INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide (NO) assays are the first noninvasive, easily repeatable tests used to measure an inflammatory mediator in the airways. Online (OL) measurement of expired NO concentration, involving exhalation into an open-ended reservoir from which levels are continuously recorded, appears to be useful in monitoring response to anti-inflammatory medications (1). However, OL measurements are flow-dependent, and pediatric patients may have difficulty maintaining a constant expiratory flow rate (4). Moreover, low ppb-range NO analyzers are neither inexpensive nor portable, making their widespread use somewhat impractical. Therefore, use of reservoir bags containing mixed VC expiratewhich may be collected in various clinics and transported to a central laboratoryhas been recommended for clinical convenience (5). Here, we show that OL and VC NO concentrations are linearly related in asthmatic and control children. These data suggest that there may not be a compelling reason to favor the use of OL measurements over more practical VC assays in children.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Subjects 5 through 18 yr of age were recruited from the pediatric and adolescent clinic. Control subjects had no history of chronic or recurring lower respiratory symptoms, no acute upper or lower respiratory tract abnormalities as assessed by history and physical examination, and no evidence of airway obstruction on spirometric testing (FEV₁/ FVC  0.8). Subjects with asthma had had at least three episodes of airflow obstruction (FEV₁/FVC < 0.8 and/or inspiratory to expiratory time ratio of > 1:3) that abated after inhalation of a ₂-agonist, but no subjects were in acute respiratory distress at the time of the study.

Testing Procedure

Patients performed spirometry (MedGraphics Pulmonary Function System Series 1070; MedGraphics, St. Paul, MN). They were then asked to blow through a flow restrictor device into a chemiluminescence NO analyzer maintaining a pressure of 20 cm H₂O until a stable plateau was reached (< 10% change in NO concentration per second), according to manufacturer's recommendations for OL NO measurement (NOA 280; Sievers Instruments, Boulder, CO). This pressure correlated with a flow rate of 7.5 to 9.0 L/min as measured by pneumotachometry (Bicore CP100; Pulmonary Monitor, Irving, CA). The chemiluminescence device was calibrated at 0 and 9.8 ppm, in accordance with the manufacturer's recommendations, and further validated at 45 ppm. If the subject was unable to maintain a constant exhalation pressure long enough to reach a stable plateau, his or her inability to perform OL measurement was recorded along with his or her age and spirometry values. For those subjects able to achieve a stable plateau on three separate attempts, OL NO concentration was taken as the best plateau after the initial peak, and the mean of three separate attempts was recorded. If the ambient NO concentration was more than 20 ppb, subjects were asked to breathe NO-depleted (0.8 ppb ± 0.4) compressed air from a high flow face mask for 2 min before OL measurements were taken (7). After completion of OL measurements, the subjects performed a slow VC maneuver through a microbial filter (VacuMed, Ventura, CA) into an 8.1-L Tedlar gas-sampling reservoir (Cole Parmer, Niles, IL) as previously described (5). This reservoir was sealed at the end of the maneuver and then opened to the intake port of the chemiluminescence device. Mixed expired NO measured in the Tedlar bag was then recorded alongside the OL mean plateau NO. Confirmation of soft palate closure was performed by viewing the nasopharynx endoscopically during the VC maneuver. Mixed VC NO in Tedlar bags was measured over a 3-h period to show that NO remains stable over time (n = 6 bags). All gases were obtained from BOC Gases (Murray Hill, NJ).

Statistical Methods

Differences between OL measurements and VC measurements of expired NO concentration were compared using a Sigma Stat program employing standard least mean square analysis. Means were compared using the Wilcoxon-Mann-Whitney test and presented ± SEM. A p value < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Thirty-three subjects (13 control subjects and 20 subjects with asthma) participated in the study. A total of eight subjects (24% of total), including six asthmatic subjects (mean age, 8.2 ± 1.8) and two control subjects (mean age, 5.5 ± 0.5), were unable to perform the OL assay reproducibly at constant flow (Figure 1). Three subjects whose anti-inflammatory regimen was unclear at the time of the study were excluded. The remaining subjects, able to perform both the OL and the VC assay correctly, consisted of 11 in the control group and 14 in the asthmatic group. Most of the asthmatic subjects who were able to perform the test were in clinical remission and receiving anti-inflammatory therapy (seven were receiving inhaled glucocorticoids alone, two were receiving a combination of oral and inhaled glucocorticoids, one was receiving oral glucocorticoids alone, and one was receiving a combination of oral and inhaled glucocorticoids and cromolyn).

View larger version (7K): [in this window] [in a new window]   Figure 1.   Online measurement of sustained flow efforts. A representative pneumotachograph tracing is presented of a child unable to sustain expiratory flow between 0.125 and 0.150 L/s.

In subjects able to perform both techniques, there was a strong linear correlation between OL and VC NO measurements with r² = 0.88 (Figure 2). The OL NO measurements in the asthmatic group were approximately 1.8-fold higher than those in the control group (21.9 pb for asthmatic subjects versus 12.0 ppb for control subjects; p = 0.029) and 1.4-fold higher for VC NO measurements (11.5 ppb for asthmatic subjects versus 8.1 ppb for control subjects; p = 0.01) despite ongoing anti-inflammatory treatment in most, in accord with previous studies (7, 8). The two groups did not have significant differences in other parameters (Table 1). Consistent with previous studies (6, 7), VC measurements were highly reproducible, with a coefficient of variation of 4.8% ± 0.7% (18 tests in six subjects), and could be performed on the sample up to 3 h after collection (change in [NO] = 4.6%; n = 6).

View larger version (22K): [in this window] [in a new window]   Figure 2.   Linear relationship between online and vital capacity NO measurements for asthmatic (circles) and normal (squares) children 5 to 18 yr of age (r2 = 0.88; n = 25).

To determine why VC measurements were closely associated with OL measurements, we studied the course of anatomic and physiologic events associated with VC measurements. Resistance to flow during the VC maneuver (3.2 ± 0.3 cm H₂O/L/min) was associated with immediate closure of the soft palate, visualized endoscopically, upon initiation of expiration (n = 7 maneuvers in two subjects) (Figure 3). The palate remained closed through expiration, though flow fell from 15.2 ± 1.0 to 3.8 ± 0.6 L/min, and expiratory pressure fell from 47.8 ± 4.3 to 7.2 ± 0.6 cm H₂O (n = 6).

View larger version (132K): [in this window] [in a new window]   View larger version (120K): [in this window] [in a new window]   Figure 3.   Soft palate closure while performing a VC maneuver into a gas-sampling bag. The soft palate was visualized fiberoptically while a subject breathed comfortably at rest (left panel ) and performed a slow vital capacity maneuver into a gas-sampling bag as described in the text (right panel ). The soft palate was observed to remain closed throughout the gas-sampling procedure.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is a common respiratory disease in childhood characterized clinically by episodic wheezing, cough, and dyspnea; physiologically by reversible airflow obstruction; histologically by chronic airway inflammation with eosinophilia; and immunologically often by increased IgE antibody to environmental allergens. Management of asthmatic airway inflammation with glucocorticoids is effective in relieving symptoms and may minimize chronic morbidity (9, 10). Most commonly, assessment of airway inflammation has been based on clinical symptoms and pulmonary function testing; however, these methods do not specifically measure degree of inflammation and may lead to either insufficient or excessive glucocorticoid therapy (9). Bronchoalveolar lavage, bronchial biopsy, and induced sputum can be used to identify the presence of airway inflammation, but these methods are invasive, making their routine use impractical (9).

Measurement of nitrosopnea (7), on the other hand, appears to be the first noninvasive marker for airway inflammation. Specifically, expression of the inducible (or inflammatory) isoform of nitric oxide synthase (iNOS) is increased in asthmatic airway epithelial cells and is associated with increased concentrations of expired NO (5, 12). Moreover, iNOS mRNA levels are increased in asthmatic bronchoalveolar lavage fluid (S. Erzurum, personal communication), and inhalation of aminoguanidine, a relatively specific iNOS inhibitor, decreases expired NO levels in asthmatic subjects (13). Glucocorticoids, likewise, decrease both histologic evidence for asthmatic airway inflammation and iNOS expression in lung cells in vitro. Importantly, glucocorticoids reverse hypernitrosopneameasured by both OL and VC techniquesin longitudinal studies of adults and children (14). Additionally, asthma may be associated with increased catabolism of the airway S-nitrosothiol storage pool, increasing NO excretion (17). Of note, expired NO has recently been demonstrated to be a more sensitive marker for childhood asthmatic airway inflammation than are serum levels of either eosinophilic cationic protein or soluble interleukin-2 receptor (18). Taken together, these data suggest that high expired NO concentrations are a marker for asthmatic airway inflammation, which may be useful in longitudinal assessment of response to anti-inflammatory therapy.

Several different techniques for measuring expired NO have been used, including VC NO (5), peak expired NO (1, 2), tidal breathing NO (8), and plateau OL NO (19, 20) measured over time. Online NO measurements have been recommended for use by European Respiratory Society (ERS) guidelines (19). However, performing the OL technique by ERS guidelines not only requires equipment that is cumbersome but also depends on techniques that are difficult to perform correctly, particularly by the pediatric population.

There has been concern that VC measurements may not be reproducible from laboratory to laboratory (20). Variation may arise because insufficient expiratory pressure is applied during sample collection in some procedures. In our hands, these measurements are highly reproducible (5), though we have found normal values to be slightly higher using the Sievers NOA 280 instrument than those measured using Thermo Environmental instruments, perhaps reflecting different calibration techniques. In this regard, both OL VC and environmental measurements made in the United States are made using ppm-range calibrated gas, as ppb standards are unstable and therefore not available.

Like OL measurements in adults, VC measurements both distinguish asthmatic from control children and decrease with steroid treatment in childhood asthma (7, 8). We have now extended these observations to show that (1) certain children may have difficulty performing expired NO measurements using the OL technique, (2) VC NO assays provide enough pressure to close the velum and prevent nasal contamination, (3) NO concentrations measured using the VC technique are stable up to 3 h for transport to a central laboratory, and (4) VC NO measurements are strongly associated with OL measurements. Of note, the trend toward a plateau in OL NO measurements ( 10 ppb) at low VC values suggests the possibility that detection of low levels of inflammation using the OL technique may be somewhat insensitive; however, firm conclusions in this regard cannot be drawn.

Separate studies using OL and VC techniques have shown that both methods are useful in identifying asthmatic patients. We have now shown that the results of the two assays are strongly related in a direct comparison. The reason for this observed association does not appear to involve the nasopharynx, as there is good evidence for soft palate closure in both techniques (Figure 3) (20). It is tempting to speculate that plateau values represent dilution of NO produced in the prealveolar airways by alveolar air coming at a steady rate from the lung periphery. If NO is produced in the branching airways at a fixed rate, and the OL and VC assays are performed in a consistent fashion, a reproducible dilution of airway NO by alveolar air would be predicted. Empirically, it appears that the dilution is greater, but equally reproducible, in VC measurements because of the greater alveolar emptying involved in breathing down to residual volume. However, the precise anatomic source of NO, even in OL measurements, remains controversial, and it has not been our goal to resolve this complex issue.

Mixed VC measurements are sensitive enough to separate asthmatic from normal subjects, and they may have significant advantages over OL measurements. First, some children may have difficulty performing the OL technique. Second, OL measurements are not as convenient for routine use in multiple clinics, whereas VC samples may be transported from a clinic to a central laboratory. Considering these advantages, as well as the strong correlation between OL and VC NO measurements, there may not be compelling reason to favor OL over VC measurements for asthmatic hypernitrosopnea in children. Additional direct comparison between OL, VC, and other techniques such as tidal measurements would be helpful in standardizing childhood measurements.