INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Exhaled nitric oxide (ENO) is increasingly used as a marker of disease activity in asthma and other lung diseases. ENO is high in asthma (1, 2) and falls rapidly after treatment with agents with an anti-inflammatory effect, such as inhaled and oral corticosteroids. Other respiratory conditions associated with a raised ENO include adult bronchiectasis (3) and viral respiratory infections in normal subjects (4).

Simple and standardized procedures allow the measurement of ENO from the lower airway (5, 6) without contamination by nasal nitric oxide (NO) which is present in high concentrations relative to the lower respiratory tract (7). This can be achieved by exhaling against resistance, which closes the vellum and excludes nasal NO, and the reproducibility of the measurements is further improved by tightly controlling expiratory flow rate (5).

ENO research protocols often incorporate more conventional measures of lung function, such as spirometry before and after bronchodilator administration. During preliminary studies in our laboratory, we noted that ENO values were perturbed following spirometry and that bronchodilatation increased ENO. We devised the following study to examine systematically the effect of these two interventions.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

We recruited four male and three female healthy control subjects aged 17-45 yr. They were all nonsmokers with no history of allergy or upper respiratory infection in the 4 wk prior to the study. We recruited 11 subjects with asthma (five males aged 15-66 yr) with a screening ENO greater than 40 parts per billion (ppb) (> mean + 1 SD of values obtained in healthy control subjects in our laboratory using the same ENO measurement technique). The diagnosis of asthma conformed to the criteria of the American Thoracic Society (1986). Exclusion criteria in the 4 wk before the study included the use of oral or inhaled corticosteroids, other anti-inflammatory agents, long-acting ₂-agonists, and upper respiratory tract infection. We also excluded those with a history of smoking within the previous 3 yr or more than 5 lifetime pack-years, pregnancy, and any other significant chronic medical disease.

All procedures were reviewed and approved by the Institutional Review Board of the University of Toronto. All subjects gave written informed consent to the study.

Study Design

There were three study days (Table 1). Day 1 assessed the spirometry effect in healthy and asthmatic subjects. In the asthmatic subjects only, Days 2 and 3 reassessed the effect spirometry followed after 1 h by bronchodilator or placebo administered double-blind in a randomized fashion. On Days 2 and 3, baseline ENO was required to vary by less than 15% or the visit was rescheduled. Bronchodilators were withheld for 8 h, strenuous exercise was prohibited for 12 h, and subjects refrained from eating for 1 h before the study.

The Effect of Spirometry on ENO

In healthy volunteers and on Day 1 in the subjects with asthma, ENO was measured at baseline and at 1, 5, 15, 30, 45, and 60 min following three maximal forced expiratory maneuvers. Spirometry (SP) was repeated at the end of the evaluation to ensure that no spontaneous change in bronchial caliber had occurred. SP was performed according to the revised standards set by the American Thoracic Society (8). Exhalations were repeated until FEV₁ agreed at the 5% level. On Days 2 and 3, the same procedure was followed during the first study hour in the subjects with asthma, allowing three estimates of the spirometry impact on ENO for each subject.

On Day 1, in 8 of the 11 subjects with asthma, ENO was measured five times over 1 h before spirometry to assess the impact of ENO measurement itself on ENO. Each subject with asthma attended on three separate study days; the time of each evaluation was kept within 1-2 h to minimize any circadian effects on ENO.

The Effect of Inhaled ₂-Agonist and Placebo on ENO

ENO was measured at 60 min after spirometry, which constituted the baseline for the effect of active or placebo inhaler. The patient then inhaled in a randomized double-blind fashion either two puffs of salbutamol 100 µg/puff (Ventolin; Glaxo Inc., Mississagua, ON, Canada) by metered-dose inhaler (MDI) on the bronchodilator day or two puffs of a matching placebo (placebo day). ENO was then monitored at 1, 5, 15, 30, 45, and 60 min after inhalation. SP was performed after the last ENO measurement to quantify any bronchodilation.

ENO Measurement Technique

ENO measurement was performed using a previously described technique (5). The seated subject (with no nose clip) inserted a mouthpiece and inhaled to total lung capacity from a reservoir of medical grade compressed air that contained less than 1 ppb NO. The subject then exhaled via high resistance and maintained a mouth pressure of 20 mm Hg, which was displayed on a pressure gauge. The positive mouth pressure caused the velum to close, thus excluding nasal NO. The resultant expiratory flow rate was 45 ml/s. The NO profile showed a rapid rise to a steady NO plateau, which was taken as the ENO value. Repeated exhalations were performed to achieve three ENO values that agreed at the 5% level.

NO Analysis

NO was measured with a rapid response chemiluminescent analyzer (NOA 280; Sievers Instruments, Boulder, CO). Daily calibration was performed with 100% nitrogen (contained < 1 ppb NO) as zero gas and a standard 1.6 ppm NO gas (Praxair, Mississauga, ON, Canada). The NO sampling rate was 200 ml/min.

Statistical Analysis

Statistical analysis was performed with the SAS analysis program (version 6.12). In the asthmatic subjects, post-hoc analysis revealed that the effect of SP was similar for the three evaluations of the spirometry effect on ENO (Days 1-3), so data were pooled. After log transformation, data were normally distributed and analyzed using repeated measures ANOVA. Post-hoc testing was performed using Dunnet's method to compare each time point to baseline. Data from the subjects with asthma on Days 2 and 3 (a two-period two-treatment crossover design with repeated measures within periods) were analyzed using a mixed effects model (1, 2, 4). This model included effects for period, carryover, treatment, time, the interaction of treatment and time, and a random subject effect. An interaction between carryover and time was tested for, but was not significant and thus dropped from the model. Data were log-transformed prior to analysis in order to produce more normally distributed residuals. All tests were two-sided and conducted at the 5% significance level. Pre- and post-SP values were compared between study days using repeated measures ANOVA with Tukey's method for post-hoc testing. All tests were two-tailed and a significant p value was less than 0.05.

The correlation between the change in ENO (the difference between ENO averaged over all time points after MDI and baseline ENO) and change in FEV₁, FVC, FEV1/FVC, ₅₀ and ₂₅ on the bronchodilator and placebo days was examined using the differences in the log-transformed data.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The Effect of ENO Measurement on ENO

The coefficient of variation in ENO measured five times over 1 h on Day 1 in eight subjects with asthma was less than 5% in all subjects with no trend to increase or decrease, showing that the ENO test itself did not cause a significant effect on ENO.

The Effect of SP on ENO in Healthy Subjects

The results are presented in Figure 1. There was a significant fall in ENO with SP (p = 0.0001). At 1 min after SP, ENO (geometric mean [confidence limits]) fell from 27.3 (22.9-32.6) to 23.6 (20.4-27.1) ppb, which constituted a mean fall of 13.0 ± 9.6% (range, 4.2-24.8%) compared with baseline. The effect was maximal at 1 min after SP, showing a gradual return to baseline levels at 60 min; ENO was significantly different from the pre-SP baseline at all time points, except at 60 min post-SP.

View larger version (13K): [in this window] [in a new window]   Figure 1.   The change in log ENO (mean + SEM) after spirometry in normal subjects. The standard errors refer to the raw data and are distinct from those used in the repeated measures ANOVA.

The Effect of SP on ENO in Subjects with Asthma

The effect and the time course of SP on ENO in the subjects with asthma was similar on all 3 d, so the data were combined. SP caused an immediate fall in ENO (geometric mean [confidence limits]) from 95.5 (80.9-112.9) to 86.9 (73.2-103.1) ppb. This constituted a fall of approximately 10% compared with baseline and similar to that seen in the normal subjects (Figure 2). Pair-wise comparisons at each time point showed that this fall was significant at 1 and 5 min post-SP (p < 0.01), while ENO progressively rose over 1 h and reached a value of 101.1 (84.9-120.3) ppb at 1 h post-SP, which was significantly higher than baseline values (p < 0.05).

View larger version (14K): [in this window] [in a new window]   Figure 2.   The change in log ENO (mean + SEM) after spirometry in asthma (pooled data, days 1-3). The standard errors refer to the raw data and are distinct from those used in the repeated measures ANOVA.

Days 2 and 3: The Effect of Placebo and Bronchodilator MDI on ENO in Asthma

The time course of ENO after placebo or bronchodilator MDI are shown in Figure 3, and the individual results are presented in Table 2. The mixed effects model confirmed the absence of a significant carryover or period effect (p = 0.4678 and 0.8792, respectively). The interaction of treatment and time was not significant (p = 0.6691), indicating that the time response curves for the active and placebo treatments were essentially parallel. The effect for treatment was highly significant (p = 0.0015), indicating that on average the active treatment ENO values were higher than the placebo values. The time effect was marginally significant (p = 0.0770).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Log ENO (mean ± SEM) after salbutamol and placebo MDI. The standard errors refer to the raw data and are distinct from those used in the mixed effects model.

Spirometric Values

Small but significant increases were seen in several spirometric variables following salbutamol inhalation (Table 3), but not on Day 1 or the placebo day. FVC, FEV₁, FEV₁/FVC, ₅₀, and ₂₅ were measured at 60 and 120 min on Day 1, and at 0 and 120 min only on Days 2 and 3. The general linear model for the conditions of study day (spirometry, placebo, bronchodilator) and intervention (SP, MDI) showed a significant interaction between day and intervention (p = 0.0016) with both day (p = 0.0003) and intervention (p = 0.0127) significant. Post-hoc testing showed a significant difference between FEV₁, FEV₁/FVC, ₅₀, and ₂₅ before and after MDI on the bronchodilator day only (p < 0.05).

The Correlation of Change in ENO and Change in Spirometry

Combining data from the placebo and bronchodilator days, the changes in log ENO before and after MDI were weakly correlated to the change in FEV₁, ₅₀, and ₂₅ (r = 0.42; p = 0.054, r = 0.40; p = 0.064, and r = 0.398; p = 0.066, respectively) but not correlated with FEV₁/FVC or FVC (r = 0.31, p = 0.167, and r = 0.34, p = 0.12, respectively).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our data address three factors of potential importance in measuring ENO. First, repeating the ENO measurement itself up to five times in 1 h did not perturb ENO values in subjects with asthma. Second, in both healthy and asthmatic subjects, the forceful maneuvers of SP caused an immediate 10-15% decrease in ENO, which took 1 h to return to baseline. Last, the inhalation of salbutamol by the subjects with asthma increased ENO by approximately 10%, a change that persisted for at least 1 h.

The effect of spirometry on ENO was small but significant with a mean fall of approxmiately 13% for baseline in normal subjects and approximately 10% in subjects with asthma. However, some subjects with asthma showed much larger falls (up to 18%). In both groups, ENO had recovered to pre-SP levels by 1 h. The precise reason underlying the ENO fall after SP is unknown but may be related to the procedure causing airway closure, similar to the observation that repeated SP may induce bronchospasm in some subjects with asthma. In this study we quantified the effect and time course of a single forced expiration session (at least three exhalations) on ENO measurement. Our findings are in agreement with Deykin and coworkers (9), who recently reported a significant ENO fall of more than 30% after repeated SP in subjects with asthma undergoing airway challenge, an effect that lasted several hours. The impact of SP stresses the importance of study design to avoid an artifactual depression of ENO. This will avoid spurious conclusions about the effect of medications or interventions on ENO.

On the bronchodilator day, both ENO and FEV₁ changed significantly in the subjects with asthma, while no change occurred in either modality on the placebo day. While the magnitude of the mean increase in ENO was small (~ 10%), one subject showed an increase of 29%, and we have observed larger changes in other subjects participating in other studies. The magnitude of this bronchodilator effect on ENO was greater than the variation in ENO (< 5%) observed over 1 h on Day 1. If NO synthesis increased due to augmented airway inflammation but was accompanied by bronchospasm, then the former might be partially masked by the latter. The bronchodilator effect on ENO may be the mechanism underlying the report by Mitsufugi and colleagues (10) that ENO rose significantly after the treatment of acute asthma in the emergency room with bronchodilators.

The precise cause of the rise in ENO after bronchodilation is unclear, but the rapid onset suggests a mechanical effect on NO output. This effect is also in keeping with our observations that ENO fell by about 20% at the time of methacholine- induced bronchospasm and returned to baseline after bronchodilator use (11) and that ENO in exhalations from FRC were about 20% less than those from total lung capacity. In attempting to explain the changes in ENO with airway caliber, we suggest that one factor may be recruitment of airways as evidenced by the increase of FVC (4.14 ± 0.25 to 4.43 ± 0.25 L) seen on the bronchodilator day but not on the placebo day (4.24 ± 0.26 to 4.08 ± 0.21 L). This recruitment of airways could improve the transfer of NO into the exhaled gas. The observation of others that end-expiratory pressure caused an increase in ENO (12) may also be related to airway recruitment. Another factor could be a change in respiratory epithelial surface area affecting the diffusion of NO into the lumen, although mucosal surface area has been reported to remain constant under different degrees of bronchodilation (15). Additionally, airway resistance is proportional to the fourth power of the radius; thus, large changes in resistance could occur for negligible changes in radius and thus surface area. Alternative explanations for the rise in NO after salbutamol cannot be ruled out, e.g., a direct effect of salbutamol on NOS activity.

The effect of bronchodilator administration on ENO has been reported previously. Our findings are in agreement with Yates and associates (16), who reported that acute administration of salbutamol caused a trend toward increased ENO in subject with asthma who were treated with inhaled corticosteroids. Salmeterol (a long-acting ₂-agonist), however, did not alter ENO, whether or not patients were using inhaled corticosteroids. In contrast to our findings, Garnier and coworkers (17) did not find any effect of methacholine bronchoconstriction or bronchodilator administration on ENO in asthma. The discrepancy may perhaps be explained by differences in the ENO measurement technique. In the latter study (17) exhaled gas from tidal breathing was collected into bags for analysis. The relatively high flow rates of tidal breathing may be insensitive to airway changes, as suggested by a recent model of NO airway diffusion (18). Lower flow rates as employed in this study (45 ml/s) and that of Yates and colleagues (16) are probably more sensitive to airway NO production. Low expiratory flows increase the transit time of exhaled gas in the airway. This amplified the absolute ENO values seen in this study (range, 97.8-132.4 ppb) and in that of Yates and colleagues (range, 143-211 ppb) (16) and also the differences before and after bronchodilation.

In summary, ENO levels may be altered by common laboratory procedures, such as spirometry and bronchodilator administration, which commonly accompany ENO measurement. In order to avoid erroneous conclusions about the effects of medications or interventions on ENO, study design must account for such potentially confounding factors.