Comparison Of A Portable Breath Hydrogen Analyzer Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The most commonly used methods to diagnose hypolactasia rely on the indirect measurement of lactose breakdown products following an oral dose of lactose. The breath hydrogen test is based on the fact that bacteria in the colon ferment undigested lactose, causing the release of hydrogen, which is then absorbed through the intestine wall into the blood circulation. In some people, the hydrogen is converted to methane by methogenic bacteria in the colon. The hydrogen and methane are transported from the capillary blood to the lung alveoli and then exhaled. An increased concentration of at least 20 ppm in exhaled hydrogen or increase of 5 ppm of exhaled methane following lactose ingestion, as compared with the pre-test value is generally accepted as the threshold to reveal lactose maldigestion(1). The combined measurements of the exhaled concentrations of hydrogen and methane provide more reliable indicators(2) of the fermentation processes in the large intestine than results based on measurements of breath hydrogen alone.

It is desirable to find an easy, cheap and reliable method of diagnosing lactose maldigestion. A rapid breath hydrogen analyzer [Sensistor AB, Linkoping, Sweden] to detect lactose malabsorption was described first by Berg who used the hydrogen in expired air to give a voltage change that can be transformed into a 'ppm' value from a calibration curve(3). Rumessen used an exhaled H2 monitor [Gas Measurement Instruments Ltd, Renfrew, Scotland] which is Hydrogen sensing device using an electrochemical cell comprising of working, reference and counter electrodes. The potential of the working electrode is held constant against the reference electrode. H2 is oxidized at the working electrode generating a current which is proportional to the amount of H2 diffusing and the signal is converted to give a digital display(4).

Since then, different devices have been constructed to facilitate measuring breath hydrogen levels in the clinical setting. These include hand held devices that give an instant result and those based in laboratories where bags of exhaled breath are taken to be analyzed after completing the lactose challenge. Bedfont EC 60 Hydrogen monitor and the newer Bedfont Gastro+ [Bedfont Scientific Ltd, Maidstone, UK] use an end tidal sampling system which allows diffusion of expired air directly into the electrochemical sensor. The electric output from the sensor is fed to a liquid crystal display which displays results in parts per million. Another device which is commonly used in the measurement of expired hydrogen is Micro H2 [Micro Medical Limited, Kent UK] which also provides an instant value for breath hydrogen using a sensor drift detection technique. A study comparing a portable breath hydrogen analyzer Micro H2 [Micro Medical Limited, Kent UK] with a Quintron MicroLyzer, Model DP [Quintron Instrument Company, Milwaukee USA] which uses gas chromatography technique where expired gas is collected in a bag and taken to this machine in a lab and analyzed was carried out by Peuhkuri et al (5). It showed a 100% concurrence between, the two methods for the diagnosis of lactose maldigestion using breath hydrogen analysis in 44 cases. They concluded that the Micro H2 appeared as reliable for measuring breath hydrogen concentrations as Quintron MicroLyzer in order to diagnose hypolactasia(5).



The study population consisted of 134 subjects, 55 patients with Ulcerative Colitis, 49 with Crohn's disease & 30 Healthy volunteers. The study populations were all of Caucasian origin. All IBD patients were in clinical remission at the time of the study as determined by Harvey-Bradshaw Index(6) score of 4 or less in those with Crohns Disease and by the Simple Clinical Colitis Activity Index(7) score of 3 or less in those with Ulcerative colitis. The study was approved by the South East Wales Research Ethics Committee.


As a reference analyser for detecting the breakdown products of lactose metabolism in exhaled air, the Quintron MicroLyzer Self-Correcting Model SC [Quintron Instrument Co. Inc., Milwaukee, WI, USA] was chosen as this was the analyser which is used as the standard in the laboratory currently where I was based as described in chapter 2

The Micro H2 [Micro Medical Limited, Kent, UK] a portable, hand-held, hydrogen monitor which is designed to give instant results was used for comparison. The measurements are obtained using 22-mm mouthpieces [Bedfont EC50-MP/200] for adults. The technical data for both the analyzers is summarised in Table 4.1.

The calibration of both analyzers was carried out with Microcan-Disposable Calibration Gas from MicroGas UK Batch [3335/1006] which contains 100 ppm of Hydrogen, 50ppm of Methane and 6% Carbon Dioxide and balance Air [UN1956] 20 litres at 20°C and 300 psig for the Quintron MicroLyzer. The Micro H2 was calibrated using Microcan gas [MCG 100] which contains 50ppm of Hydrogen UK batch [3330/0075].

Table 4.1: Technical data of the Micro H2 monitor, and the Quintron MicroLyzer SC, according to the manufacturer's manual.

Micro H2 Hydrogen Monitor

Quintron MicroLyzer SC











0.16 kg

7.2 kg

Power Supply

Single 9 V PP3

220 V

Calibration Gas

H2 50 ppm

H2 100 ppm,

CH4 50 ppm & CO2 6%

Sample size

Single breath

20mls of breath [minimum]

Results Display

Liquid Crystal Display

Liquid Crystal Display


Before the test, participants were told not to eat or drink anything other than water from midnight the night before the test which commenced at 9am. They were also told not to smoke for at least 4 hours before the test. In addition, they were advised to choose lactose-free food and avoid those foods that they recognize will produce gastrointestinal symptoms in the preceding 3 days. They had not received any antibiotic treatment or had bowel preparation for gastro-intestinal investigations for 4 weeks before the study. The subjects were given 50grams of lactose [Lactose powder BP: BN: M07001589 MS/13880/1 North Staffordshire Hospital Trust Pharmacological Services] dissolved in 300mls of water. Breath samples were obtained before the ingestion of the lactose and at 30 min intervals for 3 hours and then an hour later. On each occasion, the subject exhaled into the Micro H2 and then straight afterwards they breathed into a polyethylene bag via a one way valve. Once the bag has fully inflated, the collected sample was stored to be later analyzed by the Quintron MicroLyzer which measured both breath hydrogen and methane values. The test was defined as positive if the rise in Hydrogen or Methane or both is 20 ppm and 5 ppm above the baseline/nadir, respectively.


Sensitivity, specificity and positive and negative predictive values were calculated in order to evaluate the Micro H2 method compared with the breath hydrogen results from the Quintron MicroLyzer. Sensitivity [%] = true positives / [true positives + false negatives] X 100, Specificity [%] = true negatives / [true negatives + false positives] X 100, Positive predictive value [%] = true positives / [true positives + false positives] X 100 & Negative predictive value [%] = true negatives / [true negatives + false negatives] X 100. The Bland-Altman plot, or difference plot, was used to compare the two measurement techniques(8). In this method the differences between the two techniques are plotted against the averages of the two techniques. Horizontal lines are drawn at the mean difference, and at the limits of agreement, which are defined as the mean difference ± 1.96 times the standard deviation of the differences. The data was entered and analysed by statistical software SPSS version 12 [Chicago, USA].


Out of the total 134 participants there were 68 females and 66 males in the study. The age range was 19-86 years and their mean age was 44.98years. The mean age for male patients was very similar to female participants 44.9 Vs. 45 years and the age range was 20-86 years for male and 19-81 years for female participants.

The results obtained with Quintron MicroLyzer which was taken to be the standard, showed 39 [29.1 %] participants had a positive test and therefore 95 [70.9%] had a negative result. The test was positive in 18/66 [27.3%] men and 20/68 [30.9%] females participants. Of the 39 who had a positive test as determined by the results from the Quintron MicroLyzer, 19 [48.7%] were positive for Hydrogen [H2] production, 17 [43.6%] were positive for methane [CH4] and 3 [7.7%] were positive for both H2 & CH4.

The results obtained with Micro H2 analyzer which is only able to detect hydrogen, showed that 21 [15.7%] participants had a positive test and 113 [84.3%] had a negative test. The test was positive in 8/66 [12.1%] men and 13/68 [19.1%] females participants. The Micro H2 picked up 21/22 [95.5%] cases that the Quintron MicroLyzer had detected significant rises in breath hydrogen. However, it missed one person i.e. 1/22 [4.5%] cases. This was a 26 year old female with Ulcerative colitis for 9 years. The rise in Hydrogen from nadir was 10 ppm with Micro H2 and it was 28 ppm with Quintron MicroLyzer.

If the results of lactose metabolism were just based on hydrogen detection alone then the Micro H2 had a Sensitivity of 95.5%, a Specificity of 100%, and a Positive Predictive value of 100% and a Negative predictive value of 99.1%.

However, the gases produced by metabolism of lactose are both Hydrogen and Methane. The Micro-H2 does not measure methane levels and so it would have missed 17 other cases where there was a significant rise in the levels of this gas after lactose ingestion. This means that in the subjects studied, 17 [43.6%] cases with results from the Quintron MicroLyzer suggestive of hypolactasia, were due to methane production and would not have been picked up if only this handheld device was used. The Micro H2 therefore has an overall Sensitivity of 53.9%, a Specificity of 100%, a Positive Predictive value of 100% and a Negative predictive value of 83.2%. The results are summarized in table 4.2.

Tables 4.2: Comparison of Micro H2 results with Quintron MicroLyzer in measuring the breakdown products of Lactose metabolism.


Quintron MicroLyzer









18 [17 CH4 & 1H2]







The amount of H2 detected by the Micro H2 ranged from 0 ppm to 78 ppm when compared to 0 ppm to 227 ppm with Quintron MicroLyzer. The highest increase in the breath hydrogen concentration over the baseline after ingestion of lactose was extremely variable and ranged from 4 to 78 ppm measured with the Micro H2, or from 8 to 217 ppm measured with the Quintron MicroLyzer. The increase in hydrogen from the basal values to the highest recorded ones during the test is also shown in figure 4.1. The amount of methane detected was 0 ppm to 157 ppm with Quintron MicroLyzer with the highest rise in methane ranged from 0 ppm to 157 ppm.

Figure 4.1: Basal and Maximum Hydrogen rise between Micro-H2 and Quintron MicroLyzer.

The maximum rise in hydrogen from the base line or nadir detected by Micro H2 was 74 ppm [range 21-74] for a positive test and 13 ppm for a negative test [range 0-13]. The mean peak rise in hydrogen detected by Micro H2 for a positive test from base line or nadir detected is 34.5±13.99. The maximum rise in hydrogen from the base line or nadir detected by Quintron MicroLyzer was 212 ppm [range 20-212] for a positive test and 17 ppm for a negative test [range 0-17]. The mean peak rise in hydrogen and methane detected by Quintron MicroLyzer for a positive test from base line or nadir detected is 42.1±54.3 ppm and 36.8 ± 40.6 ppm. The results are shown in table 4.3.

Table 4.3: Summary of Hydrogen range and mean rise in participants with a positive and negative test with Quintron MicroLyzer compared to Micro-H2. SD Standard deviation, ppm Parts per million, NA not applicable

Quintron MicroLyzer


Hydrogen Range

Positive test

0-227 ppm

0-78 ppm

Negative test

0-17 ppm

0-13 ppm

Mean peak rise in Hydrogen ± SD

Positive test

71.6 ± 56.9 ppm

33.5±13.7 ppm

Negative test

4.4 ± 4.6 ppm

2.98±2.9 ppm

Methane range

Positive test

5-157 ppm


Negative test

0-4 ppm


Mean peak rise in Methane ± SD

Positive test

42.3 ± 46.3 ppm


Negative Test

0.35 ± 0.9 ppm


For all the participants, the Quintron MicroLyzer gave higher levels of hydrogen than Micro H2 15.4 ± 33.95 ppm Vs 7.5 ± 12.5 ppm respectively, and the mean difference between the results of the Quintron MicroLyzer and the Micro H2 was 7.87 ± 24.2 ppm. The rise in hydrogen in the 2 tests matches each other though the magnitude of rise as expected from the results above is smaller with Micro-H2. These are shown in the figure 4.2. The rises in both positive and negative cases are similar between the two analyzers. The intra-individual correlation between the Micro-H2 and the Quintron is shown in the figure 4.3.

Figure 4.2: Mean rise in hydrogen during positive and negative test detected by Micro-H2 compared with Quintron MicroLyzer after a challenge with 50 grams of lactose

Figure 4.3: Comparison of the Micro-H2 and the Quintron MicroLyzer hydrogen monitors in measuring the maximum increase of breath hydrogen concentration [ppm) in 134 volunteers after lactose challenge. The line is the line of linearity.

The Bland Altman's plot [figure 4.4] shows the two methods [Micro-H2 & Quintron MicroLyzer] are interchangeable in the diagnosis of Lactose malabsorption if the test is based on hydrogen production alone(8) The absolute values of the difference between the Micro-H2 and Quintron MicroLyzer were significantly related to the average level, with a Spearman's rank correlation r=-0.587and significance level p<0.0001.

Figure 4.4: Bland Altman's Plot showing Micro H2 with Quintron MicroLyzer are interchangeable in the diagnosis of Lactose Malabsorption if was based only on the detection of hydrogen in the expired breath sample.


The results from these two systems were very similar for the detection of a significant rise in breath hydrogen after a standard lactose challenge. The handheld Micro H2 analyser would have only missed one case where the hydrogen levels were found to have increased by a significant level when recorded on the Quintron MicroLyzer. The results did however, show a difference in the magnitude of the increase in hydrogen levels after a lactose load in the same subject when the two systems were compared with much higher values obtained from the Quintron MicroLyzer.

The Micro H2 analyser that was used has the ability to measure only exhaled hydrogen rather than both hydrogen and methane which is possible with the Quintron MicroLyzer. As a result the handheld system, although easier to use with an instant result, gives restricted information on the fermentation processes in the large intestine. CH4 is produced in the human intestine chiefly by an H2-utilising flora and so the adequate assessment of gut bacterial carbohydrate fermentation requires parallel measurement of both breath H2 and CH4. The assumption that H2 excretion is the only means of quantifying the amount of malabsorbed carbohydrates is questionable because methane-producing patients are likely to have a higher 'false negative' results after an oral load of lactose(9). It has been shown that as many as 40-50% of people have been identified as methane producers(2, 10). In some subjects, at least part of the hydrogen is used to produce methane, but in most hydrogen-producing subjects, part of the hydrogen is dissolved into the blood stream and exhaled through the lungs(2, 11). A study by Bjorneklett showed the prevalence of CH4 production in a group of 120 healthy subjects, determined by a single midday breath sample, was 44%, with no significant difference between sexes and no correlation to age(2, 12).

H2 and CH4 are only produced by bacteria, and carbohydrates are the primary substrate for their production, the presence of either gas in breath will signal the breakdown of carbohydrates in intestinal tract. Methanogenic bacteria are able to convert H2 to CH4 within the colon(13-15). There is a complex interaction between H2 and CH4 production which was discussed in chapter 3.

In the study by Peuhkuri there were 44 volunteers [34 female and 10 male] with no gastro-intestinal disease. Our study is a larger study with 134 participants, with nearly equal sex distribution [66 males and 68 females] and also involving healthy volunteers and disease groups although they were all in remission. Peuhkuri et al showed that the diagnoses were the same in 100% of the cases with the two breath hydrogen analyzers, the Micro H2 and the Quintron MicroLyzer(5). In contrast our study shows that Micro H2 would diagnose 95% of those with hydrogen production but only diagnose in 53.9% of cases that would be diagnosed using Quintron MicroLyzer this is due to the inability of the Micro H2 to measure breath methane concentrations.

The Micro H2 gave higher peak levels of hydrogen than the Quintron MicroLyzer 51.8 ppm [SD 86.0 ppm] and 32.3 ppm [SD 49.2 ppm] respectively in Peuhkuri's study, and the mean difference between the results of the Micro H2 and the Quintron MicroLyzer was 19.5 ppm [SD 42.6 ppm](5). In my work the Micro H2 gave lower peak levels of hydrogen than the Quintron MicroLyzer 33.5 ppm [SD 13.7 ppm] and 71.6 ppm [SD 56.9 ppm] respectively, and the mean difference between the results of the Quintron MicroLyzer and Micro H2 was 41.2 ppm [SD 46.7 ppm]. The results from the Quintron MicroLyzer were clearly higher than Micro-H2 meter. There are several potential reasons why this may have differed from the Peukhuri paper. Firstly, I used the Quintron MicroLyzer Self-Correcting Model SC which in addition measures CO2 and corrects the results for expired CO2 when compared to Quintron MicroLyzer DP model used in the study by Peukhuri. This reduces a source of error in trace-gas analyses which is contamination of the alveolar sample with dead space air during its collection. This is based on the concept that CO2 is present in alveolar air at a constant concentration, while it is absent in room air. If alveolar air is erroneously mixed with room air, the concentration of CO2 will be reduced so will be the trace gases present. By knowing the degree to which CO2 was diluted, a correction is applied to calculate true alveolar concentration of the gas.

The higher values may also result from the different concentration of gases used to calibrate the analyzers: 100ppm with Quintron MicroLyzer and 50ppm with Micro H2. Also, the sampling techniques differ between the analysers and this may have an effect on the hydrogen concentrations measured. The Micro H2 measures the hydrogen concentration of total breath and shows the highest peak value after the end of the breath. Even though the subjects were allowed to exhale slowly through the mouthpiece of the Micro H2 analyser, methods of blowing are varied. With the Quintron MicroLyzer, the gases are collected in a bag and then hydrogen is measured from the collected breath and the whole breath sample is used in the analysis. However it should be noted that my comparison was conducted in a standardised fashion using a uniform protocol to exclude any known sources of error. A set lactose load was used with a predetermined sampling schedule. All participants were nil by mouth except water for at least 9 hours before the test. They were also not allowed any medications, told not to smoke on the morning of the test and smoking was not allowed during the test. All those with IBD had their disease in remission as determined by standard criteria. Measurement of breath hydrogen and methane, together with lactase genotype, should now form the current best practice for investigation of lactose sensitivity(16).

The Micro H2 is very easy to use, portable and gives instantaneous results. It is cheaper to buy and has lower maintenance costs. The Quintron MicroLyzer needs to be calibrated before it is used each time. Every sample takes at least 2 minutes to analyse and it takes about 30 minutes for analysis of samples taken from one person which leads to higher labour costs. The Micro H2 correctly picked up the rise in hydrogen in 95.5% of cases compared to Quintron MicroLyser. The hand held devices still has a role to play in the diagnosis of lactose malabsorption. In clinical practice you could argue that the breath samples for analysis of lactose malabsorption should be anlaysed first by a hand held device and if the results are suggestive of malabsorption (raised hydrogen) then no further analysis is needed. Those samples that are negative should have assessment using Quintron MicroLyser for diagnosis. This way the two analysers could be used complementary to each other which has benefits like ease of use and reduced laboratory time and cost in diagnosis of lactose malabsorption. A handheld device should now be developed to measure expired methane similar to hydrogen, with the advantages associated with a hand held device; this would be of great help when formally assessing a patient for lactose sensitivity.


The results show that the Micro H2 was not accurate in diagnosing lactose malabsorption because of its inability to measure methane levels. It was, however, reliable in measuring breath hydrogen concentrations alone after an oral dose of lactose. The fact that it cannot measure breath methane concentrations is a serious weakness, because this inability leads to an under diagnosis of lactose sensitivity. I would therefore advocate using the Quintron MicroLyzer with analysis of both hydrogen and methane levels to determine hypolactasia in clinical practice.