductivity between the middle and highest part of the breathing were also shown in Fig. 3. The two subjects, one obese and one control, have been chosen with similar spirometry indexes. They display similar EIT images, suggesting that they have similar performances and findings in EIT assessment.

Fig. 3
Fig. 3
Full size image

Lung conductivity image of (A) an obese subject and (B) a non-obese subject obtained from EIT.

Discussion

In the current study on lung function assessment by spirometry and EIT, we demonstrated moderate to strong correlation for lung function parameters (FEV1 and FVC) measured by both methods among obese and non-obese subjects. The findings in this study suggest that EIT is a potential tool for lung function measurement in patients regardless of their body build.

Currently, spirometry has been the gold standard in lung function assessment. However, spirometry has its own drawbacks which include the lack of spatial resolution and difficulty for patients who have hearing impairment or at extreme age to follow the instructions. As such, there is a need to look for alternative method for lung function measurement, with EIT being proposed.

Currently, the use of EIT among obese patients is mainly limited to those under general anaesthesia, to assess for development of atelectasis17, pretty much the same as the conventional use of EIT, which is limited to critically ill patients18,19. EIT demonstrated its role in guiding positive end-expiratory pressure (PEEP) optimization among patients with acute respiratory distress syndrome (ARDS) which improved lung compliance, reduced mechanical power and lowered driving pressure compared to traditional methods20. EIT has also been demonstrated to offer noninvasive bedside assessment of global and regional ventilation changes in the intensive care unit (ICU) setting, using various EIT indices such as tidal impedance variation, end-expiratory lung impedance, global inhomogeneity index and regional ventilation delay21. The benefits of EIT in having real-time tomographic functional images of the lungs allowed EIT to be introduced and used in dedicated aforementioned settings. It is also not surprised to see the use of EIT among coronavirus disease (COVID-19) patients with moderate-severe ARDS as a measure to titrate PEEP as in other ARDS patients22. Among obese patients, EIT also demonstrated its role in respiratory failure management among obese subjects, by reducing the mortality of these critically ill obese patients23.

The use of EIT among non-critically ill patients was also explored in recent years. There was a study suggesting the use of EIT for lung function assessment demonstrated good correlation among healthy non-obese subjects24. There is also ongoing development of a novel guided breathing paradigm which consists of a periodic inhalation and exhalation at 12 breaths per minute. Previous report suggested that such breathing paradigm could reflect both global and regional lung function changes among non-obese healthy subjects. This novel close-to-effortless guided breathing paradigm can potentially overcome a major hurdle of spirometry24. Regional lung assessment, including three-dimensional assessment, has been assessed with EIT as well19,25. EIT has also been shown to be useful in real-time monitoring, such as in acute pulmonary embolism26.

While EIT has been demonstrated to be able to predict standard spirometry indicators over a wide dynamic range, the data is mainly limited to healthy non-obese subjects. To our best knowledge, this is the first study reported the use of EIT in predicting spirometry values in obese subjects. The results are not only significant but also show comparable correlation as in subjects without obesity. This is a promising finding to suggest the high degree of correlation of lung function parameters measured by spirometry and EIT, which supports its use among subjects who are obese.

Th Gense-EIT lung device used in this study was a palm size and light-weight console with dimensions of 15.2 × 11.0 × 4.4 cm3 and < 300 g (Fig. 4). The portable console functions together with a reusable and disinfectable 16-channel electrode belt mobile app interface, and cloud-based processing pipeline. The portability function is supported through the integration of a power and battery management module enabling the complete system to operate with a constant power supply of 3.3 V; thus it can function either through a power socket or a Li-ion battery. This device has a major advantage over the other EIT devices which require a higher power supply ranging from ± 5 V (total of 10 V) to ± 15 V (total of 30 V). The Gense-EIT lung device allows clinical use in ambulatory settings given its small size with the lower power supply needed. Currently, the cost for 1 EIT exam is as low as USD 30, which is affordable when compared with spirometry. The costs of the Gense-EIT device is also cheaper than computed tomography machine by a factor of 100 or more, and also cheaper than traditional hospital-grade spirometry device by a factor of 50.

Fig. 4
Fig. 4
Full size image

(A and B) Gense-EIT device and (C) schematic diagram on the process of data acquisition/processing by EIT.

There are several limitations in this study to be addressed. Firstly, all the subjects did not have underlying respiratory diseases based on the screening questionnaire and medical record, together with relatively normal lung function in both spirometry and EIT. This may affect the applicability among patients with underlying respiratory diseases. Yet, the establishment of the basis with healthy subjects is needed before extending the subject to diseased subjects. Secondly, all the subjects included are Chinese which affects the generalizability to other ethnic groups, though there have not been any reports on the important difference in lung function assessment across different ethnic groups by EIT or spirometry while the lung function assessment by spirometry was based on local reference that is well established27. Thirdly, the presence of corrupted signal data due to inadequate electrode contact was caused by perspiration. This issue was particularly prevalent during the summer months in which this study was performed, when many subjects experienced heavy sweating. The mean age of the excluded subjects was 61.9 ± 17.2 years, with 9 (64.2%) being male. The mean BMI of the excluded subjects was 25.9 ± 4.9. Overall, the excluded subjects showed similar baseline characteristics as the whole cohort. Although remedial actions were taken, including allowing affected participants 10–20 min of rest in an air-conditioned room prior to testing, sweating persisted in some subjects, impacting the quality of the EIT data of almost 20% of the cohort. Based on this observation, we will recommend the patients to have a quick body cleansing mainly aiming at removing the sweat before EIT examination to ensure the quality of the EIT signals collected to be suitable for analysis. Another limitation is a relatively small sample size in this pilot study, and we believe conducting a larger scale study that includes obese and non-obese subjects, as well as including patients with or without underlying respiratory diseases, is warranted. While subjects without the symptoms or history of chronic respiratory diseases were included, we still noted some patients having impaired lung function, which could be due to undiagnosed respiratory diseases. However, this would not impair the interpretation of the findings as the purpose of this study was to assess the correlation of lung function assessment by spirometry and EIT among obese and non-obese subjects. Also, the lung function parameters did not significantly differ among the obese and non-obese subgroups.

Yet, what is presented here is just the beginning of the journey of EIT measurement among obese subjects. Moving a step forward, further refinement of EIT technique with higher resolution, more extensive lung coverage with more belts, potential three-dimensional reconstruction, as well as effort-independent breathing paradigms shall be the future of EIT development, which makes EIT a lung function assessment tool that can be done at low costs, high accuracy, and risk free.

Conclusion

EIT technology has comparable performance among obese and non-obese subjects. It can suggest that EIT is a potential tool for lung function measurement in obese subjects as in non-obese subjects.