© 2007 American Thoracic Society doi: 10.1164/rccm.200605-642ST
An Official American Thoracic Society/European Respiratory Society Statement: Pulmonary Function Testing in Preschool ChildrenTHIS OFFICIAL STATEMENT OF THE AMERICAN THORACIC SOCIETY (ATS) AND THE EUROPEAN RESPIRATORY SOCIETY (ERS) WAS APPROVED BY THE ATS BOARD OF DIRECTORS, SEPTEMBER 2006, AND THE ERS EXECUTIVE COMMITTEE, DECEMBER 2006
Section 1. The Next Frontier Peter D. Sly (Chair), Nicole Beydon, Stephanie D. Davis, Claude Gaultier, Enrico Lombardi, Mohy G. Morris, Janet Stocks Section 2. Clinical Implications Sheila A. McKenzie (Chair), Paul Aurora, Francine M. Ducharme, Michael J. R. Healy, Bent Klug, Paul C. Seddon, Janet Stocks
Section 3. Spirometry Paul Aurora (Co-chair), Howard Eigen (Co-chair), Hubertus G. M. Arets, Stephanie D. Davis, Marcus H. Jones, Janet Stocks, Robert S. Tepper, Daphna Vilozni
Section 4. Tidal Breathing Measurements Paul C. Seddon (Chair), Julian L. Allen, Karin C. Lødrup Carlsen, Oscar H. Mayer
Section 5. The Interrupter Technique Enrico Lombardi (Chair), Hubertus G. M. Arets, Nicole Beydon, Hans Bisgaard, Claude Gaultier, Bent Klug, Sheila A. McKenzie, Peter J. F. M. Merkus, Paul C. Seddon, Peter D. Sly
Section 6. The Forced Oscillation Technique Francçois Marchal (Chair), G. Michael Davis, Francine M. Ducharme, Graham L. Hall, Zoltán Hantos, Ellie Oostveen
Section 7. The Multiple-Breath Inert Gas Washout Technique Per M. Gustafsson (Chair), Janet Stocks, Paul Aurora, J. Jane Pillow, Monika Gappa
Section 8. Bronchial Responsiveness Tests Nicole Beydon (Chair), Hans Bisgaard, Claude Gaultier, Enrico Lombardi, Michael Silverman, Peter D. Sly, Janet Stocks, Nicola M. Wilson
In older children, measuring lung function is integral for understanding respiratory physiology and for clinical assessment. Pulmonary function tests for infants and children younger than 2 years are used as both research and clinical tools. The usefulness of these tests has benefited from approximately 15 years of work by joint American Thoracic Society (ATS)/European Respiratory Society (ERS) working parties and task forces (1, 2). However, children aged 2 to 6 years old represent one of the major challenges in lung function assessment. Evaluating lung function in this age group is important, not only for clinical reasons but also due to the considerable growth and development of the respiratory system that occurs, with associated changes in lung mechanics (3). Children commonly present with recurrent cough and wheeze during this period. Many of these children will lose their symptoms as they grow, yet others will continue to have asthma that persists into adult life (4). The treatment implications of these two clinical patterns are different, yet we are currently hampered by a lack of objective assessments to help distinguish between these two patterns. In addition, children recovering from chronic neonatal lung disease and children with cystic fibrosis (CF) are prone to recurrent or persistent respiratory symptoms. Objective assessments of pulmonary function in these children would be expected to improve clinical management. The importance of continuous, longitudinal assessments of lung function from birth throughout childhood cannot be underestimated in understanding the evolution and natural history of disease processes. Preschoolers present a number of special challenges. The children are generally too old to sedate for pulmonary function testing (PFT), as is done with infants, and measurement of lung function under anesthesia is neither ethically acceptable nor physiologically relevant to clinical management. Children in this age group are not able to voluntarily perform many of the physiological maneuvers required for the pulmonary function tests used in older children and adults. They have a short attention span and are easily distracted. Due to these issues, the children need to be engaged and encouraged by the operator to participate in the test.
A number of pulmonary function tests have been attempted in conscious children within the preschool age group. These include the following: standard spirometry (5–11), maximal flow referenced to functional residual capacity ( As has been stressed by the ATS/ERS Working Party on Infant Pulmonary Function Testing, no matter which test is being used the operator must be given access to raw data from the equipment. As the field develops and the knowledge of respiratory physiology in this age group expands, having access to raw data will allow investigation of different and more appropriate algorithms and may result in improved disease discrimination. The joint ATS/ERS task force has produced recommendations for the tests currently used in the preschool age group. Each section of this document was written by a subcommittee of the present task force, and includes the current knowledge and recommendations to guide technical and clinical practice. These recommendations were based on reliable scientific evidence, documented by references, and validated by the subcommittee experts. However, in many situations, insufficient data exist to make definitive recommendations. This document highlights the current state of knowledge and where further data are needed. Recommendations will need to be revised periodically until sufficient evidence has been collected to make definitive guidelines in certain situations. This document will address the following topics: (1) clinical implications of PFT in preschool children, (2) spirometry, (3) tidal breathing measurements, (4) the interrupter technique, (5) the FOT, (6) gas washout techniques, and (7) bronchial responsiveness tests. Specifications for equipment used in an infant/preschooler pulmonary function laboratory have been previously reported (32), and a review of these systems and their hygiene aspects is beyond the scope of these recommendations. However, it is important to highlight that the total apparatus dead space should be minimized where possible, although this requirement does not preclude the use of bacterial filters, and should in general be lower than 1.5 to 2 ml/kg body weight (32). The main aim of these recommendations is to provide a resource for the user of these preschool techniques, to facilitate good laboratory practice, interpretation of measurements, and comparison among centers. These recommendations are expected to help the development of future methodological research in either single- or multicenter clinical studies, which are needed to support strong recommendations. Manufacturers may refer to the technical aspects of this document for developing proper equipment and software. The ideal pulmonary function test in preschool children is one that is applicable to any age so that longitudinal studies can be conducted monitoring individual children from infancy to adulthood, simple to perform, safe, reproducible, sensitive enough to detect changes with growth and distinguish clearly between health and disease, and acceptable to both the subject and parents. As with pulmonary function tests in infants, special attention must be paid to the measurement conditions under which the tests are performed, and the impact of these measurement conditions on the accuracy of test results must be considered. A pulmonary function laboratory that is "preschool-aged child friendly" is of the utmost importance. These young children must be made to feel comfortable in the laboratory environment if they are to perform the measurements accurately. The pulmonary function technician has a significant impact on the comfort level of the child. This type of environment may be achieved through a combination of friendly conversation, songs, or through distraction with a videotape or book. During tidal breathing, the level of distraction must be enough to take the child's attention away from his or her breathing, but not so exciting that the child breathes irregularly. Accurate measurements of height and weight using calibrated stadiometers and scales are essential; however, these procedures can be challenging in an active preschooler. Safety and hygiene requirements have been covered in adult guidelines, but it should be noted that additional safety precautions are necessary for preschool subjects. These include, but are not limited to, the need for constant adult supervision while the child is in the laboratory. For accurate interpretations of the lung function data, particularly where longitudinal assessments are to be made, it is essential to record data on environmental and hereditary factors likely to impact on lung growth, including the following: sex; ethnic group; family history of asthma and atopy; cigarette smoke exposure, both pre- and postnatal; allergen exposure, including pets; and relevant current and past medical history and medication use. The developmental stage of the preschool-aged child will be an important determinant of the child's success at performing pulmonary function tests. This influence will be greatest in tests requiring more active cooperation from the child. For example, young children frequently have difficulties in performing the forced expiratory maneuvers required for spirometry. They can either blow "hard" or "long," but frequently cannot blow both hard and long. Measurements that can be made during tidal breathing, such as forced oscillation, the interrupter technique, and gas washout techniques, may be more suitable for the child unable to accurately perform spirometry. If forced expiratory measurements are to be performed, these should be performed after the tidal measurements, because it is easier to "wind up" young children than to wind them down. In addition, deep inhalation may change bronchial tone in children with asthma. The physiological developmental stage of the respiratory system must also be considered in determining which outcome variables are applicable to this age group. For example, recent studies have demonstrated that the forced expiratory volume at 1 second to forced vital capacity (FEV1/FVC) ratio in healthy 5- to 6-year-old children is approximately 90 to 95% (5, 6, 30, 31, 33), and is even higher in younger children. In older children and adults, the physiological and clinical utility of FEV1 is due to its location on the effort-independent (flow-limited) part of the maximal forced expiratory flow–volume (MEFV) curve, which descends to lung volumes as low as 85 to 90% of exhaled vital capacity and reflects intrinsic properties of the respiratory system. The ability to maintain flow limitation at low lung volumes depends largely on the strength of the chest wall muscles to maintain sufficient driving pressure. It is highly likely that children in the preschool age group will not have the chest wall muscle strength to maintain flow limitation to lung volumes as low as 90% of exhaled vital capacity. Although this concept is not new (34), forced expiratory volumes at 0.75 second (FEV0.75) or at 0.5 second (FEV0.5) have not been adopted in clinical practice. Systematic research will be needed to determine the appropriate outcome variables for spirometry in this age group. The answer to the question "Which test should be used in the preschool age group?" depends on the clinical/research question being asked. As is the case in other age groups, no one test will answer all questions. The interrupter technique is easily implemented and is suitable for use in epidemiological studies, particularly those involving measurements in the field. Measurements capable of reflecting changes in the lung parenchyma, such as gas washout techniques and, potentially, forced oscillation, are likely to be more suitable for detecting early lung disease in a condition such as CF, which is known to start in the lung periphery. The clinical and research role of measuring bronchodilator responses and of provocation testing will need to be evaluated. Again, systematic studies using a number of tests will be needed before we know with certainty the place for each test in our clinical armamentarium. In summary, measurement of lung function in preschool-aged children is now feasible. However, much work remains to be done in standardizing how these tests are performed, and in understanding the most appropriate role for the various tests in the study of growth and development of the respiratory system and in the clinical management of children in this age group.
It is now recognized that, given encouragement and suitable measurement conditions, most children between 2 and 6 years old can undertake PFT. Although there is little doubt about the value of these tests in clinical or epidemiological research, their influence on clinical management in an individual remains debatable. The clinical usefulness of any measurement depends on how well it can discriminate between health and disease and how reproducible it is from day to day so that disease progression and response to treatment can be assessed within each child individually. Considerable further work is required to develop appropriate reference data for this age group, with which to reliably distinguish the effects of disease from that of growth and development, together with information on within-subject repeatability and the relative sensitivity and specificity of these tests for distinguishing health from disease. In the meantime, the following recommendations apply:
Evidence accumulating over the last 5 years indicates that PFT in children aged 2 to 6 years produces technically satisfactory measurements using tidal breathing, the interrupter technique, forced oscillation, spirometry, and multiple-breath washout (MBW) methods (26, 31, 35–43). However, the extent to which these measurements are clinically useful in the management of the individual child needs careful consideration. This section will consider the evidence base for the clinical value of lung function measurements in the individual preschool child. It should be noted that many of the issues raised may be equally true in the older child or adult (44), and the clinical value of infant pulmonary function tests also has to be determined (45). It is often claimed that the assessment of a pulmonary function test will help diagnosis, assist prognosis, monitor disease progress, and measure the effect of therapeutic interventions (46). An objective test would supplement history and physical examination in subjects with respiratory problems, which are notoriously difficult to obtain in childhood wheezing disorders (47). The evidence base of clinical decision making (i.e., deciding what is the best test or group of tests for the individual) lags far behind that for treatments (44). A recent review of the value of PFT in adults suggested that many tests used for diagnosis and for assessing a known condition were not supported by high-level evidence (48). For the clinician, it is important to know how helpful PFT is in distinguishing health from disease and in monitoring disease progress in the individual. The epidemiologist and clinical researcher are more interested in comparing measurements in groups and describing the average effects of interventions. PFT could also be helpful in monitoring progress and the response to treatment in children suffering from wheezing disorders, CF, and chronic lung disease of infancy.
Under perfect conditions, most pulmonary function tests discussed in this document can be undertaken successfully in the majority of preschool children older than 3 years. These tests were developed by researchers primarily interested in their application to groups of children to understand the progress of lung development and disease, and the effect of interventions. For their application to the management of individuals, feasibility depends on many more factors than just whether the patient can undertake the test. For example, of 72 preschool children with and without CF, only 58% were able to produce an acceptable forced expiration lasting 1 second, although 73% could manage an FEV0.5 (11). In other words, the international quality-control requirements for spirometry, which are commonly derived from studies in adults, could not be met, although alternative criteria may be feasible. The ATS has made recommendations for training and qualifications of personnel conducting pulmonary function tests. Unlike the measurement of peak flow in a respiratory clinic by the physician managing the patient's care, preschool tests require time and patience by technicians trained in the techniques that can help young children to perform at their best, who can maintain the equipment, and who can understand the procedure well enough to know when a result is or is not acceptable. In some cases, a laboratory operator may not meet normal training criteria for PFT but has particular skills in working with young children. In such cases, flexibility is recommended. Some of the tests are suitable for use in the ambulatory setting or in the community, but most require laboratory equipment.
Choosing Reference Data Reference equations are essential to express pulmonary function in relation to that which would be expected for healthy children of similar age, sex, body size, and ethnic group. The choice of reference equations directly influences the interpretation of pediatric pulmonary function data, and this can have a significant impact on patient care and research (49–52). Most lung function data are normally distributed or can be transformed to such, so that 90% of "normal" values are found within the range of mean ± 1.65 SD (with 95% within ± 1.96 SD). Lung function variables in healthy subjects and those with respiratory symptoms and/or disease often overlap to such an extent that a normal lung function measurement does not exclude disease. Clearly abnormal lung function measurements will often, but not necessarily, be associated with symptoms and disease. The tests for which preliminary reference data are available are listed in Table 1. Ideally, data from healthy children should be evenly distributed across the age range from 2 to 6 years, but in many studies, there are few data in children younger than 4 years, and this could result in distortion of any derived prediction equations. Inspection of the datasets should identify those in which there are a disproportionate number of older children, although, regrettably, plots of raw data are not always presented in published reports. Display of the raw data plotted against height or age also allows the potential user to assess whether linear regression is appropriate when modeling the data and whether data are normally distributed about the regression line (e.g., whether approximately equal numbers lie above and below 2 SDs from the regression line). Evidence for differences among ethnic groups and between sexes should also be considered. Data from spirometry, and resistance measured by Rint and plethysmographic sRaw, have so far shown similar results for boys and girls, but some sex differences may exist with respect to the FOT technique (see SECTION 6).
The most important consideration when choosing reference data is that the method, equipment, and software used to collect the data should be the same as that used by the clinician for his or her patients. In the oscillation technique, regression equations for total respiratory resistance can differ considerably (16), which may reflect the different methods used. This is true also for the interrupter technique, in which the most important consideration is the calculation of pressure (53), which differs according to the algorithm used (see SECTION 5). Particular caution is required when undertaking techniques such as plethysmography and spirometry using commercially available equipment. The default prediction equations from such equipment will almost always be based on reference data derived from older subjects, possibly resulting in serious misinterpretation if applied to preschool children.
Using Reference Data
Height or age as the main predictor?
Expressing results.
In this document, the different aspects of variability are named as follows: repeatability is the within-occasion (short-term) variability and reproducibility is the between-occasion (long-term) variability. Therefore, variability refers to either short- or long-term changes. To evaluate the variability of a test, it is necessary to know both the within-occasion repeatability and between-occasion reproducibility of the measurements in healthy children and in those with respiratory complaints. If PFT is to be used to measure the effect of an intervention, such as response to a bronchodilator, then the variability of the test over the same time interval as the expected response to treatment should be known. Only then can the clinician decide whether any observed change can be ascribed confidently to the bronchodilator (or other intervention) rather than simply reflecting intrinsic variability of the measurement or disease state over the same period (57). Variability may differ between studies due to factors such as technical differences and differences in population, or issues such as the interval between repeated tests, methods of analyses, inclusion criteria for technically acceptable data, and so forth.
Within-Occasion Repeatability
The definition of a single "measurement" varies among different techniques. Thus, with the interrupter technique, it is generally reported as the median of five or more satisfactory readings, whereas with spirometry, the best of three technically acceptable readings is usually reported. The intrameasurement repeatability is usually expressed as a coefficient of variation (CV), which is the SD expressed as a percentage of the mean (i.e., 100 x SD/mean). For example, the repeatability of Rint assessed by CV may differ among studies (see the tables in the online supplement) (25). The within-occasion intermeasurement repeatability is often reported as the coefficient of repeatability (CR)—that is, twice the SD of the mean difference between two series of baseline measurements, performed a few minutes apart, without any intervention, in a group of children. The CR defines the limits above and below an individual measurement within which 95% of second measurements will lie. Finally, some authors have expressed variability using the SD of the mean difference between two measurements and divide it by
Between-Occasion Reproducibility
The arguments about the potential usefulness of tests in clinical decision making have been discussed recently (44). Although clinical decision making is not an exact science, it may be guided by international recommendations. Because diagnosing mild asthma does not necessarily imply treatment with drugs, such as corticosteroids, then the typical threshold values that have been quoted recently for a positive test in preschool children (Table E5) are probably useful, provided the clinician knows the likelihood of a false-positive test (Figure 1). This is where clinical judgement and other tests, such as tests of atopic status, must complement PFT (60).
Before the clinician selects a pulmonary function test, he or she must know what the results are likely to be in children with the disorder(s) that is being considered. There is a large overlap between measurements in children with mild asthma or isolated cough and healthy children (40).
Asthma Because of the great overlap of measurements between healthy subjects and those with previous wheeze, the diagnostic accuracy of baseline PFT is generally very poor in any age group. Bronchodilator responsiveness (BDR) has been recommended in the workup of adults and children with asthma for whom measurements of BDR give a much better diagnostic profile than that obtained from baseline lung function data. For example, in a study of 48 healthy and 82 previously wheezy children aged 2 to 5 years or younger, 76% of those with asthma had a BDR (expressed as a ratio of baseline Rint–postbronchodilator Rint) of 1.22 or greater (40). The sensitivity for this value was therefore 0.76. By contrast, although 70% of healthy children had values below 1.22 (indicating a specificity of 0.70), 30% did not—that is, their BDR was 1.22 or more (giving a value for 1-specificity [i.e., false positive] of 0.3). Plotting sensitivity and 1-specificity against each other for a range of baseline and BDR Rint values produces receiver operating curves (ROC) (62) (see Figure 1). The diagnostic profiles for PFT for asthma and their threshold values are detailed in Table E5. It should be noted that the confidence intervals for these figures are quite wide. Rint, FOT (measured at 5 Hz), and plethysmographic sRaw appear to have similar profiles for the thresholds given (Table E5). Data to calculate the specificity and sensitivity of BDR measurements for wheeze in preschool children measured using spirometry are not available. The data available from a small group of young children suggest that the diagnostic profile of spirometry for BDR may be poor because there is so much overlap between measurements in children with and without lung disease (39). Although challenge testing to demonstrate bronchial hyperresponsiveness in preschool children is possible (see SECTION 8), the feasibility of pharmacological challenges in consecutive children younger than 5 years as a clinical tool outside a research laboratory has yet to be reported. The accuracy and repeatability of the bronchial hyperresponsiveness tests are dependent on the technique used (63, 64).
CF Lung Disease |
TOP
CONTENTS
Section 1. The Next...
maxFRC) (
to obtain the within-subject SD (SDw) (