Скачать 111.77 Kb.
Breath ammonia levels in a normal human population study as determined by photoacoustic laser spectroscopy
T Hibbard1 and A J Killard1,2*
1Biomedical Diagnostics Institute, National Centre for Sensor Research, Dublin City University, Dublin 9, Ireland
2Department of Applied Sciences, University of the West of England, Coldharbour Lane, Bristol BS16 1QY, UK.
Photoacoustic laser spectroscopy was used as a technique to measure real-time levels of ammonia in exhaled human breath in a small, locally recruited, normal healthy population (n=30). This yielded an average level of breath ammonia of 265 ppb, ranging from 29 to 688 ppb. Although average levels were marginally higher in male volunteers, this was not statistically significant. In addition, no correlation could be found between age, body mass index, or breath carbon dioxide levels. Monitoring of the daily routine of two individuals showed a consistent decrease in oral breath ammonia concentrations by the early afternoon (post-prandial), but was followed by a gradual increase towards late afternoon. However, in a comparison of oral and nasal breath in two volunteers, nasal breath ammonia levels were found to be significantly lower than oral levels. In addition, the daily variation was only seen in oral rather than nasal measurements which may indicate that significant background levels are predominantly of oral origin and that nasal sampling is the preferred route to eradicate this background in future studies. These results provide a healthy human breath ammonia baseline upon which other studies may be compared.
Human breath, diagnostics, volatile compound, ammonia, photoacoustic spectroscopy
There are at least 1,000 trace volatile compounds (VCs) in human breath . Furthermore, VCs found in human breath have been linked to various physiological conditions. For example, nitric oxide is indicative of pulmonary dysfunction, acetone has been linked to diabetes, and pentane is suggestive of lipid peroxidation . Hence, understanding the correlation between breath compounds and the condition of the human body has attracted increasing interest in clinical diagnostics.
Among breath compounds, ammonia has proven to be an important trace gas. It is a key component of the nitrogen cycle which results from protein metabolism and arises naturally from metabolic processes in the stomach and intestines . When the liver and/or kidneys are not functioning correctly, ammonia found in blood cannot be filtered out of the body properly. The increased levels due to liver dysfunction make it possible for hyperammonaemia to develop causing neurological disorders such as hepatic encephalopathy . Where the kidneys are not operating efficiently, conditions such as uraemia , acidosis, and edema  may also develop. Hence, focus on measurement of ammonia levels in the blood is of great value to those with potential liver or kidney disorders. Any excess ammonia that has not been filtered out of the body is capable of diffusing into the lungs and can be exhaled in breath . Breath ammonia levels have been correlated with blood urea (R2=0.95) and creatinine levels (R2=0.83) using optoacoustic spectroscopy . However, development of technologies that are capable of measuring ammonia in breath has proven to be a challenging task.
Breath is a complex matrix where flow rate, temperature, humidity, and gas concentration are multifarious, time-dependent variables. The goal is to isolate the particular VC signal from all other molecular interferences that can arise in the stream of breath. In an attempt to gain a deeper understanding of how conditions in the human body correlate with the constituents of breath, techniques such as chemical ionisation, gas chromatography, laser spectroscopy, and chemical sensing have been used. However, there is still a lack of data in this area due to the limited number of instruments available which are further deterred by their individual limitations. Instruments based on chemical ionisation can detect breath ammonia as low as 10 ppb, but due to complexity are still in the development of a more portable version . Gas chromatography is capable of quantifying breath ammonia at levels of 14 ppt, but the need to pre-collect the samples removes the possibility for real-time analysis . Laser spectroscopy can reach 10 ppb, although the differentiation of molecules that fall into categories of similar frequency ranges and spectral regions can prove difficult . Chemical sensing methods are capable of detecting 18 ppb, but condensation from breath can decrease the accuracy of readings . Due to such drawbacks, a need for additional research is necessary for more conclusive comparisons to be made. Recently, a form of photoacoustic laser spectroscopy has been developed specifically for the real-time monitoring of trace gases in breath. Using this technique eliminates the need for pre-collection which removes the potential of contamination or loss of sample. Furthermore, by using high laser strength, increasing the intensity of photoacoustic signals received from chosen gas molecules, and reducing noise signals from the adjoining interfering gases, the sensitivity of a photoacoustic instrument is usually orders of magnitude higher than conventional measurement technologies . However, it must be acknowledged that specificity errors can arise where impeding molecules of similar gas species are absorbed from the background. When attempting to detect gaseous ammonia, photoacoustic systems can typically use a wavelength of 1531.68 nm .
The work here outlines the application of photoacoustic spectroscopy to the analysis of a small cohort of healthy human volunteers to determine baseline ranges of ammonia concentration. This was also correlated against other biometric data including gender, age, body mass index (BMI), and breath carbon dioxide levels since previous studies have suggested that a relationship may exist among healthy people. The effect of sampling route (nasal vs. oral) and daily routine (post-prandial) was also investigated. Oral was examined more extensively in this preliminary work since it has shown a strong correlation with blood ammonia levels in previous literature. However, a quick nasal study was performed to see if it may bear more potential for future works.
2. Experimental Procedures
A breath sample was defined as a relaxed full exhalation from the volunteer after a relaxed full inhalation. Over a period of ten minutes, breath measurements were made using the NephroluxTM photoacoustic laser ammonia analyser (Pranalytica Inc., CA, USA) (Fig. 2.1). The NephroluxTM
Figure 2.1. Breath ammonia measurement using photoacoustic spectroscopy. The typical procedure for volunteer breath collection using the NephroluxTM.
underwent a multipoint ammonia calibration during development at Pranalytica which generated a strong R2 of 0.9986. Photoacoustic spectroscopy (PAS) is a well-known technique that has the advantage of being relatively simple while still capable of high sensitivity detection. Previous literature has explained the process of PAS in depth . To summarise, every 41 seconds, the system pulled a breath sample into the photoacoustic chamber by way of a vacuum flow. As ammonia (NH3) and carbon dioxide (CO2) breath gases entered the photoacoustic chamber of the NephroluxTM, the tunable CO2-laser  excited the gas molecules. The light energy used for absorption was then converted to heat . Pressure from this temperature change was detected by the chamber microphone during the de-excitation process . The photoacoustic signal, S, provided a magnitude in arbitrary units displayed as:
S = SmPC (1) where Sm is the sensitivity of the microphone (in units volts per pascal), P is the power of incident laser radiation (in units of watts), C is a constant for the cell-geometry, measurement conditions, and modulation frequency (in units of pascal centimetres per watt), and is the absorption cross-section of the transition being interrogated (in units of inverse centimetres) . The CO2-laser used by the NephroluxTM is tuneable for gas selectivity allowing for ammonia detection as low as 1 ppb ±10% and carbon dioxide down to 0.1% ±5% (of reading) to generate results within 120 seconds. Once the breath was analysed by the signal processor, the data was transmitted through the single-board computer system of the NephroluxTM and saved to an external lap-top (Fig. 2.2).
Figure 2.2. Ammonia gas molecule measurement using photoacoustic spectroscopy. Schematic of the NephroluxTM system. The CO2 laser enters the photoacoustic chamber and excites the gas molecules. As de-excitation occurs, the pressure is released and detected by the microphone where it is transmitted to the signal processor . Figure reproduced from reference .
A cohort of 30 normal healthy human volunteers was locally recruited following institutional ethical approval. Testing was carried out over a period of approximately six months. Following a questionnaire, only those with no breathing abnormalities or related medical conditions were entered into the study population. Gender, ethnicity, age and BMI were recorded. For breath sampling, volunteers were asked to sit in a relaxed position while breathing into a face mask which was connected to a breath delivery tube. All volunteers were asked to breathe using either oral or nasal respiration. Furthermore, each breath sample was from a relaxed full exhalation where breath concentrations were averaged. For the purpose of clinical analysis, it is useful to define whether whole breath averaged concentrations or peak alveolar concentrations are being observed. When focusing specifically on the alveolar concentrations, the compounds exclusively from within the lungs are monitored. With this study, however, whole breath examination took place which provided samples consisting of both the alveolar region and dead space (e.g. trachea) within the respiratory system . Previous literature has described ammonia as being a sticky substance which could potentially adhere to the walls of the system . To prevent this adsorption, the tubing of the NephroluxTM was constantly heated while breath consistently flowed into the system. Each sample was an average of five breath measurements.
A single male and female volunteer were studied for five consecutive days to observe daily variation in their oral ammonia breath concentrations. The volunteers fasted from midnight to 10:00 am at which point they were sampled. They were again sampled at 13:00 pm (post-prandial), and at 16:00 pm following a normal daily routine. In addition, a single volunteer provided three repetitions of nasal and oral breath on two separate occasions.
3. Results and Discussion
Oral Breath Ammonia and Statistical Confidence
The survey population had a mean age of 34. There were 19 female and 11 male volunteers, and the average body mass index was 25.18 kg/m2. Of these volunteers, 20 identified themselves as ethnically Irish, with several Europeans, two Americans and one Asian. The biometric data, mean oral breath ammonia, and mean oral breath carbon dioxide concentrations from the 30 volunteers are compiled in Table 3.1. The mean breath ammonia range was calculated by averaging five breath samples per individual. The mean oral breath ammonia of the 30 volunteers was 264.9 ppb with a standard deviation of ±182.8 ppb while the inter-individual (i.e. averaged five breath samples) deviation was ±43 ppb. Mean oral breath ammonia concentrations recorded from this test population ranged from 28.8 ppb to 687.8 ppb.
Previous literature has shown that oral breath ammonia ranges for healthy populations of different sample sizes can vary. Research using selected ion flow mass spectrometry (SIFT-MS) observed oral breath ammonia concentrations of 220 ppb to 550 ppb from a population of five healthy volunteers , and 200 ppb to 1800 ppb from a separate cohort of six volunteers  whereas, a larger group of 30 volunteers produced an oral breath ammonia range of 248 ppb to 2935 ppb , and another of 200 volunteers gave 112 ppb to 2865 ppb . Understanding the significance of the sample size is helpful in relating the resulting breath ammonia ranges to the
Table 3.1. Compilation of biometrics, mean oral breath ammonia, and mean oral breath carbon dioxide concentrations for 30 volunteers (mean = average of five breath samples per person).
rest of the human population. In other words, larger sample sizes are more likely to generate results that are closely related to the target population as well as produce a smaller margin of error . In population statistics in which a sample is used, a measure of confidence is needed in order to know if the sample results can be related to the larger population. Assuming that the sample studied here was not statistically abnormal, it is safe to claim that the results of a confidence interval calculation for the sample will be true 95% of the time . In calculation of the confidence interval for the given sample size and oral breath ammonia concentrations, it can be claimed that there is 95% confidence that the interval of 196.7 ppb to 333.2 ppb contains the true mean of the local population. There was a standard error of 33.3 ppb as well as a skewness of0.48 and a kurtosis of -0.70. It should be recognised that the population sample (n=30) examined in this study was of a small group where metabolic ammonia concentrations could be related to local factors such as ethnicity, diet and localised environmental factors, and may result in a mean and range distinct from those of other research.
Oral Breath Ammonia and Gender
The 19 female volunteers ranged from a mean minimum of 45 ppb to a mean maximum of 688 ppb, with an average of 250 ±189 ppb. The 11 male volunteers had a mean minimum of 29 ppb and a mean maximum of 598 ppb, with an average of 290 ±177ppb showing that the female standard deviation was slightly higher than that of the males. To compare the breath ammonia levels of females against males, a box plot analysis was used (Fig. 3.1). From the box plot, the oral breath ammonia of females and males combined gave a median of 254 ppb where the upper quartile (Q3) was 415 ppb and the lower quartile (Q1) was 91 ppb. The upper quartile represents the median of the top half of the data set. Hence, the upper 25% of the total ammonia concentrations exist between Q3 and the Max Limit. The lower quartile is indicated by the median of the bottom half of the box data. This area between Q1 and the Min Limit contains the lower 25% of the total concentrations. The breath ammonia of the females was skewed towards a lower median of 194 ppb (Q1 = 94 ppb, Q3 = 409 ppb) while the males leaned more towards a higher median concentration of 308 ppb (Q1 = 160 ppb, Q3 = 402 ppb). Further comparison of the ammonia levels between females and males via a t-test indicated a p-value of 0.57. For the t-test, a null hypothesis was used stating that the two populations will produce similar breath ammonia, and any differences were due to chance. With no predictions or expectations of a
|Human pth (1-34) Specific elisa kit Enzyme-Linked ImmunoSorbent Assay (elisa) for the Quantitative Determination of Human Parathyroid Hormone 1-34 Levels in Plasma||Study programme: Optics, Lasers and Optical Spectroscopy|
|Does ratification of human-rights treaties have effects on population health?||Belief seems to be a universal aspect of human behavior-typical of children, fanatics, psychotics, and primitive tribesmen as well as most of today's normal|
|Pattern of drinking and fatal injury: a population-based follow-up study of Finnish men||Population dynamics of the northern short-tailed shrew, Blarina brevicauda: insights from a 25-year study|
|Study of serum ferritin, HbA1c, nitric oxide, uric acid levels in type 2 diabetes mellitus”||II. Catalogue Description The course of human life, including the factors which impinge on the developmental continuum between normal and pathological conditions. III. Course Description|
|Laser-Ablation Rates Measured Using x-ray Laser Transmission||Руководитель программы: проф. В. А. Иванов Кафедра оптики Научный доц. Ю. Э. Скобло Рецензент: доц. А. А. Пастор Study of the population processes of neon atom|