Pediatric Pulmonology 41:50–54 (2006)
Original Articles
Impulse Oscillometry vs. Body Plethysmography in Assessing Respiratory Resistance in Children Waldemar Tomalak, PhD,* Jakub Radlin´ski, PhD, Jacek Pawlik, MD, PhD, Wojciech Latawier, MSc, and Andrzej Pogorzelski, MD, PhD Summary. In 334 children aged 5–18 years, we compared the results of plethysmographic measurements of airway resistance (Raw) with oscillometric (impulse oscillometry; IOS) assessment of respiratory properties of the respiratory system (resistance (R) at 5, 20, and 35 Hz). All three resistances correlated significantly with plethysmographic Raw, and the strongest correlation was seen for R5 (r ¼ 0.64). R5, R20, and R35 were significantly greater than Raw in the whole group. In the group of children with obstruction (FEV1%FVC below lower limit of normal), R5 was still greater than Raw, while R20 and R35 were not. The Bland-Altman analysis comparing plethysmographic measurements with oscillometric R5 revealed a significant difference between Raw and R5 in the whole group, which disappeared in the group of obstructed patients. Oscillometric assessment of resistive properties of the respiratory system of the lung requires less patient cooperation than does plethysmography. As the results of measurements using oscillometric R5 are similar to those obtained by plethysmography, IOS may be useful in diagnosing children with obstructive respiratory diseases. Pediatr Pulmonol. 2006; 41:50–54. ß 2005 Wiley-Liss, Inc.
Key words: lung function testing; plethysmography; forced oscillation technique; impulse oscillometry; children.
INTRODUCTION
Body plethysmography as a method for measuring airway resistance, and the forced oscillation technique (FOT) to evaluate the resistive properties of the respiratory system, have the same father.1,2 Both were introduced in the mid-1950s. However, the further development of the two techniques and their popularity have been quite different. While body plethysmography is widely used and its diagnostic value is well-established, this is not the case with the forced oscillation technique. Several reasons seem to be responsible for that, including the very complicated theoretical background, the sophisticated analysis of data collected, and the measurement of respiratory system properties for frequencies far beyond the range of normal breathing, causing problems with interpretations. In 1993, a modification of the forced oscillation technique, impulse oscillometry (IOS), was introduced by Jaeger as user friendly, commercialized apparatus offering measurements of respiratory system resistance (Rrs) and reactance (Xrs) at a number of frequencies.3 The approach of IOS differs from original FOT idea by applying a rectangular pressure impulse rather than pseudorandom ß 2005 Wiley-Liss, Inc.
pressure wave (being the sum of several sinusoidal pressure waves), but offers the same advantages, i.e., minimal requirements for the cooperation of the patient, rapid, easy and reproducible measurements. This makes FOT or IOS measurements especially interesting in pediatric population. Several works have been published showing the usefulness of IOS in different pathological conditions and in different groups of patients, including preschool children,4,5 cystic fibrosis patients,6 and asthmatic children.7 Department of Physiopathology of the Respiratory System, National Research Institute for Tuberculosis and Lung Diseases, Rabka Branch, Rabka, Poland. *Correspondence to: W. Tomalak, National Institute for Tuberculosis and Lung Diseases, Rabka Branch, J. Rudnik Str. 3, 34-700 Rabka, Poland. E-mail:
[email protected] Received 12 May 2005; Accepted 18 May 2005. DOI 10.1002/ppul.20310 Published online 5 October 2005 in Wiley InterScience (www.interscience.wiley.com).
IOS vs. Body Plethysmography
There are still few studies comparing different techniques of measuring the resistive properties of the respiratory system. In 2001, Hellinckx et al.8 published a comparison of data collected in 49 patients aged 8–70 years (mean, 24 19 years). They showed an agreement between IOS resistance at 5 Hz and plethysmographic airway resistance (Raw), with a value of R2 ¼ 0.25. As IOS measurements are less demanding, they could potentially replace plethysmographic assessment of resistive properties of the respiratory system. The aim of our study was to assess the comparability of the two techniques in children. MATERIALS AND METHODS
We analyzed the results of tests in 337 children aged 5–18 years. There were 165 girls and 172 boys. All children were treated at our institute because of chronic respiratory diseases. There were 197 children with asthma and other allergic diseases (e.g., allergic rhinitis, atopic dermatitis), 114 with cystic fibrosis, and several cases of bronchiectasis and lung fibrosis. Children were examined in a stable period, free of exacerbations. Table 1 shows the biometric characteristics of the group and the basic spirometric indices. All children underwent measurements of plethysmographic airway resistance (Raw) and impulse oscillometry in random order. To avoid any possible alteration of forced expiratory maneuvers on bronchial tone, spirometry was performed as the last measurement in the sequence. Measurements were made using Jaeger’s MasterLab1 system. Prior to measurements, the system was calibrated according to the protocol, and oscillometric measurements were validated with a standard impedance of 2 kPa provided by the manufacturer. IOS parameters (resistance and reactance at 5, 10, 15, 20, 25, and 35 Hz, noted as R5, X5, etc.) were obtained from measurements lasting 45 sec, during spontaneous breathing, while patients supported their cheeks with their hands. At least three reproducible trials were done according to the procedure described by the manufacturer. Raw was measured during rhythmic breathing, with the patient’s cheeks supported, and the value of Raw was calculated from the slope of pressureflow relationship between 0.5 l/sec. Flow-volume loops
were recorded, according to American Thoracic Society guidelines.10 To compare both techniques, we chose plethysmographic Raw and oscillatory parameters R5, R20, and R35. Statistical analyses consisted of calculating means and standard deviations, linear regressions, and dispersions from the line of identity, using the approach of Bland and Altman.11 Analyses were performed on the whole group, and in subgroups divided according to spirometric criteria for obstruction, i.e., FEV1%FVC being below the lower limit of a normal (LLN) value.10 There were 76 subjects (42 girls and 24 boys) with FEV1%FVC below LLN, with mean FEV1% predicted 55.7 23.9 (range, 19.0–105.0), and 261 with FEV1%FVC within normal limits (123 girls and 138 boys; mean FEV1% predicted, 98.6 22.7; range, 26.0–155.0). RESULTS
In the studied group, we observed a very wide range of resistances for both techniques. Table 2 shows the values of different indices in the whole group and both subgroups. The mean values of R5, R20, and R35 were higher than plethysmographic Raw in the whole group (P < 0.001 by paired t-test) and in the nonobstructed group. In the group with obstruction, R5 was statistically significantly higher, while R20 and R35 were significantly lower than plethysmographic values. All resistances correlated significantly with FEV1. Correlation coefficients were 0.62 for Raw, and 0.66, 0.54, and 0.47 for R5, R20, and R35, respectively. The correlation of oscillometric resistances and plethysmographic Raw was also significant. The highest r value was for R5 (r ¼ 0.64), while correlation coefficients for R20 and R35 were much lower (0.39 and 0.32, respectively). Figures 1–3 show the correlations between plethysmographic Raw and oscillometric resistances for the whole group. The Bland-Altman analysis revealed substantial difference between the two techniques of measuring resistance. At 5 Hz (Fig. 4), the correlation coefficient for the relationship between the difference (R5 Raw) and the mean was low, but significant. This effect became nonsignificant
TABLE 1— Biometric Characteristics of Children Participating in Study1
N Age (years) Height (cm) FEV1%pred9 FVC%pred9 1
Boys
Girls
Total
172 11.5 3.3 (5–18) 145.8 20.4 (103–194) 94.6 26.9 (21–156) 99.8 22.8 (32–138)
165 12.5 3.4 (6–18) 147.1 15.9 (109–181) 83.0 30.3 (19–142) 88.0 26.2 (27–142)
337 12.0 3.4 (5–18) 146.4 18.3 (103–194) 88.9 29.1 (19–155) 94.1 25.2 (27–142)
Values are means SD and (ranges).
51
Raw, R5, R20, and R35: kPa/l/sec; values are means SD and (ranges). P values were calculated with Student’s t-test for unpaired samples for comparison between obstructed and nonobstructed groups and with t-test for paired samples for comparison between Raw and oscillatory R. *P < 0.001.
1
0.42 0.36 (0.05–2.22) 0.71 0.31* (0.17–2.01) 0.49 0.16* (0.12; 0.93) 0.49 0.15* (0.10–0.95) 0.32 0.28 (0.05–2.22) 0.66 0.28* (0.17–1.98) 0.48 0.16* (0.12–0.93) 0.46 0.12* (0.19–0.70) 0.76 0.42 (0.15–1.93), P < 0.001 0.92 0.35* (0.45–2.01), P < 0.001 0.54 0.14* (0.31–0.92), P ¼ 0.004 0.52 0.14* (0.29–0.95), P ¼ 0.06 Total Normal FEV1%FVC (% predicted) FEV1%FVC (% predicted) below LLN
Raw
R5
R20
R35
Tomalak et al.
TABLE 2— Resistances in Whole Group of Children and in Subgroups1
52
Fig. 1. Scatterplot of resistance (R) at 5 Hz and plethysmographic Raw. Solid line, regression line; dashed line, line of identity.
in children with FEV1%FVC below LLN (P ¼ 0.06; Fig. 5). DISCUSSION
Our results show that for 5, 20, and 35 Hz, oscillometric resistances were higher than plethysmographic Raw. The resistances measured by the impulse oscillometric method at all frequencies correlated significantly with plethysmographically assessed airways resistance. The correlation was the strongest for 5 Hz. Assessment of resistive properties of the respiratory system in children is not easy. Active cooperation during measurements, required by classic methods such as plethysmography, is a limiting factor. Oscillometric measurements require less cooperation, and the results are obtained during spontaneous breathing. Thus this technique is especially suitable for children, and was used even in 2-year-olds.5 Moreover, reference values for the technique were evaluated,12,13 but there was a lack of comparative studies. The results of Helinckx et al.8 revealed that the correlation between Raw and R5 was fairly good, with an R2 value of 0.25, and with the values of R5 slightly greater than Raw values.
Fig. 2. Scatterplot of R20 and Raw. Solid line, regression line; dashed line, line of identity.
IOS vs. Body Plethysmography
Fig. 3. Scatterplot of R35 and Raw. Solid line, regression line; dashed line, line of identity.
Such results are expected, since oscillometric resistance reflects total respiratory system resistance. R5 values were greater than Raw (mean difference, 0.29 0.29 kPa; P < 0.001 by unpaired t-test). Also, mean R20 and mean R35 were greater than plethysmographic Raw. Interestingly, in a subgroup of patients with FEV1%FVC below the lower limit of normal, R5 values were greater than plethysmographic resistance (mean difference, 0.16 0.36 kPa/l/sec), but mean values of R20 and R35 were lower (0.22 0.39 and 0.24 0.40 kPa/l/sec). For R20, the difference between groups was still significant (P ¼ 0.004), while for R35 it was not (P ¼ 0.06). Resistance at 5 Hz is defined by the manufacturer (Jaeger) as total respiratory resistance,3 and data from clinical studies show that respiratory system resistance at low frequencies allows the best discrimination between healthy individuals and those with various obstructive disorders.14 In addition, the results published by Schmekel and Smith15 showed that during provocation with cold air, in 20 asthmatic subjects, R5 increased significantly from a baseline value of 0.37 0.09 kPa/l/sec to 0.58 0.22 kPa/l/sec (P < 0.001), while R35 did not (from 0.35 0.06 kPa/l/sec to 0.36 0.08 kPa/l/sec). The negative frequency dependence of resistance is a common feature observed in forced oscillatory measurements. It may be amplified by the presence of airway obstruction, or upper airway motion effect16 (minimized in our study by
Fig. 4. Bland-Altman graph for comparison of R5 and Raw in whole group (n ¼ 337).
53
Fig. 5. Bland-Altman graph for comparison of R5 and Raw in group of children with obstruction (FEV1%FVC < LLN value).
supporting the cheeks). This may explain why the correlation becomes weaker with increasing frequency, and why the difference between Raw and oscillometric resistances at 20 and 35 Hz is lower than at 5 Hz. Indeed, at 35 Hz, the difference between oscillometric R and Raw was not significant. Thus the diagnostic value of oscillometric resistances at higher frequencies seems to be weak. The question arises of whether plethysmographic Raw and oscillometric R5 are interchangeable. A Bland-Altman analysis (Fig. 4) showed that there exists substantial difference between the two techniques. From the theoretical point of view, Raw obviously reflects the resistance of the bronchial tree, while oscillometric measurements reflect total respiratory system resistance (Fig. 4), so the two methods are not directly comparable. However, the difference between them decreases with increasing resistance. An analysis in a subgroup of obstructed subjects showed that the difference between the two methods becomes nonsignificant (Fig. 5). This fact favors IOS measurements, which are easier than plethysmography. Although those results suggest an interchangeability of the two measurements, there are some limiting factors. First of all, there is no information from the oscillometric measurement concerning the quality of results. For classic FOT measurements, a measure of signal-to-noise ratio (a coherence function) is calculated for each frequency of interest, which is not the case with IOS measurements. The FOT excitation signal contains only several harmonics, while the frequency content of IOS excitation signals is not limited. Moreover, the lower the frequency, the more problems might be encountered due to the influence of spontaneous breathing. This may lead to a lack of reliable measurements at lower frequencies. Secondly, the IOS apparatus has not been evaluated critically according to published FOT recommendations,17–19 formulated on the basis of calibration, input signals and frequencies, data processing, and acceptance and validation of measurements. Such studies should be undertaken to assure the quality of measurements with IOS.
54
Tomalak et al.
CONCLUSIONS
Our study, performed on a large group of children, shows great potential for measurements of resistive properties of the respiratory system with impulse oscillometry. The technique appears able to differentiate patients who have obstructive airway disease. As the requirements for patients’ cooperation are less stringent than for plethysmography, this makes the technique more favorable for examining children. REFERENCES 1. DuBois AB, Botelho SY, Comroe JH. A new method for measuring airway resistance in man using body plethysmograph: values in normal subjects and in patients with respiratory disease. J Clin Invest 1954;35:327–335. 2. DuBois AB, Brody AW, Lewis DH, Burgess DF. Oscillation mechanics of lung and chest in man. J Appl Physiol 1956;8:587– 594. 3. Vogel J, Smidt U. Impulse oscillometry. Frankfurt: PMI Verlagsgruppe; 1994. 4. Vink GR, Arets HG, van der Laag J, van der Ent CK. Impulse oscillometry: a measure for airway obstruction. Pediatr Pulmonol 2003;35:214–219. 5. Klug B, Bisgaard H. Measurement of lung function in awake 2– 4-year-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;21:290–300. 6. Villa Asensi JR, de Miguel Diez J, Angelo Vecchi A, Salcedo Posadas A, Neira Rodriguez MA, Sequeiros Gonzalez A. Assessment of lung function using forced mpulse oscillometry in cystic fibrosis patients. Arch Bronconeumol 1998;34:520–524. 7. Vink GR, Arets HG, van der Laag J, van der Ent CK. Impulse oscillometry: a measure for airway obstruction. Pediatr Pulmonol 2003;35:214–219.
8. Hellinckx J, Cauberghs M, De Boeck K, Demedts M. Evaluation of impulse oscillation system: comparison with forced oscillation technique and body plethysmography. Eur Respir J 2001;18:564– 570. 9. Willim G, Haluszka J, Tomalak W, Kapustianyk I. 1990. Computer assisted evaluation of the state of respiratory system. Rabka: Institute for Mother and Child; 1991. p 1–31. 10. American Thoracic Society. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991;144:1202–1218. 11. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurements. Lancet 1986;81;(8476):307–310. 12. Klug B, Bisgaard H. Specific airway resistance, interrupter resistance, and respiratory impedance in healthy children aged 2–7 years. Pediatr Pulmonol 1998;25:322–331. 13. 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: 64–71. 14. Duiverman EJ, Neijens HJ, van der Snee V, Smaalen M, Kerrebijn KF. Comparison of different indices from dose-response curves to inhaled methacholine determined by multiple frequency oscillometry and forced expiratory flow-volume curves. Bull Eur Physiopathol Respir 1986;22:433–436. 15. Schmekel B, Smith HJ. The diagnostic capacity of forced oscillation and forced expiration techniques in identyfying asthma by isocapnic hyperpnoea of cold air. Eur Respir J 1997;10:2243– 2249. 16. Peslin R, Duvivier C, Gallina C, et al. Upper airway artifact in respiratory impedance measurements. Am Rev Respir Dis 1985; 132:712–714. 17. Zwart A, Peslin R. Mechanical respiratory impedance: the forced oscillation method. Eur Respir Rev 1991;1:1–237. 18. Zwart A, van de Woestijne KP. Mechanical respiratory impedance by forced oscillation. Eur Respir Rev 1994;19:115–237. 19. Oostveen E, MacLeod D, Lorino H, et al. The forced oscillation technique in clinical practice: methodology, recommendations and future developments. Eur Respir J 2003;22:1026–1041.
