Effect of fluticasone 250 μg/salmeterol 50 μg and montelukast on exhaled nitric oxide in asthmatic patients

BACKGROUND: Monitoring noninvasive biomarkers of inflammation is an important adjunct in asthma therapy.

Our goal was to prospectively evaluate bronchial and alveolar sites of inflammation using NO as the signal and the response following inhaled corticosteroids and add-on montelukast, a cysteinyl leukotriene receptor antagonist.

METHODS
Patient selection FENO was measured in 31 nonsmoking, clinically stable asthmatic patients (24 women), mean (± SD) age 43±9 years, who were regularly followed in a tertiary referral outpatient clinic for management of mild to severe persistent asthma (34,35).They all took inhaled fluticasone 250 μg/salmeterol 50 μg (F/S) twice a day for longer than a year.Aerosolized albuterol sulfate from a metered dose inhaler (MDI) and/or nebulizer, or a combination aerosolized albuterol sulfate and ipratropium bromide MDI and/or nebulizer, were used as acute relievers.All asthmatic patients were clinically stable for the previous eight weeks at baseline.
The classification of asthma guidelines (34,35) was modified in two ways: to include only treated asthmatic patients, and to determine severity stratification after 180 μg albuterol sulfate using an MDI.'Mild' refers to a forced expiratory volume in 1 s (FEV 1 ) of at least 80% of predicted value, 'moderate' refers to an FEV 1 of 79% to 61% of the predicted value and 'severe' refers to an FEV 1 lower than 61% of the predicted value.
All asthmatic patients had previous high-resolution, thinsection computed tomography of the lung with no, or trivial, emphysema present (36).
All asthmatic patients studied gave informed consent for participation.The present study was approved by the Institutional Review Board.
Asthmatic patients were not taking oral corticosteroids or leukotriene inhibitors within eight weeks of the measurement of baseline FENO.Optimal medical therapy was confirmed by evaluating clinical status and serial spirometry obtained during routine outpatient clinic visits, every one to two months over three to five years, in each asthmatic patient, to ensure all baseline values were equal to the best values previously documented.Patient protocol: Single-blinded, nonrandomized, fixedsequence crossover intervention.Visit 1 (Baseline): All asthmatic patients who were clinically stable for longer than eight weeks were taking F/S twice a day for longer than a year.Visit 2: After four weeks of taking F/S plus montelukast 10 mg daily (M).Visit 3: After four weeks of taking M plus S. Visit 4: After four weeks of taking S only.
Attempts were made to ensure asthmatic patients adhered to their medication schedule through weekly telephone calls and/or office visits.
Measurement of exhaled NO gas exchange FENO was measured in every asthmatic patient at three separate constant expiratory flow rates: 100 mL/s, 150 mL/s and 200 mL/s, in triplicate.The mean of three values (that were required to be within 10% of each other to be acceptable) was reported using a Sievers NOAi 280 chemiluminescence analyzer (Ionics Instruments, USA) with varying expiratory airflow resistors.Most importantly, Tsoukias and George (33), Tsoukias et al (37) and George et al (38) have shown that the relationship between alveolar NO concentration (CANO) and expiratory flow is linear above a threshold of 50 mL/s.To avoid nasal NO contamination, a mouth pressure greater than 5 cm H 2 O was used, as previously recommended (39).The NO analyzer was calibrated, at minimum, daily, with a known NO concentration (45 ppm), and before each patient or control subject was tested, with NO-free air.The technique of Tsoukias and George (33,37,38) was used to calculate bronchial NO maximal flux and steady-state (CANO), as previously published (1,7).One of the authors (CFT), solely responsible for obtaining and calculating NO gas exchange values and all other lung function data, was blinded to patient medication and dose.

Other lung function studies
Asthmatic patients, who were clinically stable for at least eight weeks, were instructed to continue all their medications, but to withhold inhaled albuterol sulfate and ipratropium bromide 8 h before testing, and long-acting beta 2 -agonists 48 h before testing.Techniques for measuring spirometry, thoracic gas volumes and airway resistance using plethysmography and diffusing capacity have been previously published (40,41).

Statistical methods
Healthy subjects and asthmatic patients were compared using ANOVA on the ranks (nonparametric ANOVA, Kruskal-Wallis test).Alternatively, the Wilcoxon rank sum test was used.Data distribution by visit was tested using the Shapiro Wilk W statistic.If the P value was greater than 0.15, and W was greater than 0.94, then the hypothesis of normality was not rejected.Otherwise, it was rejected, and further analyses were performed using natural log-transformed data.ANOVA for repeated measurements was used to test for differences between different visits (General Linear Model Procedure in SAS, using individual asthmatic patients as the random variable).Pair-wise comparison of asthmatic patients where the general test indicated significant differences were performed using posthoc linear contrasts and Student-Newman-Keuls tests to keep the overall experimental error lower than 0.05.P<0.05 was considered statistically significant.SAS, version 8.02 for Windows, was used for analysis (SAS Institute Inc, USA).

RESULTS
Baseline lung function and NO gas exchange Results of baseline lung function studies in 30 asthmatic patients, who completed the present study without any missing data, are described in Table 1.An additional patient declined to continue the present study following an asthma exacerbation during the first week.Results of baseline and subsequent NO gas exchange data following drug intervention are described in Table 2 and Figures 1 to 4.

Spirometry
There were no statistical differences for spirometry in asthmatic patients during visits 1 to 4.
Juniper asthma symptom score Juniper scores (42) were significantly higher (P=0.03)when taking S alone (visit 4) when compared with the three other regimens (visits 1 to 3) (Table 2).

NO gas exchange
Results of FENO (ppb) at 100 mL/s, CANO (ppb) and bronchial NO maximal flux (nL/s) are described in Table 2 and Figures 1 to 4 for different regimens.Values for FENO were significantly higher (P=0.007)when taking S alone compared with the three other regimens.Values for bronchial NO flux were significantly higher (P=0.03)when taking S alone, compared with F/S and F/S plus M.There was no significant difference for bronchial NO between S plus M, and S alone.Compared with healthy patients, 33% of all asthmatic patients had abnormal FENO (greater than 2 SD from the normal mean) and bronchial NO, despite the use of F/S, F/S plus M, and S plus M. Values for CANO were similarly elevated in all regimens, and abnormal in 20% of the asthmatic patients, when compared with normal values.

DISCUSSION
A novel observation in the present study identified no significant differences among CANO values when comparing F/S, F/S plus M, M plus S and S alone.Moreover, while there were significant differences for bronchial NO suppression for F/S and F/S plus M versus S alone, there were no significant differences between S plus M versus S alone, and between F/S versus F/S plus M. These compartmental differences would be masked if analyses were limited to FENO alone at a single expiratory flow rate.
The results in this single-blinded, nonrandomized crossover study of 30 asthmatic patients were similar to our   pilot observations (43) following four-and six-week washout of F in 13 asthmatic patients comparing similar therapeutic interventions.Furthermore, previous studies have noted that the vast majority of changes in FENO following treatment and subsequent washout with inhaled corticosteroids occurred within several days to one week (17,44).Sandrini et al (23) have also previously demonstrated, in a randomized, double-blinded, placebo-controlled crossover design lasting two weeks, M significantly reduced FENO after day 1 of treatment, with maximal effect occuring by day 7.These observations reinforced our efforts to design an efficient and cost-effective study, yet with sufficient washout time to avoid the carryover effects of anti-inflammatory controller medications.The current bronchial NO results are consistent with bronchial biopsy studies that demonstrate reduced antiinflammatory changes with montelukast versus inhaled fluticasone (45).Results in the present study noted the inability of inhaled corticosteroids to significantly suppress alveolar NO.Our previous observations using systemic corticosteroids (1), and studies by Berry et al (9), reported that five to 10 days of 30 mg prednisone resulted in the marked reduction of alveolar NO in asthmatic patients.This suggested that bronchoreactive, inducible NO synthase in the distal airways and/or alveoli responds to oral, but not inhaled, corticosteroids.The current NO observations lend support to previously published studies that reported equivocal results for the incidence of clinical exacerbations in asthmatic patients when comparing inhaled fluticasone plus salmeterol versus fluticasone plus montelukast (46,47).
Our observations of both increased bronchial and alveolar NO are consistent with results obtained by Berry et al (9) and van Veen et al (48) in asthmatic patients taking inhaled and oral corticosteroids, with a similar magnitude of expiratory airflow limitation as the present study.Elevated alveolar NO has also been noted in steroid-naïve patients with mild, intermittent asthma, with minimal expiratory airflow limitation and presence of nocturnal symptoms (8).The present results also complement the observations of Mahut et al (49), who noted increased alveolar NO in mildly symptomatic childhood asthmatic patients on inhaled corticosteroids, but with normal or near-normal spirometry.Alternatively, previous studies by Högman et al (50) and Shin et al (51) noted normal values for CANO in adult and childhood asthmatic patients, despite elevated bronchial NO.
The relationship between increased FENO, an inflammatory marker, and persistent expiratory airflow limitation has been further defined.In a recent publication (1), we noted that after taking 180 μg aerosolized albuterol, an FEV 1 of 76% of the predicted value or lower, and a FENO 100 mL/s of at least 28 ppb, could stratify clinically stable asthmatic patients at very high risk for future exacerbations requiring oral or parenteral corticosteroids over an 18-month observation period (1).
In addition to elevated alveolar NO, corroborative evidence suggests there may be a concomitant lung parenchymal component in moderate or severe persistent asthma.This includes the presence of inflammatory cells in both bronchoalveolar lavage fluid (52) and in lung parenchyma obtained by fibre optic bronchoscopy transbronchial lung biopsy in nocturnal asthmatic patients (53).Moreover, we recently reported unexpected and unexplained loss of lung elastic recoil properties not due to emphysema in stable, nonsmoking, chronic asthmatic patients with persistent, abnormal expiratory airflow limitation (40,41).However, there was no correlation with abnormal NO gas exchange.These physiological observations also challenged the prevailing concept that intrinsic small and large airway remodelling is solely responsible for expiratory airflow limitation in chronic, persistent asthma.Furthermore, Mauad et al (54,55) have noted fragmentation and decreased elastic fibres in subepithelial layers in both large and small airways, but not in alveoli.
The present observations noted increased bronchial and alveolar NO levels in chronic asthmatic patients taking inhaled corticosteroids, with no further NO suppression with the addition of M. This suggests concurrent, separate, incomplete or nonsuppressed proximal and distal airway inflammatory sites in the asthmatic patients studied.
SUPPORT: This study was funded by an investigator-initiated grant from Merck & Co, Inc, Whitehouse Station, New Jersey 08889, USA, to AFG.Merck & Co, Inc, had no input on data collection, analyses or development of the manuscript.
Gelb et al Can Respir J Vol 15 No 4 May/June 2008 194 Nitric oxide in treated asthmatics Can Respir J Vol 15 No 4 May/June 2008 195

Figure 1 )Figure 2 )Figure 3 )Figure 4 )
Figure 1) Measurement of fractional exhaled nitric oxide (FENO) at 100 mL/s, including mean (+), median, first and third interquartile range, and fifth and 95th percentile in asthmatic patients on four different treatment regimens.Visit 1 (baseline): fluticasone 250 μg/salmeterol 50 μg (F/S) twice a day; visit 2: post four weeks F/S and montelukast 10 mg daily (M); visit 3: post four weeks M and S; visit 4: post four weeks S only.*P<0.05 when comparing different treatment regimens.ppb Parts per billion

TABLE 1
Results of lung function in asthmatic patients at baselineLung function results are presented as mean ± SD. %pred Percent predicted; D L DIffusing capacity; D L CO SB Single breathing diffusing capacity for carbon monoxide; F/S Inhaled corticosteroids, fluticasone 250 μg/salmeterol 50 μg twice a day; FEV 1 Forced expiratory volume in 1 s; FVC Forced vital capacity; V A Alveolar volume

TABLE 2
Results of lung function studies and nitric oxide (NO) gas exchange for visits 1 to 4 in asthmatic patients and healthy subjectsResults for asthmatic patients are expressed as mean±SD and median (1-3 interquartile range).Results for healthy subjects are expressed as mean ± SD (1,7).
*Visit 4 is different from visits 1 to 3, P<0.007; **Visit 4 is different from visits 1 and 2, P=0.03.%pred Percent predicted; CANO Alveolar NO concentration; F Fluticasone 250 μg twice a day; FENO fractional exhaled NO; FEV 1 Forced expiratory volume in 1 s; FVC Forced vital capacity; M Montelukast 10 mg once a day; ppb Parts per billion; S Salmeterol 50 μg twice a day