WO2008060165A9

WO2008060165A9 - Improvements in or relating to breath collection methods and apparatus

Info

Publication number: WO2008060165A9
Application number: PCT/NZ2007/000333
Authority: WO
Inventors: Randall Alexander Allardyce; Ketan Lad; James Christian Neilson; James Geoffrey Chase
Original assignee: Syft Technologies Limited
Priority date: 2006-11-12
Filing date: 2007-11-12
Publication date: 2009-10-08
Also published as: WO2008060165A1

Abstract

A method of sampling the breath of a patient by directing a sample of the breath into a venturi sample tube, measuring the sample fluid pressure and the mass flow rate and separating the breath sample into a 'dead space' and an 'alveolar' fraction based on a profile of the breath sample.

Description

Improvements in or relating to breath collection methods and apparatus

Background to the Invention SIFT-MS is a technique that is used to monitor volatile components in air in real time. The basis of the technique is that precursor ions (commonly H₃O⁺, O₂ ⁺ and NO⁺) are generated in a vacuum chamber at the upstream end of a flow tube. The ions are then mass selected using a mass filter and injected into the flow tube against a pressure gradient by use of a Venturi nozzle.

The mass selected precursor ions are then entrained in a carrier gas and flow down the flow tube. Helium is usually chosen as the carrier gas because it has a low molecular weight and thus the energy transfer in collisions between ions and the carrier gas in the injection process is minimised. In some cases, helium is mixed with other higher molecular mass inert gases to reduce diffusive losses of the ions. Reducing the energy of the collisions reduces the extent of fragmentation of the precursor ions during injection.

A known flow of sample may be introduced to the flow tube by means of a heated capillary tube and chemical reactions will take place between the analyte species and the precursor ions. The extent of the reaction is monitored by measuring the reduction of intensity of the precursor ion signal, and the magnitude of product ion signals at the downstream end of the flow tube. From the comparison of primary

(precursor) and product ion signals, the identity and concentration of volatile species in the sample may be calculated if the reaction rate and flow dynamics of the system are known.

Currently there is no standardised method of sampling breath with SIFT-MS instruments or other similar real time direct breath capable instruments. Because a SIFT-MS instrument has the ability to analyse whole air samples, a standardised method for breath analysis would be highly advantageous for research into breath testing for clinical diagnosis. Standardising the way in which breath is sampled will allow a consistent method of measuring analytes that could become an industry and research standard. Previous methods of testing breath using the SIFT-MS technique involved blowing past the capillary inlet using straws. However, simply blowing past the capillary with a straw is not an ideal way of providing a sample as the subject may not be able to reach the inlet. In addition, different levels of condensation, moisture and temperature conditions will result in sample variability.

There are two basic methods of providing breath samples to real-time analysis techniques such as SIFT-MS. The first is via a direct sample from the patient. In this method the breath sample is taken and analysed simultaneously. The advantage with this method is clearly the immediacy of the results. However it is difficult to control the precise portion of breath that is "seen" by the instrument. In a practical sense, the analysis- will be done on the entire breath sample regardless of the fact that the majority of useful data will be contained only within the alveolar section. The second method samples the breath indirectly and stores the sample for later analysis. This method lends itself to more precise control over which particular section of the sample is analysed.

The subject using the direct breath collection device must introduce their breath through an inlet to the SIFT-MS instrument capillary tube. The amount of resistance the subject experiences will depend partly on the length of tube prior to the capillary.

Therefore, to minimise resistance and reduce sample dead space and consequently the sample processing time, the capillary is placed as close to the inlet as possible.

This will ensure the shortest possible distance is travelled by the breath to the capillary. Because each unit of the direct breath, collection device will have a capillary, individual calibrations for the capillary tube on each unit are required.

Because polar compounds such as ammonia have a high affinity for water, they are likely to be trapped in condensed moisture inside the device. If water vapour condenses inside the device there would be an observed lower level of polar compounds relative to expected or known values. To ensure that the capillary will not be blocked and that important compounds will not adhere to the sampling apparatus, a direct breath collection device is generally operated at an elevated temperature, preferably greater than the boiling point of water. A device for direct breath collection for a SIFT-MS instrument requires regular sterilisation, and generally would provide provision for a disposable mouth-piece to prevent patient to patient contamination and be able to be disassembled for sterilisation. Consequently any such device would require these characteristics.

The analytes of interest, which provide physiological information about a subject are exhaled in the alveolar portion of breath. Therefore, a method is required to determine when the alveolar portion is exhaled.

PRIOR ART

The first attempted method of detecting alveolar breath involved monitoring exhaled CO2 to indicate exhalation of analytes that ate of interest. Exhalation of alveolar breath occurs during the plateau region of the expiration curve. This approach comprises detecting alveolar samples based on increasing gas exchange to CO2 that occurs only in the alveoli. Available CO2 detectors vary in size, cost and response times. A Nellcor NPB-75 handheld capnograph (CO2 detector) unit, (Tyco healthcare) has a response time of 2.5 seconds, which is too slow to obtain a CO2 reading to decide when to collect alveolar breath samples. All detectors of the same price and size as the NPB-75 have similar response times. Ideally, the capnograph would need to have a sub-second response time to be suitable as a sensor for real time detection of alveolar breath.

Alveolar breath collection (BCA) devices are known. One device, designed by Phillips (1997) "Method for the Collection and Assay of Volatile Organic Compounds in Breath" Analytical Biochemistry, 247, 272-278 is used to collect alveolar breath samples onto sorbent traps for analysis with the GC-MS technique. Using breath samples collected with the BCA, the levels of VOCs in the room air were compared to expired alveolar breath. This device was used for research purposes and is not commercially available.

A majority of the previously designed alveolar breath sampling devices are based on the Haldane-Priestly tube. The Haldane-Priestly tube collects an alveolar breath sample but does not allow the flexibility of collecting breath from any specified region of an exhalation, nor does it allow for the separate collection of the deadspace volume. Thus, it is effectively a feed-forward design that may not work in extreme cases where the exhaled breath volume from a particular subject is significantly larger or smaller than normal. More importantly, the Haldane-Priestly tube does not offer programmable flexibility to collect the best sample

The alveolar breath collection device designed by Raymer et al. - RAYMER, J. H., THOMAS, K. W., COOPER, S. D., WHITAKER, K. A. & PELLIZZARI, E. D. (1990) "A device for sampling of human alveolar breath for the measurement of expired volatile organic compounds" Journal of Analytical Toxicology, 14, 337-344 is used for breath testing research and is not a commercially available product. The device collects samples of alveolar breath into evacuated stainless steel containers. The collected samples were analysed using the GC-MS technique.

The end-expiratory air sampling device by Yeung et al. - YEUNG, C. Y., MA, Y.

P., WONG, F. H., KWAN, H. C₁ FUNG, K. W. & TAM, A. Y. C. (1991) "Automatic end- expiratory air sampling device for breath hydrogen test in infants" The Lancet, 337, 90- 93, is used to collect breath samples for research on breath hydrogen concentrations in infants. Samples are used to monitor breath hydrogen for clinical diagnosis of lactose malabsorption. The device collects and stores breath samples in a syringe and they are analysed using the GC-MS technique. This device was designed for research and is not commercially available.

The device by Dyne et al. - DYNE, D., COCKER, J. & WILSON, H. K. (1997) "A novel device for capturing breath samples for solvent analysis" The Science of the Total Environment,^, 83-89 is designed for collection of alveolar breath samples and storage onto an adsorption tube. The adsorbed sample is then thermally desorbed and analysed with the GC-MS technique. This device is commercially available under the name "Bio-VOC sampler". This breath collection device is another design based on the Haldane-Priestly tube: Breath is exhaled into the device and is exhausted at the other end. When the subject stops exhaling, one-way valves at both ends of the reservoir hold the alveolar sample inside the device. The device consists of a rigid casing that houses a collapsible Tedlar sampling bag. Once the breath is captured inside the sampling bag, the device is attached to a thermal desorption tube where the sample is stored for analysis with GC-MS at a later stage. The breath sample is stored on the sorbent material by collapsing the Tedlar bag inside the casing and forcing the sample through the sorbent tube. The commercially available version is made with a plastic external case and does not use the internal Tedlar bag.

The Alveosampler is a disposable version of the Haldane-Preistly tube and is a commercially available product manufactured by Quintron. It permits single-patient use of an alveolar breath sample device that stores the collected sample in a standard syringe.

Two other devices collect breath samples by different means. The CO2 controlled alveolar gas sampler by Schubert et al. - SCHUBERT, J. K., SPITTLER, K.- H., BRAUN, G., GEIGER, K. & GUTTMANN, J. (2001) "CO2-Controlled sampling of alveolar gas in mechanically ventilated patients" Journal of Applied Physiology, 90, 486-492 is used to collect breath samples from mechanically ventilated patients. Exhaled CO2 levels are monitored to detect the alveolar breath region, which is approximately 6%. Alveolar breath samples are stored on activated carbon sorbent traps. The stored samples are thermally desorbed and analysed using the GC-MS technique. This device is not commercially available and is used for research purposes only.

Yeung et al.- YEUNG, C. Y., MA, Y. P., WONG, F. H., KWAN, H. C, FUNG,

K. W. & TAM, A. Y. C. (1991) "Automatic end-expiratory air sampling device for breath hydrogen test in infants" The Lancet, 337, 90-93. measures the exhaled alveolar CO2 concentration to collect breath samples. Neither of these approaches offer the flexibility in use required as a research device for SIFT-MS breath sample collection.

Apart from the device by Schubert et al. (ibid), known devices use sorbent traps to store the gas sample. Stored samples are then analysed using the GC-MS technique after thermal desorption. All these devices are primarily designed to interface with GC-MS and thermal desorption instruments, and are not necessarily the ideal solution for the SIFT-MS technique. Although SIFT-MS can analyse thermally desorbed samples, quantitative analysis is much more complex than analysis of whole air samples. In addition, the time required for thermal desorption of collected samples voids the real time ability of SIFT-MS and also requires further complex equipment. Thermal desorption is also not suited for collection and analysis of highly water soluble compounds such as ammonia and ethanol.

The alveolar breath collection device by Schubert et al. (ibid) collects whole air samples, but the volume of sample is only 1 ml in 0.1 ml increments per breath. The minimum volume of sample required for an analysis with a SIFT-MS instrument would be hundreds of millilitres. Thus, the alveolar breath collection device by Schubert et al. would be a very time consuming method to collect the sample volume required for analysis using the SIFT-MS technique.

US Patent specification 5,465,728 (Philips) describes and apparatus which has a body with first and second ends. The body defines a chamber and has a breath entry portal, a breath exit portal and a sampling portal. The device also includes a jacket to maintain the desired temperature and a sample container with a pump for moving selected samples from the chamber into the sample chamber.

US Patent specification 6,582,376 (Baghdassarian) discloses a device which has a hollow body having two valved outlets. The concentration of specific gases is monitored to determine the presence of alveolar breath which when present is moved through one of the valved outlets into a collection reservoir.

US Patent specification 5,432,094 (Delente) describes an apparatus for detecting alveolar breath by directing the patient's breath into a container which is sealed. A detector is positioned in the container which^' indicates whether the sample is a true alveolar sample.

US Patent specification 5,072,737 (Goulding) relates to a method of measuring the metabolic rate of a patient and in particular the measurement of the consumption of oxygen in the inspiration gases and the measurement of carbon dioxide in the expired breath. These measurements are made by flow meters which provide flow weighted averages. The apparatus includes a flow sensor to provide to provide information regarding the rate of flow, a pressure sensor, and an oxygen and a carbon dioxide sensor. US Patent specification 4,297,871 (Wright) relates to a breath sampling device which is capable of collecting breath preferably close to the end of the expiration. This is arranged by detecting a change in the pressure of the breath passing through the instrument. Wright teaches that when the rate starts to fall, this will indicate the presence of alveolar breath. The device uses a venturi to generate suction which acts on a displacement element so that when the suction is reduced, the displacement element is released and a sample is collected. This device is not capable of measuring and recording breath profile data, then applying that information to determine if they are 'normal' breaths for the patient. Consequently it is not possible with Wright to provide a reproducible sample collection.

US Patent specification 6,726,637 (Phillips) discloses a method of collecting, concentrating and analysing VOCs in alveolar breath. The specification discloses a device similar to that of US 5,465,728 (Phillips) except that it includes a condensation unit downstream from the sampling portal.

US Patent specification 6,540,691 (Phillips) discloses a process of collecting a representative sample of alveolar breath from a mammal and determining the alveolar gradients of n-alkane in the breath.

US Patent specification 5,848,975 (Phillips) discloses a method of collecting samples of alveolar breath into bags. The method has been particularly designed to detect H. pylori infection.

US Patent specification 5,996,586 (Phillips) discloses a method of detecting lung cancer by collecting a measured quantity of alveolar breath, analysing the breath for the presence of a marker for lung cancer and comparing the analysis with a sample of breath from a mammal free of lung cancer.

US Patent specification 6,221 ,026 (Phillips) discloses yet another process for determining the presence of disease in mammals which comprising the collection of a representative sample of alveolar breath and of ambient air, analysing both the samples to detect the content of n-alkanes, determining the alkane profile and comparing that with a mammal known to be free of the disease.

Other triggers suggested included other breath components, water clusters and measuring the volume of the breath exhaled. Recognition of alveolar breath is achievable by monitoring other commonly occurring analytes found in breath or looking for water clusters (masses 55 and 72 on SIFT-MS). However, detection of other analytes and water clusters would require SIFT-MS to detect in real time, which would not be appropriate for use with a remote breath collection device. Therefore, use of other analytes and water clusters as triggers of alveolar breath is not considered viable.

None of the known devices or methods offer collection of samples in the form of whole air that would simplify analysis in comparison to analysis of thermally desorbed sample when using the SIFT-MS technique. Additionally, they do not provide the desired programmable flexibility desired for the study of sample delivery effects on the concentration of VOCs measured from exhaled, breath.

Therefore, there is a need for a suitable, simple device for the remote collection of breath samples for analysis using the SIFT-MS technique.

OBJECT OF THE INVENTION

It is an object of this invention to provide an improved device for the collection of alveolar breath for immediate or later analysis.

SUMMARY OF THE INVENTION

According to one aspect of this invention there is provided a breath collection device which enables the measurement of breath through the device and the sampling of the breath based on the exhaled profile of the breath, means being provided to apply the information obtained to subsequent breaths from the patient to determine whether the breath samples fall within an acceptable standard established by the information. Preferably the device includes a venturi flow tube through which the breath is passed.

Preferably the collection is measured by an averaging pitot tube associated with the venturi flow tube.

Preferably the averaging pitot tube is connected to a pressure transducer.

Preferably the collection of a breath sample is triggered by the exhaled breath profile.

Preferably the breath collection device is provided with a disposable mouthpiece.

Preferably the breath collection apparatus enables samples to be collected in a

Tedlar bag.

Preferably the Tedlar bag or bags are located within an airtight housing the interior of which is under partial vacuum pressure.

Preferably the partial vacuum pressure is supplied by a diaphragm pump.

Preferably the measurement of the flow of breath is conducted at the inlet and throat of the venturi flow tube.

Preferably the mass flow rate of the breath through the venturi flow tube is determined by using the known cross sectional areas of the venturi flow tube and the. pressure drop across the converging section of the venturi flow tube.

The invention also provides a method of collecting alveolar breath, comprising the steps of: analysing more than one breath sample from a patient to determine a breath profile, utilizing the information obtained from such analysis to determine whether the captured breath sample meets the desired profile rejecting the breath samples that do not come within the desired profile, and capturing the breath samples that come within the desired profile in a sample collection device.

Preferably the sample collection device comprises a SIFT-MS instrument and the sample is either immediately or later analysed in the SIFT-MS machine.

Preferably the sample is stored in the sample collection device for later analysis.

Preferably the breath sample will pass directly into the sample collection device without passing through a pump.

Preferably software architecture is provided to operate sequences required for the collection and measurement of breath passing through the venturi flow tube and to analyse the sample collection based on the exhaled profile.

Preferably the sequences comprise purging of the device, data entry, average breath data acquisition, breath fractionation and breath collection summary.

Preferably the sample collection device is located within an air tight chamber, the interior of which is under a partial vacuum pressure to ensure that the sample air does not pass directly through a pump.

It is to be understood that this invention relates to two methods of sampling breath with SIFT-MS. Firstly the breath can be sampled directly into the SIFT-MS instrument and secondly the collection of the sample into a storage media that can' be analysed at a later stage.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred forms of the invention will now be described with the aid of the accompanying drawings, wherein Figure 1 is a schematic view of one form of a breath collection device according to the present invention.

Figure 2 is a graph which illustrates the VOC concentration changes with respiration.

Figure 3 is a graph illustrating differing VOCs concentrations.

Figure 4 is a graph illustrating a direct breath acetone comparison with alveolar bag acetone.

BEST MODE OF PERFORMING THE INVENTION

A schematic diagram of a device suitable for the remote breath collection is illustrated in Figure 1. In this Figure the patient blows into a mouthpiece 1 of a known construction. Preferably the mouthpiece is disposable to prevent cross contamination.

The mouth piece is connected to a stainless steel venturi sample tube 3 which is open at 3a to enable excess breath to be exhausted. A stainless steel tube 4 is connected to the venturi and includes at least two branches 5 and 6 each of which is controlled by on-off electrical control valves 7 and 8.

A collection chamber 9 which is suitably connected to an exhaust pump 10 is controlled by an electronic module 15 which includes means to receive a control from a signal generator 16 and from a pitot tube pressure transducer 17 which is connected to the sample tube 3 so the system will recognize when breath is being expelled through the sample tube 3. The module also includes control means for the exhaust pump 10 to maintain a vacuum within the chamber 9. The collection chamber 9 includes an appropriate opening such as a lid 18 so access can be obtained to the interior of the chamber. The chamber houses one or more collection devices 19 which are preferably but not necessarily Tedlar bags which are connected at 20 to the control valves 7 and 8.

The purpose of a vacuum within the chamber 9 is to provide the necessary vacuum pressure to enable the breath samples to be drawn into the Tedlar bags thereby providing a method of sample collection that does not involve the sample passing through a pump. The vacuum will also ensure that the lid 18 of the chamber will be held shut by the vacuum pressure.

The Tedlar bags 19 represent only one type of product that would be suitable for the role of storing samples. Any material that is suited to storage of trace amounts of Volatile Organic Compounds (VOCs) will provide a similar capability. Another material that is commonly used for this purpose is Teflon.

Other commercially available air sampling products may also be employed. For example, the chamber 9 can be substituted by a Syft Sample Case by Syft Technologies Ltd, or the Vac-U-Tube or the Vac-U-Chamber by SKC. However, these commercially available devices collect samples by manual operation from the user. For example, the switches on the Syft Sample Case must be turned by the user to initiate the collection of a sample. In the system of the present invention, the breath collection device is automated using program controlled solenoid valves that are opened to collect the appropriate (typically alveolar) breath sample.

The sampling is effected by running test breaths that are used by the system to establish the correct point at which to sample later breaths and to check that all subsequent breaths fall within an acceptable standard. This is effected by separating the breath sample into 'dead space' and 'alveolar' fraction and collecting each fraction in separate containers. In one application of the method of this invention, the patient breathes into the venturi sample tube 3 and after passing the averaging pitot tube and pressure transducer 17, the breath is separated and directed into the respective containers. The separation is generally arranged by the device assigning the first part of the tidal volume as the 'dead space' fraction. The breath samples that come within the desired profile can then be stored and/or analysed.

The collection chamber 9 was designed with the intent to make it as easy and fast as possible to set up for sample collection. Previous sample cases utilised a box inside which the Tedlar bags were placed, with a sealable lid placed on top. In the device of the present invention, the inlets 20 on the Tedlar bag valves are inserted into the attachment points for the control valves 7 and 8, and the case is placed over them. In operation, the vacuum within the collection chamber 9 will create a seal between the lid 18 and the chamber 9. The pump 10 also creates the necessary vacuum pressure inside the case so the samples will be drawn into the Tedlar bags. The pump 10 is preferably situated in its own individual case, directly attached to the collection chamber holding the Tedlar bags. Overall this design allows easy access to the Tedlar bags and if the lid is transparent, the user can clearly see when the bags are full.

The venturi sample flow tube 3 allows the sample fluid pressure at the inlet and at the throat to be measured and the mass flow rate is determined by using the known cross sectional areas and the pressure drop across the converging section of the Venturi flow tube. This measurement is preferably effected by averaging Pitot Tubes or Anubars. The tube 3 may include the usual pressure taps one of which will be arranged to face the fluid flow inside the venturi flow pipe and the second is positioned behind the first, but faces in the opposite direction to the fluid flow. The purpose of the second pressure tap is to measure the stagnant pressure of the fluid flow and hence obtain a differential pressure reading between the two pressure taps to assess flow rate. The tubing is selected to minimise the dead space volume, but must be large enough so that there would not be too much resistance to collect the sample.

Pressure readings from the taps 21 and 22 can be read with a suitable commercially available pressure transducer connected to the pressure taps.

Once outside the sample tube, the breath inlets are connected directly by means of non contaminating tubing to the solenoid sampling valves 20 through the electrical control valves 7 and 8. When the solenoid valve or valves 20 are opened the pressure difference between the chamber 9 and the venturi tube 3 draws the sample in through the breath inlet and into the Tedlar bag. The solenoid valves are connected as close as possible to the breath inlet to minimise the dead space volume from the solenoid to the breath collection point, thus minimising room air residing inside the tubing prior to breath collection. Figure 2 is a graph which illustrates the VOC concentration changes with breath plotted over time.

Figure 3 is a graph which illustrates differing VOC concentrations collected using the device and method of the present invention. The 'bag V and the 'bag 2' show the differing VOC concentrations which are collected into the Tedlar bags.

Figure 4 is a graph illustrating a direct breath acetone measurement compared with alveolar bag acetone and shows a close and linear relationship between two samples and demonstrates the efficient function of the device.

There are a range of mouthpieces currently used in clinical applications for breath testing. However, all of the different variations of mouthpieces have essentially the same diameter. Therefore, the inlet for the patient to use the device is designed to connect with currently used disposable mouthpieces. The majority of available mouthpieces are made from plastic that have incorporated biological filters to prevent the possible spread of infection from patient to patient, although some are disposable cardboard tubes without filters.

To control the various functions, software and electronic hardware are utilised. The invention also provides software architecture to operate sequences required for the collection and measurement of breath passing through the venturi flow tube and to analyse the sample collection based on the exhaled profile.

While the breath collection system of the present invention was developed primarily for use with a SIFT-MS instrument, it is. to be understood that the system may also be utilised with known forms of Mass Spectrometry or any system designed or able to measure analyse breath samples.

Two methods of sampling breath are available as a result of the present invention. The first method is the direct sampling of breath and in particular alveolar breath into a MS instrument and preferably a SIFT-MS instrument. In this case the

SIFT-MS instrument is connected directly in place of one of the sampling collection bags and there is no requirement for a diaphragm pump 10 as the SIFT-MS instrument provides a constant sampling flow rate. The second method is the collection of a breath sample remotely into a storage medium for later analysis.

The present invention provides the following characteristics: 1. Minimum flow resistance: for clinical reasons, the patient providing the breath sample should not experience undue resistance while exhaling into the device.

2. Contamination: subjects are not exposed to risk of infection from any previous subjects who used the device.

3. Condensation: To avoid condensation during collection, the parts in direct contact with the subjects' breath are heated above 37°C and are made from a non contaminating substance, such as stainless steel or FEP.

4. The collection chamber is large enough to collect samples into at least two 1 litre Tedlar bags.

5. The device can easily be dismantled for sterilisation as parts of the device in direct contact with the breath flow may need to be autoclaved regularly.

6. The collected breath sample does not pass through any pumps, which otherwise may create another external VOC source and loss of analyte on the pump internals.

7. The device will automatically collect the samples at the relevant collection points without the operator actively collecting them.

Having described preferred methods of putting the invention into effect, it will be apparent to those skilled in the art to which this invention relates, that modifications and amendments to various features and items can be effected and yet still come within the general concept of the invention. It is to be understood that all such modifications and amendments are intended to be included within the scope of the present invention.

Claims

WHAT WE CLAIM IS:

1. A breath collection device which enables the measurement of breath through a device and the sampling of the breath based on the exhaled profile of the breath, means being provided to apply the information obtained to subsequent breaths from the patient to determine whether the breath samples fall within an acceptable standard established by the information..

2. The breath collection device of claim 1 , wherein the device includes a venturi flow tube through which the breath is passed.

3. The breath collection device of claim 2, wherein the collection is measured by an averaging pitot tube associated with the venturi flow tube.

4. The breath collection device of claim 3, wherein the averaging pitot tube is connected to a pressure transducer.

5. The breath collection device of claim 1, wherein the collection of a breath sample is triggered by the exhaled breath profile.

6. The breath collection device of claim 1 , wherein the breath collection device is provided with a disposable mouthpiece.

7. The breath collection device of claim 1 , wherein the breath collection apparatus enables samples to be collected in a Tedlar bag.

8. The breath collection device of claim 7, wherein the Tedlar bag or bags are located within an airtight housing the interior of which is under partial vacuum pressure.

9. The breath collection device of claim 8, wherein the partial vacuum pressure is supplied by a diaphragm pump.

10. The breath collection device. of claim 2, wherein the measurement of the flow of breath is conducted at the inlet and throat of the venturi flow tube.

11. The breath collection device of claim 2, wherein the mass flow rate of the breath through the venturi flow tube is determined by using the known cross sectional aress of the venturi flow tube and the pressure drop across the converging section of the venturi flow tube.

12. A method of collecting a breath sample, comprising the steps of: analysing more than one breath samples from a patient to determine a breath profile, utilizing the information obtained from such analysis to determine whether the captured breath sample meets the desired profile rejecting the breath samples that do not come within the desired profile, capturing the breath samples that come within the desired profile in a sample collection device.

13. The method of collecting a breath sample as claimed in claim 12, wherein the sample collection device comprises a SIFT-MS instrument and the sample is either . immediately or later analysed in the SIFT-MS machine.

14. The method of collecting a breath sample as claimed in claim 12, wherein the sample is stored in a sample collection device for later analysis.

15. The method of collecting a breath sample as claimed in claim 12, wherein the breath sample passes directly into the sample collection device without passing through a pump.

16. The method of collecting a breath sample as claimed in claim 12 including software architecture to operate sequences required for the collection and measurement of breath passing through the venturi flow tube and to analyse the sample collection based on the exhaled profile.

17 A breath collection device substantially as herein described with reference to Figure 1 of the accompanying drawings.