INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Wheezing in infancy is a common problem, affecting 30 to 40% of all infants (1). The majority of recurrently wheezy infants are symptom free by 6 yr of age; however, approximately 25% are persistent wheezers (2). Epidemiologic and physiologic studies indicate that persistent wheezers have an atopic predisposition, whereas infants with transient wheezing tend to have diminished lung function early in life (1). These observations suggest two plausible hypotheses: (1) that the persistent wheezers have asthma and airway obstruction caused by airway inflammation; and (2) that transiently wheezy infants have small-caliber airways rather than airway inflammation.

In order to test these hypotheses, a simple, noninvasive technique is required that provides information about airway inflammation. Such a technique for assessing airway inflammation in infants could be used to develop rational approaches for treatment strategies, based on the presence and persistence of airway inflammation, and would be a valuable tool for investigating the etiology of asthma.

A potentially useful noninvasive marker of airway inflammation is exhaled nitric oxide (eNO) (4). eNO is thought to be increased in asthmatic individuals because of the induction of inducible NO synthase (iNOS) as a result of airway inflammation (4, 5). Inducible forms of NO synthase are inhibited by glucocorticoids (6, 7). It has therefore been suggested that eNO may be a useful measure of airway inflammation, and may hence constitute a useful tool for the diagnosis and management of airway inflammation. However, important methodologic factors can affect eNO levels (8, 9). One problem is the ability to measure eNO of specifically pulmonary origin, since exhaled air may be greatly contaminated by nasal eNO. Silkoff and colleagues showed that in adults, contamination of exhaled air with nasal NO can be avoided by oral exhalation through a resistance sufficient to produce a positive expiratory pressure and to thus close the vellum (8). Another factor that greatly influences the level of eNO is expiratory flow (8). A number of recent studies have shown that the effect of flow on eNO levels can be controlled by measuring eNO during exhalation from TLC at constant expiratory flow (8, 9). Therefore, methods controlling for the effect of flow and excluding nasal NO levels allow reliable measurements of eNO during a single exhalation.

Although eNO has been measured in infants previously, the techniques that have been used for this do not control for flow or exclude nasal NO from the expirate. Comparisons of infants based on these techniques may not be valid if the infants have different expiratory flow patterns, resulting, for example, from age differences or respiratory disease, or because the extent of nasal NO contamination of the expirate is unknown.

The aims of the present study were to demonstrate that a modification of the technique described by Silkoff and colleagues could be used to measure eNO in infants, and to investigate factors that might influence eNO levels.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirty infants, aged 3 to 24 mo, were recruited for the study. Sixteen had a history of three or more wheezing episodes (wheezy group). Of these, none had regular treatment with antiinflammatory drugs and all were well on the study day. The other 14 infants had no history of wheezing and had no significant respiratory illnesses (healthy group). Family history of atopy was obtained from parental questionnaires, and the two groups of wheezy (w) and healthy (h) children were subdivided into three subgroups as follows: (1) positive maternal and paternal first-degree family history of atopy (n = 9; w = 6); (2) positive single maternal or paternal family history of atopy (n = 13; w = 5); and (3) negative parental family history for atopy (n = 8; w = 5). The infants' median (range) body length was 77 cm (58 to 88 cm), and their median (range) weight was 10.6 kg (6.3 to 13.8 kg). All of the infants were studied in the supine position and while asleep following an oral dose of chloral hydrate (80 mg/kg). Written informed consent was obtained from all parents, and the study was approved by the Medical Ethics Committee of the Princess Margaret Hospital for Children.

Lung Function Measurements

We measured forced expiratory volume in 0.5 s (FEV_0.5) as an index of airway function, using the raised volume rapid thoracic compression (RVRTC) technique in 27 of the 30 infants (10, 11). Immediately before forced expiration, lung volume was raised during three consecutive cycles to an inflation pressure of 20 cm H₂O by using a fan pump connected to a face mask via a computer-controlled circuit (Inflate-all; Coleman, Inc., Wicket, KS). Following the first and second inflations, the infants exhaled passively.

At end-inspiration of the third inflation cycle, forced expiration was achieved by compressing the chest and abdomen with an inflatable jacket to transmit to the airway a pressure of 20 cm H₂O above inflation pressure at end-inspiration. Mouth pressure was measured at the face mask and flow was measured with a pneumotachograph (No. 3719; Hans Rudolph Inc., Kansas City, MO). The signals were collected and analyzed by computer with data acquisition and analysis software (Labdat-Anadat 5.2; RHT-Infodat, Montreal, Canada) that enabled digital integration of flow so that forced expiratory volume could be calculated. The mean FEV_0.5 was calculated from three technically acceptable forced expiratory maneuvers (10, 11).

eNO Measurements

A modification of the RVRTC technique was used to measure eNO by inflating the lungs and by performing an expiration against an expiratory resistance (Figure 1). An intravenous cannula (Insyte 16Ga; Becton Dickinson, Salt Lake City, UT) was inserted into the expiratory limb of the system to increase expiratory resistance. Three inflation cycles to 20 cm H₂O preceded the eNO measurement. After the first and second inflations the infants exhaled passively. After the third inflation, jacket pressure was increased manually during expiration, using a 3-L calibration syringe (Model No. 5530; Hans Rudolph) to achieve a constant mouth pressure of 15 cm H₂O and hence a constant expiratory flow (50 ml/s). For three subjects, eNO levels were also measured at a lower expiratory resistance (i.e., higher flow of 100 ml/s) to demonstrate flow dependence of eNO measurements.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Modified RVRTC technique to measure eNO.

A two-compartment mask was tightly fitted on the face of the infants and eNO was measured from the mouth compartment together with mouth pressure. A rapid-response chemiluminescence analyser (Model 280; Sievers, Boulder, CO) was used to measure eNO at a sampling flow of 200 ml/min. To verify that a closure of the velum occurred during determination of eNO, end-tidal CO₂ was measured in the nasal compartment (Normocap CO₂ Monitor; Datex Instrumentarium OY, Finland). The background level of NO on all study days was below 5 ppb. The analyzer was calibrated daily with a zero-NO gas and two gases of known concentrations of 9.5 ppm and 21 ppm NO.

Statistical Analysis

Sex (male or female) and disease status (wheezy or healthy) were analyzed as binary variables. Family history of atopy was analyzed as an ordinal variable (no history of atopy in either parent, history of atopy in either mother or father, or history of atopy in both parents). All other variables were analyzed as continuous variables. Bivariate analysis was done with Pearson's correlation coefficients, unpaired two-tailed t tests, and one-way analysis of variance (ANOVA) (12). Generalized linear models (logistic and linear regression) were used to model the effects of multiple covariates on categorical and continuous outcomes (12). Potential confounders included in the multiple regressions were sex, age, weight, length, and FEV_0.5 level.

The coefficient of variation (CV), as an indication of the ability of the measurement method to achieve reproducible eNO plateaus, was determined with the pooled SDs from three technically acceptable measurements in each child.

Statistical significance was defined at the nominal 5% level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 1 shows the demographic data for all of the infants, together with the mean (SD) plateau eNO levels and FEV_0.5 values. A plateau eNO level was achieved for each child (Figure 2). The CV derived from pooled SDs was 4.73%. There was no significant correlation between eNO and age (r = 0.09; p = 0.62), weight (r = 0.01; p = 0.95), length (r = 0.02; p = 0.91), or FEV_0.5 (r = 0.3; p = 0.13) (Figure 3A). FEV_0.5 measurements were positively correlated with length (r = 0.58; p = 0.001), and were significantly greater in healthy than in wheezy infants after adjustment for length, as indicated by linear regression (beta coefficient = 28.1, SD = 13.8; p = 0.05) (Figure 3B).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Simultaneous eNO and expiratory flow profiles of an infant, obtained with the single-breath technique and constant expiratory pressure.

View larger version (8K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   Figure 3.   (A) Scatterplots of eNO plotted against FEV0.5 in 27 children. (B) Scatterplots of FEV0.5 plotted against body length in wheezy (n = 16) and healthy infants (n = 11).

End-tidal CO₂ measured at the nose was not changed during expiration, indicating that the velum was closed during the maneuver. Flow dependence of eNO was shown in three subjects with mean (SD) eNO values at an expiratory flow of 100 ml/s of 18.8 ppb (0.4), 15.5 ppb (0.1), and 6.8 ppb (2.4), and mean (SD) eNO values at an expiratory flow of 50 ml/s of 30.2 ppb (0.5), 34.5 ppb (0.7), and 12.6 ppb (0.6), respectively.

Levels of eNO were significantly higher in wheezy than in healthy infants (Figure 4). The mean (SD) levels of eNO were 18.8 ppb (12.4) in the healthy group and 31.8 ppb (18.3) in the wheezy group (t₂₆ = 2.30, p = 0.03). Binary logistic regression suggested that this difference was independent of age, sex, weight, length, family history of atopy, or the FEV_0.5 level.

View larger version (9K): [in this window] [in a new window]   Figure 4.   Box plot of eNO in healthy and wheezy infants. Whiskers represent range, box represents median, 25th, and 75th percentiles.

There was an effect of parental atopy on eNO that was independent of whether infants wheezed (Figure 5). ANOVA indicated that the mean levels of eNO were highest in the group of infants with both a maternal and a paternal positive family history of atopy (41.1 ppb); were lower in the group of infants with either a maternal or a paternal positive first- degree family history (24.2 ppb) of atopy; and were lowest in the group of infants with no family history of atopy (10.8 ppb) (F_2,27 = 12.27, p < 0.001). Generalized linear modeling indicated that this relationship was independent of age, sex, weight, length, the symptom of wheezing or the FEV_0.5 level (F_2,24 = 26.54, p < 0.001).

View larger version (13K): [in this window] [in a new window]   Figure 5.   Box plot of eNO according to family history of atopy. Whiskers represent range, box represents median, 25th, and 75th percentiles.

The effect of parental atopy was also evident in the wheezy group analyzed separately. ANOVA indicated that the mean levels of eNO in wheezy infants were highest in the group of infants with both a maternal and a paternal positive first- degree family history of atopy (43.5 ppb); were lower in the group with either a maternal or a paternal positive first-degree family history of atopy (39.8 ppb); and were lowest in the group with a negative family history of atopy (9.6 ppb) (F_2,13 = 16.92, p < 0.001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have shown that the single-breath technique with positive expiratory pressure can be adapted to measure eNO in infants. A modification of the RVRTC technique with expiration against a flow resistor enabled us to measure eNO during constant expiratory flow. End-tidal CO₂ measured from the nose did not increase during expiration, showing that closure of the velum occurred during the forced expiratory maneuver. Therefore, eNO levels measured in this way are unlikely to be contaminated by nasal NO. Using this method, we observed that the mean level of eNO was significantly higher in a group of young wheezy infants than in healthy infants. There was no relation between FEV_0.5 and eNO when the groups were combined or in either group of infants analyzed separately. In part, this might be due to the wide variation in FEV_0.5 in the age group studied. However, airway inflammation might be present in young children without causing airway obstruction, or eNO levels might reflect processes in addition to airway inflammation.

Four of the wheezy infants in our study had relatively low levels of eNO (i.e., less than 20 ppb). This would be expected if some infants wheeze because of relatively small-caliber airways rather than from airway inflammation. There might also be different temporal patterns of eNO expression, with some infants who wheeze having transiently increased eNO levels and others having persistently elevated eNO levels. In adults with asthma, there are data suggesting that inflammation is present when there is mild disease, and that inflammation persists even when symptoms are absent (13). Therefore, the persistent wheezers described by Martinez and colleagues might have persistent inflammation, in which case longitudinal assessment of eNO might predict, in infancy, children likely to have asthma (2, 3). More information is needed about temporal changes in eNO during infancy and their association with wheezing.

We also observed that four infants in the healthy group had relatively high levels of eNO (i.e., greater than 20 ppb). Questionnaire data indicated that both parents of three of these four infants had atopy, and that the proportion of infants included in the study from atopic families was greater than that in the general Western Australian population. Given these observations, we decided to examine whether a parental history of atopy had an independent effect on levels of eNO. Surprisingly, we found a significant effect, with levels of eNO highest in infants with both parents affected by atopy and lowest in those with neither parent affected by atopy, independently of whether or not the infants were wheezy. A plausible explanation for this observation in infants is that an increased level of eNO reflects immune deviation associated with atopy that might already be present before the development of symptoms. An independent effect of atopy on eNO levels is supported by data from a study of healthy, nonasthmatic primary-school children showing that levels of eNO increase with the number of positive skin reactions to common allergens (14). An alternative explanation for these data in infants and children is that increased levels of eNO in asymptomatic, atopic children reflect the presence of subclinical inflammation. Therefore, future studies should examine the relations between eNO levels and other markers of airway inflammation in infants and young children.

We conclude that the single-breath technique with positive expiratory pressure is a feasible technique for measuring eNO in infants, since it effectively controls for flow dependence and contamination by nasal NO. The technique is minimally invasive, and can be combined with other measures of lung function. The measurement of eNO is potentially a useful marker of airway inflammation in infants and young children. Longitudinal assessment of eNO from early infancy, in order to determine its relation to symptoms, atopy, lung function, and outcome, might significantly increase understanding of the pathophysiology of illnesses producing early wheezing, and enable the development of effective strategies for reducing morbidity from asthma.