Standardisation issue | Recommendation | Strength of recommendation, rationale and comments |
General | Standardise sampling, storage, and assay within an individual study. | 3. Comprehensive uniform standardisation of EBC is not recommended. However, within a single study, it is recommended to identify potential confounding issues and control for them. If using different devices within one study, rigorous comparisons should be made to assure that differences in temperature, collector surface, cleaning agents, salivary trapping ability, duration of collection and other characteristics do not cause differences in marker levels. |
Sampling issues | Methodology needs to be well-detailed. | 5. Basic requirement of scientific manuscripts. Sufficient detail should be provided to assure that the technique can be reproduced. |
Device | Manuscripts should clearly delineate the device used. If a commercial device, note the name and manufacturer and precisely specify any modifications. If a custom device, clearly detail the device, provide sufficient diagrams to allow a reasonable understanding of the equipment employed. | 4. At the current level of knowledge, this detail is necessary to help determine what factors may indeed be relevant for the various markers. Most information for the commercial devices is readily available and does not need to be reiterated in manuscripts. |
Materials | If a custom device, detail the materials that are employed, particularly for the surface in contact with the EBC. | 4. This information will enhance knowledge of optimum materials for different markers. Different markers will have differing tolerance for materials that line the condensers. The materials themselves, or the compounds used to clean the equipment, may contaminate the sample. This needs to be investigated for each marker. |
Temperature | Specify temperature or range | 4. It remains unclear for many markers what the optimum temperature of collection is. There is an expectation that colder sampling is better for unstable mediators, however this has not been proven, and is not necessarily correct. The temperature of collection should be noted. For the commercial devices, the following recommendations are forwarded. ECoScreen: The temperature of the collector should be measured at the beginning of collection. RTube: The temperature of the cooling sleeve or the freezer in which it is kept should be measured or recorded. |
Saliva trap | Recommended | 5. Gross salivary contamination certainly occurs occasionally in EBC sampling. Some subjects/patients profusely drool. A clear system for preventing these occasional contaminations should be in place, or samples should be assessed for salivary contamination. |
Duration of collection | Duration of collection should be recorded. | 2. Sample volume, minute volume, and duration of collection are three potential, interrelated values that might be chosen for standardisation within a single study. Sample volume is difficult to determine during sample collection in most systems. Minute volume can be determined readily. Duration is simplest. Concentrations of many markers in EBC may be completely independent of any of these values. It needs to be determined for each study/marker whether these values need to be controlled. |
Noseclips | Probably should be worn during oral EBC sampling. | 2. There are two reasons for wearing noseclips for the collection of EBC: first is to minimise nasal airway lining fluid entry into the airstream during inhalation; second is to keep all exhaled air exiting through the mouth and not the nose. Note that the nasal contamination issues for EBC collection are entirely different than for exhaled NO, the latter of which is performed during exhalation manoeuvres only). Reasons not to wear noseclips include discomfort. There are no data to support this recommendation. |
Contamination | Test all materials that contact EBC and assure adequate controls are in place. | 4. NOx are notorious laboratory surface contaminants, and rigorous methods for assuring no contamination of collection equipment, pipettes or sample containers is necessary. Some sample containers may leach out NOx, which suggests that assays should be performed as soon as possible after collection. The pH of EBC may be affected by contaminants, and the system of collection and storage should be assessed for this. |
Subject issues | ||
Storage | Unless proven unnecessary, store samples in the coldest temperature available. | 3. This seems a reasonable precaution, as it is likely that freezing will not significantly alter levels on markers in EBC. |
Stability of marker in storage | Data should be presented regarding marker stability in EBC, or previous publications addressing the stability of the specific marker referenced. | 4. Loss of marker over time in storage, or even with just a brief delay before analysis may lead to Type II errors and may be one cause of wide ranges of values seen in a given marker among different publications. |
Stabilisation of marker | When possible, this should be performed. | 2. Addition of a marker-free protein to EBC sample may increase or lessen the loss of an unstable marker. Early derivatisation of a reactive compound to a stable compound may allow later assay. Each marker needs to be independently analysed to determine optimum systems. |
Assay | In all cases, use assays proven to be sufficiently sensitive and specific for the marker of interest in EBC. | 5. Issues of assay sensitivity and specificity are of paramount importance when working with EBC. The Task Force considers it likely that the majority of variability seen within and between studies is based on difficulties with the assays. |
Timing | Assays should be performed as soon as possible if any loss of marker or contamination with exogenous marker is likely to occur. | 4. This is a reasonable precaution, as many markers are not stable. This precaution needs to be balanced with the need to maintain feasible study designs. In all cases, the effect of storage time (delay to assay) on the marker of interest in EBC should be determined. |
Validation in EBC | Assay systems should be tested for utility in EBC. | 5. Most assay systems employed were not designed for use in EBC. EBC is a very dilute fluid that is protein and buffer poor. |
Immunoassays | Assure that nonspecific binding is identified and minimised, and that appropriate controls are performed in all cases. | 4. Immunoassays in which the protein-poor EBC may lead to falsely elevated test results have been identified. This may result from unblocking of the immunoassay by the dilute, and potentially acidic EBC. Spike-recovery experiments using EBC as the matrix can be helpful to assuring assay functionality. Use of standard curves in an EBC or water matrix, as opposed to a proteinaceous matrix, should be considered. Also to be considered is the addition to EBC of protein so that the matrix of the standard solutions (generally containing substantial protein) and the EBC will be more similar and lead to less false positive results. Concentration of samples by lyophilisation and resuspension in ≤1/10th of the initial volume can bring many markers into the reliable range of immunoassays. |
Nitrogen oxides | Report precisely what was measured. Do not use the term nitrogen oxides (NOx) without providing a definition for that term in the manuscript. Clearly note the NOx that are included in the assay used. | 4. Many NOx assays reported to date are not specific for one compound. Assays may be for nitrate, nitrite, nitrosothiols, nitrotyrosine, NO and other higher oxides of nitrogen. Chemiluminesce analysis after reduction by various methods, when carefully performed, provides sufficient sensitivity for most NOx in unconcentrated EBC. Spectrophotometric tests are less sensitive and NOx are commonly near the limits of detection, especially for nitrite and nitrosothiols, thus, these assays should be used cautiously. Be aware of laboratory surface and supply contamination. Pay attention to potential contamination of the assay equipment (including injection supplies), storage containers, pipettes, as well as the sample collection equipment. |
pH | Report if de-aerated (or gas-standardised) and by what means. If not de-aerated, note the timing of the measurement after collection. | 5. Both de-aerated and nondeaerated assays have been reported. Measurement of pH after deaeration is probably the most technically validated EBC measurement in the published literature. De-aeration is a misnomer, a better term may be “gas-standardised.” Completeness of removal of CO2 by gas standardisation is not clear. However the pH does stabilise during this process (described in text). The gas-standardised EBC pH is not a direct measurement of airway lining fluid, but appears to reflect capture of acids volatilised from the airway lining fluid. CO2 is not appreciably more volatile from a low pH fluid, and, therefore, is not of interest in EBC when attempting to identify airway acidification. If not de-aerating, the pH will be substantially affected by CO2 moving in and out of the EBC. Many investigators consider the CO2 to be unwelcome noise in the system. Others consider it relevant and believe that gas standardisation is unnecessary. |
Spectrophotometric and other assays | Assure sufficient controls, and that the assay is in range. | 4. Many spectrophotometric assays identify their target at or near the lower limit of detection. At the lower limit, there is generally substantially increased variability. Indeed, variability in EBC sampling and assays may be predominated by assay issues (as opposed to collection issues) in may people. |
Dilution issues | Consider volatile and nonvolatile constituents of EBC differently. Present findings of nonvolatiles cautiously in the absence of a dilution factor or relevant ratio. | 4. Nonvolatile substances are derived either from aerosolised particles of airway lining fluid or from a chemical alteration of a volatile molecule into a nonvolatile molecule. When considering the airway lining fluid, a dilution factor is necessary in order to confidently determine absolute concentrations in the airway lining fluid. Ratios of marker to total protein, urea or conductivity have been proposed and hold promise, but have not yet been used extensively. |
3. Volatile components of EBC do not benefit from having a dilution marker. Issues that determine volatile components include the composition of the source fluid, the water-partition coefficient, and others. | ||
1. Beware of the possibility if a volatile marker (such as NO) reacting in solution to become a non-volatile (nitrate). This is speculative. | ||
New markers | Sceptically consider the specificity and sensitivity of the assay. Determine the possibility of contamination. Determine stability in storage. | 3. In all cases, careful scientific investigation should be undertaken for new (and older) markers identified in EBC. |
EBC: exhaled breath condensate; NO: nitric oxide; NOx: nitrogen oxides. Issues relevant to the table: 1) Optimum collection and assay systems differ substantially for the markers measured in EBC. The Task Force believes that no one standardisation is wise to force upon the field. It is not scientifically appropriate to standardise the collection technique broadly for all purposes, as this would have the effect of greatly limiting innovation, and optimisation for one marker will make another marker's collection or assay suboptimal. There will be disagreement between labs as to normal levels of a biomarker until all aspects of an EBC sampling and assay procedure are standardised, but such standardisation should be marker specific. 2) The Task Force has endeavoured to initiate the standardisation process by making recommendations. These must not be considered rigid or absolute requirements by reviewers of papers or grants, as often the level of evidence supporting the recommendations is not high. 3) Strength of recommendation has been indicated (5: unequivocal data and/or unanimity among Task Force experts; 4: compelling data or when data are unnecessary, strong consensus; 3: little data, or data unnecessary, with consensus; 2: little or no data, majority opinion. 1: no published data, opinions of panel).