Pediatric Pulmonology 43:11931197 (2008)

Transient Reference Values for Impulse Oscillometry for Children Aged 318 Years
ski, PhD,2 ena Nowowiejska, MD,1 Waldemar Tomalak, PhD,2* Jakub Radlin Boz 1 2 Grzegorz Siergiejko, MD, Wojciech Latawiec, PhD, and Maciej Kaczmarski, MD, PhD1
Summary. Impulse oscillometry (IOS) is a technique of assessing mechanical properties of respiratory system by means of measuring resistances and reactances in a number of frequencies during tidal breathing. It is especially useful in preschool children, however has also been validated in older children and adults. The aim of the present study was to construct equations describing normal values of oscillatory parameters in pediatric population of healthy polish children. Six hundred twenty-six healthy children aged 3.118.9 years (278 boys and 348 girls) completed the study. Analysis revealed that body height was the best predictor and resistances are best described with exponential model while reactances with linear one, with correlation coefcient r reaching the value of 0.9. Oscillometric resistances decrease with height, while reactances increase. Reference values for children and adolescents will allow not only the interpretation of the measurement, but also will make possible to study changes of oscillometric indices during growth. Pediatr Pulmonol. 2008; 43:11931197. 2008 Wiley-Liss, Inc. Key words: children; adolescents; impulse oscillometry; reference values.

INTRODUCTION

Impulse oscillometry (IOS) introduced in early 1990s is an alternative technique of studying respiratory system properties especially in children. As the measurements are made during tidal breathing and require less cooperation than spirometry, the IOS is used with success in children from 2 years of life. IOS measurements have been shown especially useful in preschool children2in assessing bronchomotor response to different stimuli, however, the usefulness in older children is also documented.3 IOS derived resistances have been shown to correlate strongly with plethysmographic airway resistance3,4 and spirometric FEV1, while reactances seems to be useful in evaluating bronchomotor response.5 Recent joint ERS/ ATS statement6 on pulmonary function testing in preschool children, which summarizes the application of different techniques in children have stressed the importance of the forced oscillation technique (including IOS) in children. The usefulness of the technique is obvious not only in preschool children, but also in older ones, as IOS is able to identify airway obstruction and the response to bronchodilatators and bronchoconstrictors. It may be fully performed in settings ranging from eld studies to the emergency room. In the last few years several papers concerning reference values have been published7,8 but they were concentrated rather on the youngest groups of children. As IOS offers the unique chance to study respiratory system properties starting at the age of 27,9 through childhood and adolescency, we have attempted to create equations for
2008 Wiley-Liss, Inc.

normal values for oscillometric parameters in children and adolescents aged 319 years.
MATERIALS AND METHODS

The study was performed in kindergartens, primary and secondary schools as well as in lycees in Bialystok area in northeast Poland. School authorities have been contacted and childrens parents gave informed consent. The study has been approved by an Ethic Committee of Bialystok Medical Academy. Prior to the measurements a short questionnaire oriented to past respiratory diseases has been distributed to the parents. Exclusion criteria for the study, according to the recommendations of the GAP Conference were as
1

IIIrd Department of Pediatrics, Medical Academy, Biaystok, Poland.

2 Department of Physiopathology of Respiratory System, National Research Institute for Tuberculosis and Lung Diseases, Rabka Branch, Poland.

Grant sponsor: Ministry of Science and Higher Education; Grant number: 2PO5E 09027. *Correspondence to: Waldemar Tomalak, PhD, Department of Physiopathology of Respiratory System, National Research Institute for Tuberculosis and Lung Diseases, Rabka Branch, 34-700 Rabka, J. Rudnik str. 3b, Poland. E-mail: wtomalak@zpigichp.edu.pl Received 20 March 2008; Revised 9 July 2008; Accepted 16 July 2008. DOI 10.1002/ppul.20926 Published online 5 November 2008 in Wiley InterScience (www.interscience.wiley.com).

1194
10

Nowowiejska et al.
TABLE 1 Age, Height, and Weight of the Children Participating in the Study Boys Number Age (years) Range Mean SD Height (cm) Range Mean SD Weight (kg) Range Mean SD 278 3.218.9 10.6 4.4 98193 144.8 25.0 1493 41.6 19.8 Girls 348 3.118.9 10.9 4.5 95185 141.9 21.6 1486 38.5 16.0

follows: familial asthma, diagnosed chronic respiratory illness, obesity, premature birth, smoking (active or passive). Included children taking part in the study were examined by a physician in the day of measurements. Children with respiratory tract infection in preceding 4 weeks were also excluded. Among the children considered healthythree 3-year old and thee 4-year old refused to perform measurements. Twenty children 3-year old, ve 4-year old, and two 5-year old were not able to complete the protocol. In the oldest group of children (1519) seven refused to take part despite the parents agreement. In total626 (all of Caucasian descent) children were successfully examined. The overall success rate was 96%. The biometric characteristics of the group is given in Table 1. Figure 1 shows the age distribution for the boys and girls. The measurements were made in the sitting position with noseclip on, using IOS setup by Jaeger. The pneumotachograph has been calibrated each day prior to the measurements with a 3-L syringe, and the validity of IOS calibration was tested every time against reference impedance of 0.2 kPa/L/sec supplied by manufacturer. During the measurements the cheeks were supported by hands of investigators (for younger children), or by the children themselves. The measurements lasted for 45 sec, during which the children were asked to breathe tidally. Then, the procedure was repeated to obtain two sets of data

which did not differ more than 10% between each other. In a majority of children two measurements were needed (a maximum of four measurements were necessary in four children). For analysis mean values from the measurements were taken. IOS measurements bring resistances (R) and reactances (X) at 5, 10, 15, 20, 25, and 35 Hz and also the so-called resonant frequency (F)at which reactance X 0. Data sets were analyzed for boys and girls separately and for both sexes together. First, a multiple linear analysis using age, weight, body height and their combinations was performed to nd out the best predictor for IOS derived parameters. Then according to the results of multiple

Boys
60 60

Girls

50

40

Frequency

Frequency 5 10 Age [yrs] 15

30

20

10

10

20

30

40

50

10 Age [yrs]

15

Fig. 1. Histogram showing the distribution of the children with respect to age.

Pediatric Pulmonology

Reference Values for IOS for Children and Adolescents

Boys
1.6

1195

R [kPa/l/s]

<= 110 cm 110 cm 150 cm > 150 cm

0.8

1.2

<= 110 cm 110 cm 150 cm > 150 cm 5 10 15 F [Hz] 20 25 35 5 10 15 F [Hz] 20 25 35

0.0

0.4

Girls
1.6

R [kPa/l/s]

<= 110 cm 110 cm 150 cm > 150 cm X [kPa/l/s] 0.5 0.1 0.1 0.3 0.5 0.5 X [kPa/l/s] 0.1 0.1 0.3 0.5 <= 110 cm 110 cm 150 cm > 150 cm 10 15 20 25 35

0.0

0.4

0.8

1.2

10

15

20

25

35

F [Hz]

F [Hz]

Fig. 2. Resistances and reactances at different frequencies in three different height ranges for boys and girls.

linear regression analysisfour models using height as independent variable have been analyzed: linear, exponential, logarithmic and multiplicative (for reactances: linear and logarithmic). The equations were constructed basing on the best correlationthat is, the model was chosen according to the highest correlation coefcient between independent and dependent variables. The analyses have been made using R (TM) statistical package and Microsoft EXCEL.

RESULTS

Figure 2 presents frequency course of resistances and reactances in boys and girls in three different height ranges (<110 cm; 110150 cm; and >150 cm). The resistances decrease slightly with frequency, while reactances increase with increasing frequency. The analysis showed, that for all variables except X25 in boys and girls height was the best predictor. Further

TABLE 2 Coefcients of the Regression Equations for Boys and Girls Boys Parameter R5 R10 R15 R20 R25 R35 X5 X10 X15 X20 X25 X35 Fn a 0.0171 0.0177 0.0162 0.0149 0.0138 0.0134 0.0035 0.0021 0.0020 0.0014 0.0039 0.0016 0.0101 b 1.855 1.765 1.475 1.209 1.023 1.059 0.699 0.377 0.317 0.161 0.063 0.482 4.164 r 0.881 0.89 0.878 0.854 0.846 0.855 0.850 0.706 0.696 0.605 0.223 0.680 0.739 RSD 0.2305 0.2260 0.2208 0.2272 0.2173 0.2034 0.0540 0.0531 0.0522 0.0457 0.0422 0.0436 0.2299 a 0.0167 0.0172 0.0160 0.0145 0.0136 0.0121 0.0040 0.0023 0.0022 0.0017 0.0008 0.0015 0.0109 b 1.784 1.694 1.480 1.201 1.025 0.899 0.762 0.383 0.330 0.199 0.001 0.461 4.240 Girls r 0.866 0.874 0.866 0.830 0.815 0.813 0.845 0.676 0.677 0.610 0.374 0.599 0.718 RSD 0.2085 0.2070 0.2006 0.2114 0.2088 0.1883 0.0543 0.0532 0.0524 0.0476 0.0434 0.0429 0.2292 a 0.0169 0.0174 0.0161 0.0148 0.0138 0.0128 0.0037 0.0022 0.0021 0.0015 0.0006 0.0015 0.0101 Both sexes combined b 1.818 1.729 1.483 1.217 1.032 0.987 0.728 0.376 0.319 0.178 0.030 0.471 4.164 r 0.873 0.883 0.872 0.842 0.832 0.836 0.845 0.684 0.677 0.603 0.305 0.640 0.739 RSD 0.2185 0.2155 0.2010 0.2196 0.2130 0.1960 0.0545 0.0536 0.0529 0.0469 0.0432 0.0432 0.2299

R5, R10 . . . X35, kPa/L/sec; Fn, Hz. Models: E, exponential (X eaHb); L, linear (X a b H); a,b, model coefcients; r, correlation coefcient; RSD, residual standard deviation.

Pediatric Pulmonology

1196

Nowowiejska et al.
1.6 0 0.7 0.6 0.5 0.4 0.3 0.2 0.1

1.4

1.0

1.2

X5 [kPa*l/s ]

R5 [kPa*l/s]

0.6

0.2

0.4

0.0

100

120

140

160

180

200

100

120

140

160

180

200

10

15

.0

Fn [Hz]

0.8

20

25

100

120

140

160

180

200

height [cm]

height [cm]

height [cm]

Fig. 3. An example of model t to individual points in girls for R5, X5, and F.

analysis was performed using body height as an independent variable. Table 2 presents equations for resistances, reactances and resonant frequency in boys and girls and for both genders together. As expected, resistances at all frequencies show negative correlation to height, while reactances (except X35) increase with increasing height. Resonant frequency F decrease with height both in boys and girls. Figure 3 presents examples of the t of exponential model for resistance at 5 Hz, linear model for X5 and exponential model for F.
R5 boys
1.6

DISCUSSION

Our study performed on a large group of healthy children shows well known features: height is a main predictor of respiratory function measured with IOS as in the other studies.79 Resistances and resonant frequency decrease with increasing heights, while reactances increase. Table 2 presents coefcients of equations for boys, girls and the whole group, because some authors published reference equations without gender differentiation.8 As ours cover the age range from 3 to 19 years, separate sets of equations are also calculated
R5 girls
1.6

1.4

1.2

R5 [kPa*l/s]

1.0

0.8

0.6

0.4

0.2

R5 [kPa*l/s]

0.0

100

120

140 height [cm]

160

180

200

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Present study Malmberg 2002(6) Frei 2004(7)

Present study Malmberg 2002(6) Frei 2004(7)

100

120

140 height [cm]

160

180

200

Fig. 4. A comparison of the relationship of R5 and height in the present study (solid line) to the values of Malmberg (6, dashed line) and Frei (7, dotted line).

Pediatric Pulmonology

Reference Values for IOS for Children and Adolescents

1197

for boys and girls as in case of spirometry or body plethysmography. Table 2 contains also residual standard deviations (RSDs) values which are necessary for calculating the so called z-scores. z-score combines the percent predicted and the variability into a single number (z (measured value predicted value)/RSD) and expresses measured values in terms of RSD. The presentation of a result in terms of z scores is recommended and facilitates the interpretation of the results, that is, determination whether the result falls outside the dened lower or upper limit of normal.11 A comparison of our values of calculated R5 for boys and girls with those of Malmberg and Frei is shown on Figure 4. Our values are very close to previously published ones, however their main advantage is that they cover ages from 3 to 19 and heights from 95 to 193 cm. IOS seems to be the only technique allowing measurements of respiratory mechanics with equal ease in preschool children as well as in schoolchildren, adolescents and adults. This makes possible to compare different populations, or, what is even more potentially interestingto study the progression of lung function changes in health and disease starting from the age of 2. This feature is clearly seen on Figure 1 on reactance dependency on frequency. For younger (shorter) children (<110 cm of height) there is a marked increase of X above the resonant frequencywhich is related to different elastic properties (central airways shunt or inhomogeneities) of the respiratory system comparing to older (taller) children and adolescents. The advantage of IOS over spirometry is that oscillatory indices are measured during tidal breathing, not during forced expiratory maneuver. FEV1 is reduced if there is increased resistance to ow in the airways, and resistive properties of the airways (reected by R5 and R20) are measured by IOS. The usefulness of IOS resistances has been demonstrated in preschool children2,4,6,8 and in older ones. In a previous work from our laboratory3 it has been shown that in the group of 334 children aged 518 years the correlation between spirometric FEV1 and R5 was very strong with the value of r 0.66, for R20 r reached value of 0.54. Also reactances might be useful in evaluating children. We have shown12 greater specicity of reactances compared to resistances in assessing bronchial obstruction in children. Recent work of Gangell et al.13 also shown decreased X values in terms of z scores in young CF symptomatic children, compared to asymptomatic ones. IOS has also been tested in adultsin patients with COPD to asses its potential in assessing bronchodilatation during clinical trials14 and in adult patients with asthma in

assessing bronchodilatatory effect of ipratropium bromide.15 As it requires only passive cooperationit can be used also in patients with severe respiratory disorders, who are unable to perform spirometry.
REFERENCES
1. Vogel J, Smidt U. Impulse oscillometry. pmi Verlagsgruppe, Frankfurt 1994. 2. Klug B, Bisgaard H. Measurement of lung function in awake 2-4year-old asthmatic children during methacholine challenge and acute asthma: a comparison of the impulse oscillation technique, the interrupter technique, and transcutaneous measurement of oxygen versus whole-body plethysmography. Pediatr Pulmonol 1996;21:290300. ski J, Pawlik J, Latawiec W, Pogorzelski A. 3. Tomalak W, Radlin Impulse oscillometry vs body plethysmography in assessing respiratory resistance in children. Pediatr Pulmonol 2006;41:50 54. 4. Vink GR, Arets HG, van der Laag J, van der Ent CK. Impulse oscillometry: a measure for airway obstruction. Pediatr Pulmonol 2003;35:214219. 5. Arets HGM, van der Ent CK. Measurement of airway mechanics in spontaneously breathing young children. Pediatr Respir Rev 2004;5:7784. 6. Beydon N, Davis SD, Lombardi E, Allen JL, Arets HGM, Aurora P, Bisgaard H, Davis GM, Ducharme FM, Eigen H, et al. An Ofcial American Thoracic Society/European Respiratory Society Statement: pulmonary function testing in preschool children. Am J Respir Crit Care Med 2007;175:13041345. 7. Malmberg LP, Pelkonen A, Poussa T, Pohianpalo A, Haahtela T, Turpeinen M. Determinants of respiratory system input impedance and bronchodilator response in healthy Finnish preschool children. Clin Physiol Funct Imaging 2002;22:6471. 8. Frei J, Jutla J, Kramer G, Hatzakis GE, Ducharme FM, Davis GM. Impulse oscillometry: reference values In children 100 to 150 cm in height and 3 to 10 years of age. Chest 2005;128:1266 1273. 9. Klug B, Bisgaard H. Specic airway resistance, interrupter resistance and respiratory impedance in children aged 27 years. Pediatr Pulmonol 1998;25:322331. 10. Taussig LM, Chernick V, Wood R, Farell P, Mellins R. Standardisation of lung function testing in children. J Pediatr 97:668676. 11. Stanojevic S, Wade A, Stocks J, Hankinson J, Coates AL, Pan H, Rosenthal M, Corey M, Lebecque P, Cole TJ. Reference ranges for spirometry across all ages. A new approach. Am J Respir Crit Care Med 2008;177:253260. ski J, Pawlik J, Latawiec W. Sensitivity and 12. Tomalak W, Radlin specicity of the resistances and reactances obtained with impulse oscillometry in children. Eur Respir J 2006;28:706s. 13. Gangell CL, Horak F, Jr., Patterson HJ, Sly PD, Stick SM, Hall GL. Respiratory impedance in children with cystic brosis using forced oscillations in clinic. Eur Respir J 2007;30:892897. 14. Borrill ZL, Houghton CM, Woodcock AA, Vestbo J, Singh D. Measuring bronchodilatation in COPD clinical trials. Br J Clin Pharmacol 2004;59:379384. 15. Houghton CM, Woodcock AA, Singh D. A comparison of plethysmography, spirometry and oscillometry for assessing the pulmonary effects of inhaled ipratropium bromide in healthy subjects and patients with asthma. Br J Clin Pharmacol 2004;59: 152159.

Pediatric Pulmonology