Extending the range of test meals for gastric emptying studies by using multi-frequency EIT.

C.T Soulsby*, A. Romsauerova, R.Yerworth, L. Horesh, D.F. Evans*, D Holder

*Adult and Paediatric Gastroenterology, Bart’s and The London School of Medicine, London.

Department of Medical Physics, University College London.

Introduction. The use of EIT to measure gastric emptying was first introduced in the late 1980’s following validation studies carried out by the Sheffield Group[1]. EIT continues to be used in a small number of GI physiology units that specialise in measuring gastrointestinal motility. The system most commonly used is the Sheffield Mark I (Medical Physics Department at Sheffield University) which was the first commercially available system. Most foods have a low conductivity and cannot be imaged using the Sheffield Mark I as there is insufficient contrast between the conductivity of the food and stomach tissue to allow the region of interest to be visualised. The aim of this study was to investigate whether using a multi-frequency EIT system (UCLH Mark II) to extend the range of measurement frequencies would allow food with low conductivity to be imaged.

Methods: Subjects attended following an overnight fast. Sixteen Blue Sensor ECG electrodes were placed around the midway point between the sternum and the umbilicus; starting at the anterior midline, a tape measure was used to ensure the electrodes were evenly spaced and lay in the same plane. Electrode contact was checked using a reciprocity check, if results fell below the threshold value, electrodes were repositioned Wide bore nasogastric tubes (Corflo ™ 12 Fr Merck) were passed. The test meals were four different solutions: distilled water (DW), 0.2% saline and 0.9% saline and enteral feed (Nutrison Standard, Nutricia). Subjects drank the 400mls of each of the liquids orally as rapidly as possible (within 30 seconds). Each liquid remained in the stomach for a maximum of 60 seconds and was then rapidly aspirated via the nasogastric tube (within 120 seconds). Rapid measurement was used to minimize the effect of gastric secretions and gastric emptying. EIT images were taken for 30 second periods, with images being taken prior to ingestion, while the stomach was full and after gastric contents had been aspirated. Using the image when the stomach was most full, the ratio of the signal within the ROI to the noise in the rest of the image the signal to noise ratio (SNR) was calculated. Liquid conductivity pre and post gastric placement was measured on the Hewlett Packard impedance analyser at frequencies from 1kHz to 1MHz by placing samples of each solution were placed in a series of 3 Perspex tubes (0.10m, 0.049m and 0.033m length, diameter 0.0015m) with silver electrodes at either end.

Figure 1 showing subject with electrodes placed around the abdomen and wide bore nasogastric tube in situ

Results: As there were 30 frequencies, 30 SNR’s were calculated for each solution. Results are shown in the table below. The 0.9% saline had the highest conductivity and the highest mean SNR for all frequencies. Distilled water had the lowest conductivity and a lower SNR than 0.2% and 0.9% saline. Enteral feed had the lowest mean SNR for all frequencies. The frequencies with the highest SNR were identified and are shown in the table 1 below.

Distilled water / 0.2%saline / 0.9%saline / Enteral feed
Conductivity (Sm-1) (range) / 0.213
(0.20-0.23) / 0.60
(0.51-0.70) / 1.67
(1.62-1.74) / 0.97
Frequencies where SNR was highest (kHz) / 161.3, 203.3, 256.0, 322.5, 406.4, 512.0, 1024.0, 1625.5 / 12.7, 16.0, 32.0, 40.3 / 64.0, 256.0, 322.5, 1290.2 / 2.0, 3.2, 4.0, 5.0, 6.4, 8.0, 10.1, 12.7
Mean SNR for best frequencies / 3.70 / 4.74 / 4.73 / 4.06
Mean SNR for all frequencies / 3.50 / 3.96 / 4.12 / 3.32

Table 1

Discussion:

The higher the signal to background noise (SNR) the better the image, thus the best current frequency for imaging each of the solutions could be identified by finding the frequency that produced the highest SNR. The measurement conditions that would create this clear image with a high SNR would be when the difference between stomach and solution conductivity was greatest.

The best frequency could have been predicted by comparing the solution conductivity measured on the Hewlett Packard with the known conductivity of stomach muscle and identifying which frequencies would create the greatest conductivity difference. The range of stomach muscle conductivity is 0.5- 0.75Sm-1 [2] (at frequencies from 100Hz to 4MHz). The conductivity of the aspirated enteral feed was 0.9 Sm-1 so the greatest difference would occur at the lowest frequencies, and it was frequencies 1 to 10 which gave the highest SNR’s. For the 0.9% saline the conductivity was much higher than stomach muscle so it was expected that clear images would occur at all frequencies, which is confirmed by the high mean SNR. So, for highly conductive meals clear images can be obtained at any frequency. For less conductive meals, such as enteral feed or 0.2% saline, care should be taken to select the frequencies which will produce the best images. Although conductivity of solutions can easily be increased by adding extra free ions (such as sodium chloride) there are clinical conditions when this is not be possible. Therefore the ability to use lower frequencies to image less conductive meal offers the potential to measure gastric emptying in patients who could not previously be measured with single frequency EIT systems [3].

Referrences

1. Avill R, et al., Applied potential tomography. A new noninvasive technique for measuring gastric emptying. Gastroenterology, 1987. 92(4): p. 1019 - 1026.

2. Gabriel C and Gabriel S, Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies: www.brooks.af.mil/AFRL/HED/hedr/reports/dielectric/home.html.

3. Soulsby C, Y.E., Evans DF, Powell-Tuck J, Can electrical impedance tomographic spectroscopy (EITS) be used to measure enteral feed tolerance? Clinical Nutrition, 2002. 21: p. 48.