Spirometry in an unselected group of 6-year-old children: the darc birth cohort study

Pediatric Pulmonology 43:806–814 (2008) Spirometry in an Unselected Group of 6-Year-Old Henrik Fomsgaard Kjaer, MD,1* Esben Eller, MSc,1 Summary. This study presents reference equations for spirometric parameters in 6-year-oldchildren and evaluates the ability of spirometry to discriminate healthy children from children withasthma. Baseline spirometry and respiratory symptoms were assessed in 404 childrenparticipating in a longitudinal birth cohort study. Children with known asthma, possible asthmaand a control group also performed bronchodilator measurements. At least two acceptable flow-volume curves at baseline were obtained by 368/404 children (91%). The two best values for FEV1and FVC were within 5% of each other in 88% and 83% of children, respectively. Linear regressionanalyses for 242 children included in the reference population demonstrated height to be the mainpredictor of all spirometric indices except FEV1/FVC. FEV1, FEV75, and FVC correlatedreasonably to anthropometric data in contrast to flow parameters. Gender differences were foundfor FEV1, FVC, and FEV75, but not for flow parameters. Asthma was diagnosed in 25/404 children.
Baseline lung function in healthy children and children with asthma overlapped, although asthmaticchildren could be discriminated to some extent. Bronchodilator tests showed a difference inDFEV1(mean) between healthy children and children with asthma (3.1% vs. 6.1%, P < 0.05). At acut-off point of DFEV1 ¼ 7.8%, bronchodilator tests had a sensitivity of 46% and a specificity of92% for current asthma. Spirometry including bronchodilator measurements was demonstratedto be feasible in 6-year-old children and reference values were determined. Spirometry aids thediagnosis of asthma in young children, but knowledge on sensitivity and specificity of thesemeasurements is a prerequisite. Pediatr Pulmonol. 2008; 43:806–814.
Key words: lung function tests; spirometry; reference values; preschool children; children.3 The clinical significance of spirometry inpreschool children6–8 and in particular of bronchodilator The diagnosis of asthma in children often represents measurements in this age group14–16 remains unsolved.
a challenge to the clinician, as the disease presents in The aim of this study was to evaluate the feasibility and various ways and comprises heterogeneous symptoms.
repeatability of spirometry in an unselected cohort of Moreover, different phenotypes of wheeze exist in 6-year-old children and to determine reference equations preschool children with different clinical patterns and for different spirometric indices. The purpose was prognoses. The majority of children presenting withwheeze and cough will outgrow their symptoms,1,2 yetfor others persistent asthma will manifest. A reliable 1Department of Dermatology, Allergy Centre, Odense University Hospital, objective tool aiding the diagnosis of asthma would therefore be of great value. In order to evaluate dynamic 2Department of Pediatrics, Odense University Hospital, Odense, Denmark.
spirometry as such a tool in preschool children, measure-ments of lung function need to be both feasible, repeatable The authors declare no conflicts of interests.
and reproducible.3 Furthermore, establishment of refe-rence equations and assessment of the inter-individual Grant sponsor: John and Birthe Meyer Foundation.
biological variability of different lung function parame- *Correspondence to: Henrik Fomsgaard Kjaer, MD, Allergy Centre, ters are required for the interpretation of individual data Odense University Hospital, 29 Sdr. Boulevard, Odense DK-5000, sets. Reference equations for preschool children4–9 as Denmark. E-mail: [email protected] well as for school children10–13 have been published, but arecent joint publication from The American Thoracic Received 11 February 2008; Revised 25 April 2008; Accepted 4 June 2008.
Society (ATS) and The European Respiratory Society (ERS), emphasizes the need for development of reference Published online 10 July 2008 in Wiley InterScience equations that can cross from preschool- into school-age furthermore to evaluate the discriminatory capacity of TABLE 1— Anthropometric Data of Total Study Population spirometry and in particular bronchodilator measurements in differentiating children with asthma from healthy In an ongoing non-interventional birth cohort study of 562 unselected children concerning allergic diseases, Numbers are expressed as mean Æ standard deviation or in %.
404 children (72%) attended the clinic at 6 years of age.
A detailed description of the establishment of the cohort Daily volume check (and calibration if needed) was and the mode of selection has previously been pub- performed with a 3 L calibration syringe according to the lished.17 No differences were found between the children manufacturer’s instruction. Children were at rest at least originally declining participation in the study and the 15 min prior to lung function measurements. All measure- final cohort, nor between the cohort and the analyzed ments were carried out by the same investigator. Standing subsample at 6 years of age on a number of character- height and weight without shoes were measured. Each istics.18 Follow up examinations at 3, 6, 9, 12, 18, 36, child was instructed thoroughly to take the deepest breath and 72 months of age included detailed assessment of possible and then to blow as hard, fast and for as respiratory symptoms through extensive questionnaire long time as possible into the mouthpiece. The technique based interviews (regarding atopic heredity, medical was demonstrated by the investigator repeatedly, and history, environmental and demographic factors), clinical before any measurements were performed the child examination by a physician, skin-prick-testing and IgE practiced on the mouthpiece alone under supervision.
measurements. Measurement of lung function by dynamic Individual flow-volume curves were then made with the spirometry was attempted in all children attending the child standing wearing a nose clip. At least three clinic at 6 years of age. A subgroup of healthy children acceptable flow-volume curves were recorded whenever suitable for the establishment of reference equations for possible. All individual curves were visually inspected different pulmonary function parameters was identified.
Excluded from this group were children who fulfilled atleast one of the following criteria: (1) The flow-volume curve demonstrated a rapid rise to peak flow and a gradual flow decrease with no signs (1) Doctor’s diagnosis of asthma (or other chronic of abrupt end (sharp drop of flow) or evidence of cough or glottic closure. A distinct peak (and one (2) Wheeze, shortness of breath or episodes of cough (apart from upper respiratory infections) at least once (2) The ATS start of test criterion (Vbe (volume back within the last year or reported at least twice during extrapolated) <0.15 L) was fulfilled.19 (3) The volume-time trace was approaching a horizontal (3) Respiratory infection within the last 3 weeks.
(4) The duration of expiration was !1 sec.
Anthropometric data of the study population are From acceptable curves, the highest FEV1, FEV75, and presented in Table 1. Written consent for the study was FVC were reported even if they were not from the same obtained from the parents, verbal assent from the children.
curve. For the analyses of repeatability (short-term The study protocol was approved by the local ethics variability) these values were compared with the second highest value of the same indices from acceptable curves.
All flow indices reported (MEF25, MEF50, MEF75, MMEF, and PEF) were selected from the ‘‘best manoeu-vre’’ defined as the curve with the largest sum of FEV Spirometry was performed using the MicroLoop equipment (Micro Medical Ltd., Rochester, UK) andthe corresponding SPIDA Spirometry Software V 5.0 including a computer-animated volume driven incentive(‘‘the bubblegum kid’’). All tests were performed between Depending on the information given in the interview December 2004 and January 2006. All children were about respiratory symptoms, children could be divided completely inexperienced in performing spirometry.
(1) Children with a doctor’s diagnosis of asthma For all three criteria, a clear subjective or objective established before the present evaluation (at 6 years).
response to treatment with b2-agonists and/or inhaled (2) Children with possible asthma (defined as at least corticosteroids was demanded for the diagnosis of asthma.
one confirmatory answer to questions about wheeze,coughing episodes apart from during colds, breathing problems or reduced physical activity within the last Reference equations were generated through multiple linear regression analysis assessing the relationship (3) Children with no respiratory symptoms within the between the spirometric indices and the explanatory variables height, weight and gender in the healthysubpopulation. Since all children were the same age, Children in the first two groups were submitted to inclusion of that variable added no information. Linear bronchodilator testing irrespective of their baseline lung relationships between dependent and explanatory varia- function. A control group chosen randomly as every bles showed the best correlations (compared to quadratic, eighth child examined with no previous respiratory logarithmic and exponential models) evaluated by the disease or symptoms (apart from during colds) was explained fraction of variance (i.e., the adjusted R2) and submitted to bronchodilator testing. Four doses of the distribution of residuals. Regression coefficients were Bricanyl1, AstraZeneca (0.25 mg/dose) were admini- regarded significant when P < 0.05. The 95% prediction stered to the child via NES-spacer1 after baseline testing intervals were calculated as Æ2 RMSE (root mean squared and spirometric measurements repeated 15 min later.
error) from the regression line. Z-scores defined as The test was regarded positive if DFEV1 (as a percentage observed value-predicted value/RMSE of the reference of the baseline value) !12% (between the highest FEV1 population were calculated from our reference equations.
from technically acceptable pre- and post-bronchodilator The Wilcoxon rank sum test was used for comparison between groups. All statistical analyses were performed If any of the above symptoms were related to physical with Stata 9.1 (Stata Corporation, College Station, TX).
activity (or information was given on reduced physicalactivity), an exercise test was performed after the baseline testing (and before bronchodilator measurements). Theexercise test was adjusted to the age group and interpreted At least one acceptable flow-volume curve was produced by 376/404 children (93%). The remainingchildren either refused measurements (n ¼ 10), were unable to produce any acceptable curves (n ¼ 10) or theirdata were lost due to software error (n ¼ 8). At least two All children attending the clinic at 6 years of age were acceptable flow-volume curves were achieved by 368/404 thoroughly evaluated for asthma. If asthma was suspected children (91%). The difference between the two best and a clear cut diagnosis could not be obtained at the initial curves could be demonstrated to be 5% for 88% and 83% visit, a follow up visit was scheduled 3 months later to repeat clinical assessment and spirometric testing. Spiro- compared to 97% and 98% using a 10% repeatability metric data presented in this article are exclusively from the children’s initial visit to the clinic.
Asthma was diagnosed if one of the following criteria The data for 368 children with at least two acceptable (1) Recurrence of at least two of the three symptoms flow-volume curves were used for further analysis. Of cough, wheeze and shortness of breath within the these children, 242 (66%) were part of the reference previous 12 months (symptoms not triggered only by population; 131 girls and 111 boys. Linear regression analyses on the reference population for one explanatory (2) Doctors diagnosis of asthma prior to follow up visit variable (height, weight, gender) at a time, demonstrated (in combination with ongoing treatment).
height to be the main predictor for all spirometric indices (3) Symptoms suggestive of asthma in combination with except for FEV1/FVC. In the analysis of FEV1 and height a positive exercise- or bronchodilator test. Symptoms only, the explained fraction of variance (R2) was 47% and suggestive of asthma were isolated persistent cough, based on a single explanatory variable the reference wheeze or shortness of breath (apart from during colds), exercise induced symptoms or reduced phys-ical activity.
FEV1 ¼ À1:489 þ 0:24 Â height ðcmÞ; Fig. 1. FEV1 (L) by height (cm) for children in the reference Fig. 2. MMEF (L/s) by height (cm) for children in the reference population (n ¼ 242) (a) and outside the reference population population (n ¼ 242) (a) and outside the reference population (n ¼ 126) (b) presented with the regression line and 95% (n ¼ 126) (b) presented with the regression line and 95% prediction intervals. The dashed line (– – –) in (a) indicates the prediction intervals. The dashed line (– – –) in (a) indicates the regression line found by Eigen et al.4 for comparison.
regression line found by Eigen et al.4 for comparison.
Multivariate regression analysis with respect to height, metric parameters based on the explanatory variables weight and gender was then performed. For FEV1 the inclusion of gender (in addition to height) improved R2to 52%, whereas weight added no further explanation tothe variance. Inclusion of a gender-difference in reference equations is often done by assuming similar slope coeffi- According to the criteria outlined, asthma was diag- cients for girls and boys, making the regression line for nosed in 25/404 children (6.2%) at 6 years of age. Of these, one gender a parallel shift of the other. A more adequate 16/25 children had a doctors diagnosis of asthma before approach would be to perform gender-specific regression visiting the clinic, while 9/25 were diagnosed at the visit analysis. In doing so, the addition of weight to the equation (Table 3). Two children with asthma could not perform proved significant for boys with a slight increase in R2 spirometry (ID 3 and 13, Table 3). Lung function from 50% to 51%. The reference equation for boys based parameters measured at baseline for children with asthma overlapped the values of healthy children, although children diagnosed with asthma at this follow up could be separated from healthy children to some extent The explained fraction of variance for FVC in the crude (Figs. 1b and 2b). Depending on the cut off point chosen, linear regression analysis was 49%, but improved in the MMEF and MEF50 showed slightly higher discriminatory multivariate analysis with inclusion of gender (R2 ¼ 55%) capacity at baseline compared to FEV1, but this was not and slightly if weight was included in addition to height significant (data not shown). Children with asthma and gender (R2 ¼ 57%). In the gender-specific regression diagnosed prior to the visit could not be discriminated analysis weight was only a significant regression co- from the rest of the population as exemplified by FEV1 and efficient for boys. Height was the only significant expla- MMEF (Figs. 1b and 2b). Since the treatment of asthma in natory variable for all flow parameters, but the explained these children had already been initiated by a doctor, the fraction of variance was low as exemplified by MMEF reason for this finding is probably that a subgroup were (R2 ¼ 9%) (Fig. 2a). Reference equations for all spiro- treated with inhaled corticosteroids at time of testing, TABLE 2— Regression Coefficients (With 95% Confidence Intervals [CI]) for Different Spirometric Indices in 242 HealthyChildren (111 Boys, 131 Girls) Satisfying Our Criteria for the Healthy Reference Population *Regression coefficient not significant (P ! 0.05).
TABLE 3— Characteristics of Children Diagnosed With Current Asthma 1Current asthma, but diagnosed before the examination at 6 years of age.
2Ongoing treatment regime at time of 6 years examination. ICS: inhaled corticosteroids, b2: b2-agonists.
3Symptoms within the last year and unrelated to common colds/respiratory infections.
4Positive bronchodilator test.
5Positive exercise test.
6Asthma criterion fulfilled. Some children fulfilled more than one criterion, but listed here is the ‘‘main criteria’’ prioritized in the order 1, 2, and 3,respectively.
7Unable to perform spirometry.
8Inhaled corticosteroids prevented these children from experiencing severe symptoms within the last year (but were paused at time of 6 yearsexamination).
while the remaining children had mild asthma (treated group. A total of 8 children (9%) in the three groups could only with b2-agonists) and therefore normal lung function not achieve any acceptable bronchodilator curves. The results of all bronchodilator tests are presented in Figure 3.
The mean bronchodilator-induced change in FEV1differed between healthy subjects and those with con- firmed asthma according to our criteria (3.1% vs. 6.1%,P < 0.05), but clearly the overlap between these groups On the basis of information given in the interviews, complicates the interpretation of individual results. A cut- the 404 children could be divided into three main groups; 18 children with asthma already diagnosed before the visit asthma from the rest of the children with bronchodilator (of whom 16 children could be confirmed to have current results with a specificity of 92%, but a sensitivity of only asthma), 44 children with respiratory symptoms within the last year apart from during colds (‘‘possible asthma’’) and342 asymptomatic children. Acceptable measurementof the bronchodilator response was possible in 16/18 children with known asthma and 34/44 children with This study describes the assessment of baseline lung possible asthma. In the latter group, the bronchodilator test function and bronchodilator responses in an unselected was omitted for 8 children, since the symptoms reported group of 6-year-old children. Reference equations are occurred only once and were vague and clearly not related presented and the ability of spirometry to discriminate to asthma. Successful measurement of the bronchodilator healthy children from children with asthma is evaluated.
response was obtained for 33/37 children in the control Strict quality control of all individual curves obtained by spirometry is essential for the reliability of thesemeasurements. Some of the ATS/ERS criteria23 for theacceptability of spirometric curves are not suitable forpreschool children,24 but very intriguing work onvalidated criteria for this particular age group has recentlybeen published.3,7,24,25 Our criteria for the acceptance ofindividual curves were selected on the basis of therecommendations in these publications. The upper limitfor Vbe of 150 ml in this study (default in our software) isprobably too high in this age group, although a definitelimit still needs validation and should be used only as aguide for the mandatory visual inspection of curves.3,24,25Not all preschool children can achieve the end-of-testcriteria recommended demanding a forced expiratory time(FET) !3 sec.23 Quality control studies for our particularage group suggest a limit for FET of !1 sec (assuming noabrupt end of expiratory flow and a volume-time traceapproaching a plateau).7,24 According to the selectioncriteria described, 376/404 (93%) of the children studiedachieved one acceptable flow-volume curve, 368/404(91%) at least two. The repeatability of our measurementswas high; 88% and 83% of the children fulfilled the 5%repeatability criterion for FEV1 and FVC, respectively.
This is in agreement with Arets et al.24 (91% and 83%respectively for children <8 years), but most studiesreport lower success rates for fulfillment of this strictcriterion,5,25–27 in some cases explained by slightlyyounger study populations than ours. Almost all childrenin this age group apply to the 10% repeatabilitycriterion,4,5,25,27 as also supported by our study (97% Fig. 3. Bronchodilator responses (DFEV1(%) of FEV1 baseline) in and 98% for FEV1 and FVC, respectively).
four groups of children tested for different reasons. Indicated The normative data presented in this article has is also if a diagnosis of asthma could be confirmed accordingto the criteria described: * (Criterion 1), & (Criterion 2), and strengths and limitations. As studies on preschool children4–6,9,27 and school children10–13 mainly includes younger and older children than this study, only one prior spirometry can aid the diagnosis of asthma in some publication include a very high number of 6-year-old children if results are interpreted on a basis of knowledge children.28 On the other hand, the narrow age span of our on sensitivity and specificity of these tests. Although no population (6.10 Æ 0.14 years) is an important limitation parameter was found superior to FEV1 in discriminating to the usefulness of our reference equations. Strictly, our healthy children from children with asthma at baseline, normative data are only suitable for use within our age we found that MEF50 and MMEF were the most dis- range (5.8–7.1 years) and height range (100–130.5 cm), criminatory indices, in line with similar studies in 5- and but since age has been shown to have little or no 7-year-old children.1,7 A few studies have published effect on lung function parameters when adjusted for reduced spirometric parameters in preschool children with height,4,6,9,10,27 reference equations should primarily be asthma and wheeze.6–8 Vilozni et al.8 interestingly found chosen on the basis of the height range. The explained that children with mild asthma did not differ from healthy fraction of variance (R2) between height and FEV1 and children regarding spirometric indices, children with FVC respectively, varies between 0.34–0.85 for FEV1 moderate asthma had FEV1 and FVC values within the and 0.33–0.86 for FVC in different studies 4–9 compared healthy range, but flows at low lung volumes significantly to our values in the crude analysis of 0.52 and 0.57 for lower than healthy children. In children with severe FEV1 and FVC, respectively. Low values of the explained asthma they found all spirometric indices significantly fraction of variance will result in wide normal values, but reduced. These findings are highly important, but also if the high variance is a true reflection of the healthy underline the limitations of spirometry as a diagnostic population, the reference values should be wide. Low tool, since the children that are most difficult to diagnose values could also reflect insufficient selection of curves or overlap substantially with the healthy population regar- insufficient efforts by the children, but if quality control criteria are followed this should be a minor problem. Since Our evaluation of bronchodilator testing as a discri- no consensus has been reached on the criteria defining a minatory tool has limitations. A positive bronchodilator reference population, this matter relies mostly on arbitrary test according to the definition described is used in our (although quite similar) choices in the studies published so diagnostic criteria for asthma with the possibility of far. Anyhow, the identification of children with symptoms arguing in a circle (three children are diagnosed with in the past is far more reliable in a prospective study asthma because they have high bronchodilator responses recording symptoms when they occur and thereby in combination with symptoms). Still, our results indicate diminishing the risk of recall bias.
that there is a large overlap between healthy children and The predicted values presented in this article are in children with asthma, underlining the need for knowledge agreement with most other studies on preschool children on sensitivity and specificity of this diagnostic test. The of the same height4,5,9 but differ from the values obtained fact that a large proportion of the asthmatic children in our in a large Australian cohort study.28 Eigen et al. and study exhibits low bronchodilator responses may have Zapletal et al. found no gender differences with respect to different explanations. As mentioned, two-thirds of our FEV1, whereas Nystad et al. and Joseph-Bowen et al. both asthmatic children were diagnosed before the visit and published gender differences very similar to our results.
half of them treated with inhaled corticosteroids with the Our reference equations for flow parameters are also in aim of reducing the bronchial hyperresponsiveness in line with these studies, but as a recent published study of these children. Furthermore, the proportion of children 5-year-old children,7 we found the correlation between with severe asthma likely to exhibit the largest broncho- dilator response was low in our cohort. An important study Our evaluation of spirometry as a tool to separate regarding the use of bronchodilator tests in children with healthy children from children with asthma has certain mild intermittent asthma was conducted by Dundas et al.15 strengths as well as limitations. An important strength is They found the optimal cut-off point for previous wheeze the extensive diagnostic work-up for asthma. Most studies to be a 9% change in predicted FEV1 with a resulting rely solely on questionnaire based information in defining sensitivity of 50% and a specificity of 86%, close to the important endpoints used for discriminatory analyses findings in this study. Our bronchodilator results were also and furthermore use endpoints like wheeze or a history of confirmed by Bibi et al.14 In contrast, Marotta et al.16 asthma. The most important limitation in our analysis of found the change in FEV1 after bronchodilation to be the discriminatory capacity of spirometry is the size and higher in the control group as opposed to asthmatic the composition of the asthma group. Not only is the total children, although not statistically significant.
number of children with asthma small, two-thirds were All studies on the discriminatory power of lung also diagnosed before this examination and the children functions measurements face the same difficult obstacles.
with more severe asthma treated with inhaled cortico- The definition of asthma needs to be precise in order to steroids affecting their lung function measurements. With evaluate spirometry as a diagnostic tool. On the other these limitations borne in mind, we found that baseline hand, the group of children with more vague symptoms is difficult to categorize and these are exactly the children, 7. Turner SW, Craig LC, Harbour PJ, Forbes SH, McNeill G, Seaton where spirometry could be most helpful in aiding the A, Devereux G, Helms PJ. Spirometry in 5-year-olds-validation of current guidelines and the relation with asthma. PediatrPulmonol 2007;42:1144–1151.
In conclusion, this study confirms the feasibility and 8. Vilozni D, Barak A, Efrati O, Augarten A, Springer C, Yahav Y, repeatability of spirometry in 6-year-old children and Bentur L. The role of computer games in measuring spirometry in presents reference equations for different spirometric healthy and ‘‘asthmatic’’ preschool children. Chest 2005;128: parameters. Our study supports the use of baseline spirometry and bronchodilator tests in the diagnostic 9. Zapletal A, Chalupova J. Forced expiratory parameters in healthy preschool children (3–6 years of age). Pediatr Pulmonol 2003; evaluation of asthma in children, although clinical decisions must be based on knowledge of the sensitivity 10. Knudson RJ, Lebowitz MD, Holberg CJ, Burrows B. Changes in and specificity of these tests. Future follow up examina- the normal maximal expiratory flow-volume curve with growth tions of the cohort will not only provide important and aging. Am Rev Respir Dis 1983;127:725–734.
knowledge on the longitudinal development of pulmonary 11. Polgar G, Promadhat V. Pulmonary function testing in children: techniques and standards. Philadelphia: W.B. Saunders; 1971.
function in these children, but also enable an evaluation of lung function at 6 years of age as a predictor of subsequent 12. Quanjer PH, Stocks J, Polgar G, Wise M, Karlberg J, Borsboom G. Compilation of reference values for lung function measure-ments in children. Eur Respir J Suppl 1989;4:184S–261S.
13. Zapletal I, Samanek M, Paul T. Lung function in children and adolescents. Methods, reference values. Prog Respir Res 1987;22: 14. Bibi H, Montgomery M, Pasterkamp H, Chernick V. Relationship We thank the participating families for their immense between response to inhaled salbutamol and methacholine commitment to the project. We also express our gratitude bronchial provocation in children with suspected asthma. Pediatr to nurse Isabel Mejrup for her untiring work with the children and their families and to Professor Werner Vach 15. Dundas I, Chan EY, Bridge PD, McKenzie SA. Diagnostic for statistical advice. Funding was obtained from The John accuracy of bronchodilator responsiveness in wheezy children.
Thorax 2005;60:13–16.
16. Marotta A, Klinnert MD, Price MR, Larsen GL, Liu AH. Impulse oscillometry provides an effective measure of lung dysfunction in4-year-old children at risk for persistent asthma. J Allergy Clin 17. Johnke H, Vach W, Norberg LA, Bindslev-Jensen C, Host A, 1. Illi S, von ME, Lau S, Niggemann B, Gruber C, Wahn U.
Andersen KE. A comparison between criteria for diagnosing Perennial allergen sensitisation early in life and chronic atopic eczema in infants. Br J Dermatol 2005;153:352–358.
asthma in children: a birth cohort study. Lancet 2006;368:763– 18. Kjaer HF, Eller E, Høst A, Andersen KE, Bindslev-Jensen C.
(e-pub ahead of print). The prevalence of allergic diseases in an 2. Wennergren G, Amark M, Amark K, Oskarsdottir S, Sten G, unselected group of 6 year-old children. The DARC Birth Cohort Redfors S. Wheezing bronchitis reinvestigated at the age of Study. Pubmed; Pediatric Allergy Immunol. 2008, March 3.
10 years. Acta Paediatr 1997;86:351–355.
19. Standardization of Spirometry, 1994 Update. American Thoracic 3. Beydon N, Davis SD, Lombardi E, Allen JL, Arets HG, Aurora P, Society. Am J Respir Crit Care Med 1995;152:1107–1136.
Bisgaard H, Davis GM, Ducharme FM, Eigen H, Gappa M, 20. Pellegrino R, Viegi G, Brusasco V, Crapo RO, Burgos F, Casaburi Gaultier C, Gustafsson PM, Hall GL, Hantos Z, Healy MJ, Jones R, Coates A, van der Grinten CP, Gustafsson P, Hankinson J, MH, Klug B, Lodrup Carlsen KC, McKenzie SA, Marchal F, Jensen R, Johnson DC, MacIntyre N, McKay R, Miller MR, Mayer OH, Merkus PJ, Morris MG, Oostveen E, Pillow JJ, Navajas D, Pedersen OF, Wanger J. Interpretative strategies for Seddon PC, Silverman M, Sly PD, Stocks J, Tepper RS, Vilozni lung function tests. Eur Respir J 2005;26:948–968.
D, Wilson NM. An official American Thoracic Society/European 21. Vilozni D, Bentur L, Efrati O, Barak A, Szeinberg A, Shoseyov D, Respiratory Society statement: pulmonary function testing in Yahav Y, Augarten A. Exercise challenge test in 3- to 6-year-old preschool children. Am J Respir Crit Care Med 2007;175:1304– asthmatic children. Chest 2007;132:497–503.
22. Niggemann B, Jacobsen L, Dreborg S, Ferdousi HA, Halken S, 4. Eigen H, Bieler H, Grant D, Christoph K, Terrill D, Heilman DK, Host A, Koivikko A, Koller D, Norberg LA, Urbanek R, Valovirta Ambrosius WT, Tepper RS. Spirometric pulmonary function in E, Wahn U, Moller C. Five-year follow-up on the PAT study: healthy preschool children. Am J Respir Crit Care Med 2001; specific immunotherapy and long-term prevention of asthma in children. Allergy 2006;61:855–859.
5. Nystad W, Samuelsen SO, Nafstad P, Edvardsen E, Stensrud T, 23. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Jaakkola JJ. Feasibility of measuring lung function in preschool Coates A, Crapo R, Enright P, van der Grinten CP, Gustafsson P, children. Thorax 2002;57:1021–1027.
Jensen R, Johnson DC, MacIntyre N, McKay R, Navajas D, 6. Piccioni P, Borraccino A, Forneris MP, Migliore E, Carena C, Pedersen OF, Pellegrino R, Viegi G, Wanger J. Standardisation of Bignamini E, Fassio S, Cordola G, Arossa W, Bugiani M.
spirometry. Eur Respir J 2005;26:319–338.
Reference values of Forced Expiratory Volumes and pulmonary 24. Arets HG, Brackel HJ, van der Ent CK. Forced expiratory flows in 3–6 year children: a cross-sectional study. Respir Res manoeuvres in children: do they meet ATS and ERS criteria for spirometry? Eur Respir J 2001;18:655–660.
25. Aurora P, Stocks J, Oliver C, Saunders C, Castle R, Chazipar- 27. Pesant C, Santschi M, Praud JP, Geoffroy M, Niyonsenga T, asidis G, Bush A. Quality control for spirometry in preschool Vlachos-Mayer H. Spirometric pulmonary function in 3- to children with and without lung disease. Am J Respir Crit Care 5-year-old children. Pediatr Pulmonol 2007;42:263–271.
28. Joseph-Bowen J, de Klerk NH, Firth MJ, Kendall GE, Holt PG, 26. Crenesse D, Berlioz M, Bourrier T, Albertini M. Spirometry in Sly PD. Lung function, bronchial responsiveness, and asthma in a children aged 3 to 5 years: reliability of forced expiratory community cohort of 6-year-old children. Am J Respir Crit Care maneuvers. Pediatr Pulmonol 2001;32:56–61.

Source: http://www.darc14.dk/Portals/0/Papers/Kjaer%20et%20al%20(2008)%20Spirometry%20in%20an%20unselected%20group.pdf