Evaluation of electrical impedance tomography for determination of urinary bladder volume: comparison with standard ultrasound methods in healthy volunteers
- 18 Downloads
Continuous non-invasive urinary bladder volume measurement (cystovolumetry) would allow better management of urinary tract disease. Electrical impedance tomography (EIT) represents a promising method to overcome the limitations of non-continuous ultrasound measurements. The aim of this study was to compare the measurement accuracy of EIT to standard ultrasound in healthy volunteers.
For EIT of the bladder a commercial device (Goe MF II) was used with 4 different configurations of 16 standard ECG electrodes attached to the lower abdomen of healthy participants. To estimate maximum bladder capacity (BCmax) and residual urine (RU) two ultrasound methods (US-Ellipsoid and US-L × W × H) and a bedside bladder scanner (BS), were performed at the point of urgency and after voiding. For volume reference, BCmax and RU were validated by urine collection in a weight measuring pitcher. The global impedance method was used offline to estimate BCmax and RU from EIT.
The mean error of US-Ellipsoid (37 ± 17%) and US-L × W × H (36 ± 15%) and EIT (32 ± 18%) showed no significant differences in the estimation of BCmax (mean 743 ± 200 ml) normalized to pitcher volumetry. BS showed significantly worse accuracy (55 ± 9%). Volumetry of RU (mean 152.1 ± 64 ml) revealed comparable higher errors for both EIT (72 ± 58%) and BS (63 ± 24%) compared to US-Ellipsoid (54 ± 25%). In case of RU, EIT accuracy is dependent on electrode configuration, as the Stripes (41 ± 25%) and Matrix (38 ± 27%) configurations revealed significantly superior accuracy to the 1 × 16 (116 ± 62%) configuration.
EIT-cystovolumetry compares well with ultrasound techniques. For estimation of RU, the selection of the EIT electrode configuration is important. Also, the development of an algorithm should consider the impact of movement artefacts. Finally, the accuracy of non-invasive ultrasound accepted as gold standard of cystovolumetry should be reconsidered.
KeywordsElectrical impedance tomography Non-invasive Urinary bladder volume Ultrasound Cystovolumetry
maximum bladder capacity
body mass index
electrical impedance tomography
- L × W × H
formula calculated from length, width and height
It is estimated that around one million people in Germany suffer from bladder dysfunction . Dysfunctions of the storage and voiding function of the bladder are often associated with an overactive bladder with signs of urgency, frequent voiding and nocturia with or without urinary incontinence. Furthermore, neurological diseases and spinal cord injury may result in the loss of bladder sensation as well as uncontrolled and incomplete micturition. For therapeutic purposes like continence training based on biofeedback techniques and self-monitoring of patients, the knowledge of bladder filling in real time is essential .
Intermittent self-catheterisation is the method of choice for people with spinal cord injury and other patients suffering from neuropathic bladder to regularly empty their bladder and prevent overdistention, as the patients themselves cannot sense the urge to micturate . Nevertheless, self-catheterisation is often performed even though the bladder might not yet be full or, even worse, when the urine level has already reached a critical threshold leading to renal reflux.
Therefore, “portable” bedside ultrasound devices, so-called “bladder scan” (BS) are available that enable patients to self-check their bladder volume [4, 5, 6]. However, these devices are often bulky and the user has to be trained very well to achieve a more or less reliable outcome [2, 5]. The possible integration of an automatic non-invasive measurement system into a portable device would enable self-monitoring in patients with bladder dysfunction and also enhance their mobility.
Electrical impedance tomography (EIT) is proposed as an unobtrusive cystovolumetric technique to determine bladder volume continuously [7, 8, 9, 10]. This technique has been successfully applied in intensive care for monitoring of lung function [11, 12, 13]. For this measurement, a set of electrodes is placed around the torso and a small alternating current (AC) is injected via two of the electrodes into the body. The resulting surface voltages are measured between the remaining pairs of electrodes and provide information about the cross-sectional impedance distribution in the thoracic region and lungs, respectively.
There is no real distinction between the organ tissue of lung and urinary bladder with regard to electrical resistivity . Apart from their anatomy and function, the actual difference between these organs is that the lung is normally filled with air and the bladder is filled with liquid. However, as air is a very good isolator, whereas urine has almost the same electric conductivity as the surrounding tissue, this might lead to problems in bladder volume estimation by the EIT method. When calculating volume distributions based on differences in impedances, highest accuracy is achieved when electrical impedances between the medium in the hollow organ and the surrounding tissue differ substantially. Nevertheless, first attempts showed promising outcomes and the measured impedance shows a linear correlation with bladder volume for a given urine conductivity .
While the EIT electrode configuration and the signal analysis algorithm are well established for estimation of the lung volume, little is known about the optimal parameters for bladder measurement. Therefore, this pilot study aims to acquire preliminary data for the evaluation of (i) the EIT method for the estimation of bladder volume in comparison to standard ultrasound methods and (ii) Four possible electrode configurations with respect to their accuracy for non-invasive cystovolumetry by EIT. The EIT data are analysed by comparison with standard ultrasound and BS measurements, which are currently the gold standard for non-invasive cystovolumetry.
This prospective, monocentric pilot study was approved by the board of the local Ethics committee of the Medical Faculty at the University Hospital of RWTH Aachen (Trial registration EK 169/13) and has been registered at the WHO (Clinical Registration Number: DRKS00012871).
Prior to the EIT study, the participants were advised to keep a drinking and micturition diary (at home) using a measurement pitcher for at least 3 × 24 h, to get baseline data on micturition volumes and frequency and to ensure normal physiology of their bladder and drinking behaviour.
The “1 × 16” is the classical configuration, used in lung patient surveillance, whereas the other configurations were chosen due to their promising outcome in finite element (FEM) simulations for cystovolumetry, as has been shown by Schlebusch et al. .
Calculation of bladder volume
Volume validation was carried out by uroflowmetry consisting of a measurement pitcher, an electronic balance with a weight transducer and corresponding software (Flow, Laborie, Montreal, QC, Canada) to record the flow rate of external urinary stream as volume per unit time [ml/s] according to recommendations of the standardization committee of the International Continence Society (ICS) and urine collection in the pitcher to compare the accuracy of EIT to ultrasound and bladder scan volumetry . Validation of the actual bladder volume was performed by measuring the weight of the voided urine with an electrical scale. Immediately after voiding, calculation of RU was performed via standard ultrasound and bladder scan. Afterwards, the voided RU was collected in the weight measurement pitcher, without EIT measurement. The BCmax at the point of urgency was determined as micturition volume plus RU from the measurement pitcher. This procedure was performed twice for each participant and for each electrode configuration.
Results from the measurement pitcher (in grams) were converted into a volume using a conversion factor known from the literature, i.e. 1.02 g/ml . Because multiplication of the urine weight by this factor showed very good correlation with the urine volume (99.9%), for further measurements only the pitcher volume was used for validation of the measurements. For EIT volume estimation, one of the voiding measurements of each volunteer was used as calibration data for the estimation of BCmax and RU from the second measurement.
The EIT method
For statistical analysis the Shapiro–Wilk test was performed to test for normal distribution. Because the US, BS and EIT data showed a normal distribution, Student’s t test was used to determine significant differences between the measurement methods and the EIT configurations. A p-value < 0.05 was considered to indicate statistical significance. The software used was OriginPro (2017G, Origin Lab Corporation, Northampton, USA). If not indicated elsewhere values are shown as mean values plus/minus standard deviation.
Participants’ baseline data
Female participants had normal BMI values (mean 22.1 ± 3.0 kg/m2) whereas the values of the men tended towards overweight (mean 25.2 ± 3.9 kg/m2), but the difference for both gender was not significant. Mean baseline micturition values surveyed via the bladder diary were within the range as described for healthy men and women . Here, the mean micturition frequency in women was 8.3 ± 1.2/day and in men was 6.4 ± 2.2/day (7.1 ± 2.1/day for men and women combined). Mean micturition volume was 284 ± 93 ml in women and 317 ± 45 ml in men (combined: 305 ± 63 ml). This results in a total volume of 2337 ± 627 ml/day for women and of 1975 ± 778 ml/day for men (combined: 2110 ± 702 ml).
Comparison of baseline and study data
Evaluation of measurement accuracy of US, BS and EIT
Evaluation of electrode configurations
As shown by Leonhardt et al. EIT is a promising tool for the estimation of bladder volume in individuals with spinal cord injury . However, in individual patients, the differing amounts of body fat can alter the results or impede cystovolumetric measurements . Therefore, for this study, data on BMI were acquired to align BMI data with the impedance data, if needed. However, since the healthy participants in this study showed no significant differences in BMI there was no need to create a correlation factor for this parameter. Nevertheless, this might prove to be an issue in future studies.
During the study days, the non-physiological liquid ingestion of the (highly motivated) participants lead to a raised renal function resulting in increased diuresis . Comparison of the micturition diaries with micturition volume revealed that, in normal life, none of the participants had an abnormal drinking or micturition behaviour. Nevertheless, the resulting high bladder volumes and small amounts of RU allowed to test the standard US and BS measurements, together with the new EIT method, under extreme conditions.
Although some studies have reported the superiority and reliability of ultrasound as a clinical tool, others have reported problems related to the measurement of bladder volume [22, 23, 24, 25]. According to Carter et al., the ellipsoid US remains the best method for use in the urological clinic. However, in the present study this was not apparent since the L × W × H showed high similarity to the ellipsoid US . With respect to the validation of EIT cystovolumetry in our study, the real voided urine volume had to be used to correlate these data to our impedance signals, because the ultrasound data were not reliable.
Nevertheless, this study allowed us to compare standard US techniques with the proposed EIT-based cystovolumetry. The US methods show a considerable error for estimation of both BCmax and RU, whereas bladder EIT shows slightly better accuracy than US and BS for BCmax estimation. For the estimation of RU, all methods show decreased accuracy compared to the estimation of BCmax, and EIT, when combining all four configurations tested, shows a higher error than the standard US and BS methods. A possible explanation for the reduced sensitivity of the 1 × 16 and 2 × 8 EIT configurations to low bladder volumes is that the EIT electrode positions are fixed on the body and, at lower volumes, the bladder falls below the electrode plane behind the symphysis thereby lowering the sensitivity of the EIT system to bladder volume changes. Therefore, the electrode configuration to be used in future should cover a larger transversal volume and enable measurement of all levels of bladder filling, as realized by the “Matrix” or “Stripes” configurations.
Although our aim was a continuous measurement of the bladder volume during micturition with the EIT method, we encountered major measurement problems in form of movement artefacts and additional stomach pressure due to abdominal muscle contraction which caused volume-independent impedance changes (see Fig. 5). Therefore, we could only compare the full with the voided bladder. The impact of these comparably small muscle tensions must be taken into account in the case that EIT electrodes are integrated into fabric [27, 28].
These are additional outcomes that, on the one hand, show the limitations of bladder volume estimation itself and the need for improvement of the standard devices or development of new devices. On the other hand, they can help in the development of our medical EIT device and the specifications for a new cystovolumetry algorithm.
The use of standard ultrasound as a reference method for the development of a new EIT bladder volume device has to be neglected, as there was a large offset between the measurements and actual bladder volume. Volume validation by means of a measurement pitcher and a weight scale is recommended as reference.
EIT, in comparison to standard ultrasound-based measurement systems, shows considerable potential as a cystovolumetry system. However, before being used in everyday applications and as a continuous method, its accuracy needs to be improved by minimizing the negative influence of movement artefacts and by optimising the electrode configuration and calibration.
DL designed and performed the study, analysed the data and wrote the manuscript. CC analysed the data and wrote the manuscript. TS designed and performed the study. MR, RR analysed the data and revised the manuscript. MW, SL supervised the study and revised the manuscript. JOG designed, performed and supervised the study, analysed the data and revised the manuscript. All authors read and approved the final manuscript.
The authors thank the research group of the Department of Urology, especially Katja Stollenwerk, Carmen Fera, Isabella Zraik and Patrick Arndt for technical support. We also thank Angela Habier, from CTC Aachen, for assistance with drafting of the documents for the Ethics Committee.
The authors declare that they have no competing interests.
Availability of data and materials
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Written informed consent for publication was given by all participants.
Ethics approval and consent to participate
The study was approved by the Ethics Committee of the Medical Faculty at the University Hospital RWTH Aachen (EK 169/13, Ethik-Kommision an der medizinischen Fakultät der RWTH Aachen) and has been registered at the WHO (Clinical registration number: DRKS00012871). All investigations were performed according to the principles stated in the Declaration of Helsinki. Written informed consent was given by all participants.
This study was funded by the Federal Ministry of Education and Research, Germany (BMBF, 13EZ1128A). The sponsors had no control over the design of the study, the collection, analysis, and interpretation of data or the writing of the manuscript.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.S. Bundesamt. Diagnosedaten der Krankenhäuser ab. 2000. www.gbe-bund.de. Accessed 25 Jan 2018.
- 3.Guttmann L, Frankel H. The value of intermittent catheterisation in the early management of traumatic paraplegia and tetraplegia. Paraplegia. 1966;4(2):63.Google Scholar
- 4.Beauchamp-Parent A, Sawan M. New reconfigurable ultrasonic enuresis monitoring system. In: 10th annual international conference of the IEEE-engineering-in-medicine-and-biology-society, Hong Kong, Peoples R China, vol. 20, pp. 789–792, 1998.Google Scholar
- 15.Schlebusch T, Leonhardt S. Effect of electrode arrangements on bladder volume estimation by electrical impedance tomography. In: 15th international conference on electrical bio-impedance (ICEBI)/14th conference on electrical impedance tomography (EIT), Heilbad Heiligenstadt, Germany, vol. 434, 2013.Google Scholar
- 17.Williamson MA, Snyder LM. Wallach’s interpretation of diagnostic tests. 9th ed. Philadelphia: Lippincott Williams & Wilkins; 2011.Google Scholar
- 18.Lionheart W, Polydorides N, Borsic A. The reconstruction problem. In: Holder D, editor. Part 1 of electrical impedance tomography: methods, history and applications, Chap 1. Manchester: The University of Manchester; 2004, pp. 3–64.Google Scholar
- 24.Mendez A, Sawan M. Chronic monitoring of bladder volume: a critical review and assessment of measurement methods. Can J Urol. 2011;18(1):5504–16.Google Scholar
- 26.Tubaro A. et al. Imaging and other investigations. Incontinence, vols 1 and 2 (vol 1: Basics and evaluation—vol 2: Management) pp. 707–797. 2005.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.