INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: airway and respiratory tissue mechanics; wheezy infants; low-frequency forced oscillations; forced expiratory volumes

Until recently the available knowledge on the normal physiologic development of the airways and lung tissue was limited, and as a result, the understanding of the development of disease and responses to treatments has suffered. Epidemiologic studies have identified at least two wheezing phenotypes in infancy; those infants who wheeze in the first 3 yr of life but do not wheeze or have asthma later in childhood (transient wheeze), and those infants whose wheeze persists and who develop asthma by the age of 6 (persistent wheeze) (1). Those infants with transient wheeze had diminished premorbid maximal flows at FRC (maxFRC) and may represent infants with congenitally smaller airways (1). However, this conclusion is based on scant data obtained with a technique that does not measure airway function directly. Infants with persistent wheeze have initial maxFRC values similar to those infants without wheeze, but decreased lung function at 6 yr of age. These infants also had more frequent symptoms and were more likely to have had rhinitis, eczema, maternal asthma, and elevated serum immunoglobulin E (IgE) at 9 mo than the group of transient wheezers (1).

Clinically, when faced with a wheezy infant, it has not been possible to determine whether that infant is likely to have transient or persistent wheeze. Using the raised volume rapid thoracic compression (RVRTC) technique, Turner and coworkers (2) reported decreased forced expiratory volumes (FEV_t) from a raised lung volume in wheezy compared with healthy infants. However, many studies have failed to detect differences between infants with recurrent wheeze and healthy infants, especially when studies are performed during symptom-free intervals (3). A potential reason for this failure is that many of the techniques are unable to provide a separate assessment of the airway and respiratory tissue mechanics and hence cannot definitively identify the site of the changes in lung function. The low-frequency forced oscillation technique (LFOT) allows the measurement of the respiratory system impedance (Zrs) at frequencies including the spontaneous breathing rate (< 2 Hz). Sly and coworkers (6) demonstrated that the LFOT could be successfully applied to spontaneously breathing infants and low-frequency Zrs spectra obtained. These investigators demonstrated that a four-parameter model (7) could be applied to the Zrs data and that the simultaneous determination of airway resistance (Raw) and inertance (Iaw) and tissue mechanics (damping [G] and elastance [H]) could be obtained. Recent studies in infants with the LFOT have characterized the volume dependence of the airway and respiratory tissue mechanics (8) and the response to inhaled salbutamol (9) and methacholine (10). In a previous study, we described the relationship between length and lung function using the LFOT in a cross-sectional study of healthy infants (11). The present investigation used the LFOT and RVRTC technique to determine the airway and respiratory tissue mechanics and FEV in asymptomatic infants with recurrent wheeze and compared these values with those previously obtained in healthy infants (11).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Asymptomatic infants with a history of recurrent wheeze, who had been free of symptoms for at least 4 wk before testing, were examined. Recurrent wheeze was defined as three or more episodes of parentally reported wheeze in the past 12 mo of life (1). Infants ranged between 6 and 32 mo of age and 67 to 89 cm in length (Table 1). The infants were sedated with an oral dose of chloral hydrate (70 to 100 mg/kg) and studied in the supine position with the head supported in the midline and the neck slightly extended. Zrs spectra and FEV_t were obtained in 22 and 16 infants, respectively. Data from both techniques were available in 14 infants. The human ethics committee of Princess Margaret Hospital approved the study. Parents gave written consent and were generally present during the study.

Measurement Apparatus and Analysis

LFOT. Zrs was measured with the LFOT previously applied in infant studies (6, 8). The infant's lungs were raised three times to a transrespiratory pressure (Prs) of 20 cm H₂O. After the final inflation, the airway opening was occluded, inducing the Hering-Breuer reflex. During the resulting end-inspiratory pause, a multifrequency forcing signal, between 0.5 and 20 Hz, was applied to the airway opening by a loud speaker, via a soft-rimmed face mask. Five to 10 Zrs spectra were collected from each infant. Raw, Iaw, and coefficients of G and H were estimated from the individual Zrs data by model fitting as described previously (7).

RVRTC. Forced expiratory data were obtained over an extended volume range using the technique described by Hayden and coworkers (12). The infant's lungs were inflated three times to a Prs of 20 cm H₂O with passive deflations between each inflation. Using a jacket connected to a positive pressure reservoir, a standardized compression force was applied to the thorax and abdomen after the third inflation. Three to five forced expiratory maneuvers were performed. Forced expiratory flows recorded and forced volume-time curves and FEV_0.5 data produced.

Statistics

LFOT airway and tissue parameters and FEV_0.5 were averaged and compared with the regression equations based on length derived in a previously reported cross-sectional population of normal infants (11). Standard variants (Z scores) were calculated as the difference between measured and predicted parameter values, divided by the residual standard deviation derived from the regression equations. To determine whether lung function of the wheezy infants differed significantly from our normal population, a t test was used to test whether the Z scores were significantly shifted from zero. Z scores were not calculated for infants whose body lengths exceeded that of the previously reported healthy population (11), and thus were calculated for 20 and 13 infants for LFOT data and FEV_0.5 respectively. As the definition of recurrent wheeze was three or more episodes of wheeze in the last 12 mo of life and a number of infants (n = 11) were younger than 1 yr of age, a separate analysis of infants younger than 1 yr and older than 1 yr was carried out using a two-tailed t test. Significance was accepted at the p < 0.05 level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory system H was significantly elevated in infants with recurrent wheeze, when compared with our normal population. There were no significant differences in Raw, G, or in FEV_0.5 between the group of wheezy infants and the normal infants reported previously (11). Figures 1A, 1B, 1C, and 1D show the mean values of FEV_0.5, Raw, H, and G, respectively, for each infant, as functions of length. Because Iaw was found to have a relatively minor influence on Zrs in the low-frequency range, its dependence on length was not established in our previous study in normal infants (11), and thus, this parameter is not reported. For the group as a whole (i.e., all ages), Raw and FEV_0.5 in wheezy infants did not differ from our normal data (Z score; 0.14 ± 0.25 and 0.25 ± 0.25 for Raw and FEV_0.5, respectively), whereas H was significantly elevated and G exhibited a tendency to be elevated, with Z scores of 0.61 ± 0.20 for H (p = 0.007) and 0.41 ± 0.21 for G (p = 0.066) (Table 2).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Mean individual data for (A) FEV0.5, (B) Raw, (C ) H, (D) G in wheezy infants. The open circles correspond to those infants < 1 yr of age; the closed circles represent those infants > 1 yr. Those infants for whom Z scores were not calculated are shown as triangle symbols. The solid line represents the derived regression equation for healthy infants presented previously. The variations in data representing Z scores of 1 (long dash) and 2 (short dash) are also shown.

When the population of wheezy infants is considered as two age groups (i.e., those younger and older than 1 yr of age), significantly differing patterns of respiratory mechanical properties are seen. No significant differences in parameters were noted in recurrently wheezy infants younger than 1 yr of age, when compared with our normal data. In contrast, significant differences were noted in all parameters in those infants older than 1 yr (Table 2), with group mean Z scores of 0.58 ± 0.20 for Raw (p = 0.018), 0.79 ± 0.31 for G (p = 0.032), 1.06 ± 0.25 for H (p = 0.002), and 0.94 ± 0.22 for FEV_0.5 (p = 0.007).

We obtained both FEV_0.5 and LFOT parameters in 11 infants (five older than 1 yr of age), allowing comparisons to be drawn between the ability of the two techniques to discern normal or altered lung function. Infants were considered to have altered lung function, compared with normal, if the Z score was greater than +1 for LFOT parameters and less than 1 for FEV_0.5 (i.e., > 1 SD from regression line). Table 3 shows the number of infants with normal lung function for each lung function parameter and the subsequent number of infants who had altered lung function for the remaining parameters. Figures 2A-2D show mean individual lung function values for the wheezy infants overlying previously published mean individual lung function values of healthy infants (11). The tendency for lung function in wheezy infants to deviate from the normal range is more pronounced in older children.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Mean individual data for (A) FEV0.5, (B) Raw, (C ) H, (D) G in wheezy infants (closed circles) and healthy infants previously reported (open circles). The solid lines represent the regression equation for lung function as a function of changing body length and the upper and lower limits of the normal range (defined by Z scores of +2 and 2).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The aim of the present study was to investigate differences in airway and respiratory tissue mechanics and FEV between asymptomatic infants with a history of wheeze and healthy infants, using the LFOT and the RVRTC technique. Our results indicate that recurrently wheezy infants have increased pulmonary H when compared with a normal infant population. Younger infants (< 1 yr of age) did not exhibit altered lung function, whereas infants with wheeze persisting into the second year of life had increased airway and respiratory tissue impedances and decreased FEV.

Conflicting data on lung function in wheezy infants have been reported. Using the single breath technique, Dundas and coworkers (4) demonstrated small but significantly decreased weight-corrected respiratory system compliance (Crs) in wheezy infants compared with healthy control subjects. In contrast, two studies from the Perth group reported no differences in Crs between healthy and wheezy infants (3, 13). The reasons for these differences are not clear, because infants in these studies (3, 4, 13) were of a similar age (11 mo) and because Crs was determined using the same technique. It is possible that differences in population may explain the conflicting results. The same groups of researchers reported a similar pattern of contrast in respiratory system resistance (Rrs), with Dundas and coworkers (4) describing significant differences and Young and coworkers (13) finding no change between healthy and wheezy infants in a longitudinal study. Whereas Crs allows the quantification of the tissue elastic properties, Rrs will incorporate an airway component (Raw) and therefore leaves the potential changes in either the airways or respiratory tissues due to wheeze inseparable in Rrs. Two studies have examined Raw measured with the plethysmographic technique in healthy and wheezy infants. Gutkowski (5) observed no differences in predicted or specific Raw between wheezy and control infants, but did find increased airway responsiveness to inhaled carbachol among wheezy infants. Dundas and coworkers (4) reported significant changes in Raw obtained at end-expiration and initial-inspiration (i.e., FRC), but not end-inspiration or initial-expiration (i.e., increased lung volume) in wheezy infants. Measurements made at end-expiration or low lung volumes, where peripheral airway narrowing may occur, are more likely to detect increases in Raw due to altered airway tone than those made during mid- or end-inspiration. Because the values of Raw were calculated at half-tidal volume by Gutkowski (5), the difference in mean lung volume may explain the discrepancy between these studies. Frey and coworkers (14), using the high-frequency interrupter technique, demonstrated differences in the first antiresonant frequency between healthy and wheezy infants, and suggested that these differences may relate to airway wall compliance.

A number of investigators have used measurements of forced expiratory flows and volumes to investigate changes caused by disease in wheezy infants. Using the rapid thoracic compression technique, Stick and coworkers (3) reported significant differences in baseline maxFRC in wheezy infants when compared with control subjects, although histamine responsiveness was not significantly different between the two groups. Turner and coworkers (2) demonstrated significant differences in maxFRC and FEV_t between healthy and wheezy infants. In a longitudinal study, Young and coworkers (13) reported that lung function changes in wheezy infants were dependent upon the period of initial or subsequent wheeze. In confirmation of previous studies (1, 15, 16), these investigators (13) observed decreased premorbid absolute and size-corrected maxFRC in infants who subsequently wheezed, and these differences persisted to 6 and 12 mo of age. Similarly, the current study demonstrated that FEV_0.5 was decreased in infants in whom wheeze persisted into the second year of life only. Although measurements of maxFRC and FEV_t have shown differences in baseline lung function between infantile health and disease, the data are unable to provide separate information on the involvement of the airways and respiratory tissues: airway obstruction will lead to expiratory flow limitation, but the mechanical properties of the parenchyma will also play a role. Direct conclusions as to any alterations in airway or lung tissue mechanical properties cannot be made from measurements of forced expiratory flows as these measurements do not provide data specifically linked to structural components but rather reflect global changes in respiratory mechanical performance. However, the changes seen in FEV_0.5 in infants with recurrent wheeze in the present study more closely reflected the changes seen in lung parenchymal mechanics than in airway mechanics.

Changes that may occur in the mechanical properties of both airways and respiratory tissues with wheeze in infancy have not been systematically addressed. The advantage of the LFOT is its ability to simultaneously provide separate information on the airways and respiratory tissues. Several technical issues with our methodology deserve to be discussed. Our estimates of Raw include the resistance of the nose and the extrathoracic airways, as well as any Newtonian (frequency-independent) resistance component from the tissues. Data reported previously with the same technique demonstrated that the nasal impedance contributes approximately 45% to Raw and that this contribution is constant with length (17) (As the presence of an upper respiratory tract infection is likely to influence upper airway mechanics, we deliberately excluded infants from testing if they had any evidence of infection in the 4 wk before scheduled testing). The Newtonian component arising from the tissue contribution to Raw was found to be negligible in the case of the canine lung tissue (18), whereas we have previously shown in infants (19) that the chest wall contributes approximately 18% of the total Newtonian resistance. Consequently, the differences in Raw between the wheezy and normal infants, which one expects to reside within the intrathoracic airways, will be somewhat masked by the inclusion of the nasal impedance and the chest wall tissue component, which are most likely unrelated to wheeze.

We would also note that the estimates of G and H in the present study incorporate the chest wall and pulmonary tissue parameters. In a preliminary study on the pulmonary and chest wall components of Zrs (19), we found the contributions of the chest wall to the total G and H in healthy infants to be as large as 58% and 50%, respectively. Hence, the differences in pulmonary tissue parameters between the normal and wheezy infants are likely to be roughly twice as high as those derived from total respiratory impedance in the present study. Although the contribution of the chest wall to respiratory system resistance and elastance increases with age as the chest wall becomes relatively stiffer (20), it is unlikely that this increase with age would be systematically different in recurrently wheezy and healthy infants. One possible influence of the chest wall would arise from differences in body weight between the groups, causing alterations in H not related to wheeze. However, no significant differences in the group mean body length/ weight ratio in the range of 67 to 87 cm were noted between infants in the present study and those healthy infants reported previously (11) (7.2 ± 0.9 cm/kg and 7.3 ± 0.8 cm/kg for the wheezy and healthy infants, respectively: t test: p = 0.43).

The current study used Z scores derived from the regression equations previously reported in healthy infants (11) to illustrate differences between the wheezy infants and the healthy ones. This approach is likely to be valid provided that the study population is similar in length to the reference population. In the present investigation, we have excluded from the Z-score analysis those infants who do not fall within the range of body lengths originally published. Furthermore, by using Z scores, we have allowed for changes in both the intersubject and intrasubject variability with body length. Thus, we would suggest that the reported results reflect actual physiologic and mechanical differences between healthy and wheezy infants and not those in anthropometric properties.

It is also important to point out that there were no differences between the two groups in terms of impedance measurement and parameter estimation, because the measurement protocol, the applied oscillatory signal, and the criteria of model fitting were all identical. The average fitting errors (6.32 ± 0.2 [SD]% and 6.36 ± 0.3%, respectively) are not different in the healthy and wheezy populations, and the number and location of the Zrs data that coincided with the cardiac frequency and its harmonics and were excluded from the fitting are not statistically different in the two groups. The observation that the wheezy infants separate from the healthy control subjects more by tissue properties than Raw is rather unexpected, and the nature of the altered tissue mechanics is unclear. One may argue that inhomogeneities in the peripheral airways, which did not result in the elevation in the total Raw, may have contributed to the change in the tissue properties. Indeed, it has been demonstrated that moderate to severe increases in Raw, presumably resulting from inhomogeneous airway constriction, may cause an artifactual increase in G, i.e., a change unrelated to those in real tissue mechanical properties (21, 22).

Although we cannot discount the possibility of more marked airway inhomogeneities within the wheezy population, it is unlikely to explain the alterations seen in G and H. First, our measurements of Zrs are made at a Prs of 20 cm H₂O, and this raised lung volume will act to splint the airways open, reducing the likelihood of significant heterogeneous narrowing of bronchi that might have existed at FRC. Second, the previous investigations (21, 22) demonstrated a significant influence of inhomogeneities during moderate to severe constriction, manifesting as increases in Raw and G, and not mirrored by changes in H. In the present study, minor alterations in Raw were noted, and only in those infants older than 12 mo of age, whereas changes in H were in fact the most prominent. We therefore suggest that the changes seen in the respiratory system tissue mechanics (G and H) in the present study can be attributed to alterations in the intrinsic parenchymal properties of wheezy infants.

The mild difference in Raw between the two groups of infants is in accord with the observations by Dundas and coworkers (4) and Gutkowski (5), and is most probably related to the raised lung volume uniformizing all peripheral airway heterogeneities that might have developed at FRC to a higher degree in the wheezy infants than the normals. Because it has been shown that LFOT data can also be collected, although with a lower success rate, at end-expiration (8), Zrs measurements made at both FRC and high lung volume can be advocated for more detailed exploration of the respiratory mechanical alterations in wheeze.

In a subgroup of infants (n = 11), we were able to obtain measurements of lung function using both the LFOT and RVRTC technique (Table 3). We found that of those infants with normal tissue elastance (6 of 11), two had altered Raw, one of whom also had elevated G. None of these six infants had abnormal FEV_0.5 Z scores. In contrast, eight infants were determined to have normal FEV_0.5, half of whom had altered oscillatory mechanics. Although only limited conclusions can be drawn regarding the similarities and differences between the two techniques because of the small numbers of infants, there was a closer concordance between abnormal H and abnormal FEV_0.5 than between altered Raw or tissue damping and FEV_0.5. Although we stress that these data are not definitive, they should sound a cautionary note to those who interpret changes in FEV_0.5 as representing changes in airway mechanics in infants.

In conclusion, we have shown that the LFOT applied at a Prs of 20 cm H₂O is able to detect differences in respiratory mechanical parameters between asymptomatic infants with a history of recurrent wheeze and healthy infants. The mechanical properties of both the airways and respiratory tissues were found to be altered in wheezy infants older than a year, with the parameters representing parenchymal mechanics deviating further from normal than the airway resistance. These findings indicate that the pulmonary tissues play a major role in infantile wheeze and argue for using a technique capable of estimating the airway and tissue properties separately.