To cite this article: Nair A, Vaidyanathan S, Clearie K, Williamson P, Meldrum K, Lipworth BJ. Steroid sparing effects of intranasal corticosteroids in asthma and allergic rhinitis. Allergy 2010; 65: 359–367.
Background: Treating allergic rhinitis may have a downstream anti-inflammatory effect on the lower airways. We conducted a dose ranging study in asthma and persistent allergic rhinitis to evaluate if intranasal corticosteroids exhibit a sparing effect on the dose of inhaled corticosteroid.
Methods: Twenty five participants were randomized to receive two weeks of 100 μg/day (Low dose) or 500 μg/day (High dose) of inhaled fluticasone propionate both with intranasal placebo; or inhaled fluticasone 100 μg/day with intranasal fluticasone 200 μg/day (Combined) in a double-blind cross-over fashion.
Results: Low dose fluticasone produced a shift of 1.20 doubling-dilutions (95% CI, 0.63, 1.77); Combined fluticasone, 1.79 doubling-dilutions (95% CI, 0.77, 2.80) and high dose fluticasone, 2.01 doubling-dilutions (95% CI, 1.42, 2.61) in methacholine PC20 from respective baselines. There was a significant difference between high and low doses: 0.82 doubling dilutions (95%CI, 0.12, 1.50) but not between combined and low dose 0.58 doubling dilutions (95% CI, –0.78, 1.95). Combined treatment alone produced improvements in peak nasal inspiratory flow (P < 0.001), rhinitis quality of life (P = 0.004) and nasal NO (P = 0.01); reduced blood eosinophil count (P = 0.03), and serum eosinophil cationic protein (P = 0.02). All treatments significantly improved tidal NO, FEV1 and asthma quality of life.
Conclusions: High-dose fluticasone was superior to low dose fluticasone for methacholine PC20, demonstrating room for further improvement. Combined treatment was not significantly different from low dose fluticasone and we could not demonstrate a steroid sparing effect on methacholine PC20. Combined treatment alone produced improvements in upper airway outcomes and suppressed systemic inflammation but not adrenal function.
Asthma and allergic rhinitis (AR) frequently coexist and are considered to be different manifestations of the same inflammatory disease continuum (1–3). Braunstahl and colleagues elegantly demonstrated the presence of significant naso-bronchial crosstalk with their experiments on segmental bronchoprovocation (4, 5). Epidemiological studies have indeed shown the adverse impact of AR on asthma control, hospitalization, and quality of life (6–9). Equally, treating AR has been shown to reduce the frequency of asthma related visits to the emergency department and inpatient hospitalization (7, 9–11). Likewise, the beneficial effects of intranasal corticosteroids in AR on bronchial hyperresponsiveness have also been reported, suggesting a downstream anti-inflammatory effect (10, 12).
However, whether adding intranasal corticosteroids to patients with concomitant asthma and persistent allergic rhinitis (PAR) allows a reduction in the dose of inhaled corticosteroid required to achieve control, has to the best of our knowledge not been investigated in a prospective randomized fashion. We have therefore evaluated the steroid sparing effect of intranasal fluticasone on the dose of concomitant inhaled fluticasone by carrying out a dose response study between 100 and 500 μg/day of inhaled fluticasone (both with intranasal placebo) vs 100 μg/day of inhaled fluticasone with 200 μg/day of intranasal fluticasone, using bronchial methacholine challenge as the primary outcome. We also measured other surrogates of lower and upper airway inflammation, airway calibre, quality of life, and systemic adverse effects.
Patients were recruited from our database of volunteers with the following inclusion criteria: males and females, 18–65 years, asthmatics with FEV1 ≥ 60% on ≤1000 μg beclometasone dipropionate equivalent units (BDP) with concomitant PAR as evidenced by symptoms and a positive skin prick test to at least one perennial aeroallergen, and a methacholine PC20 < 4 mg/ml. Assessment of asthma and PAR were based on current GINA and ARIA guidelines respectively (1, 13). Patients with nasal polyposis ≥ grade 2, deviated nasal septum ≥ 50%, oral corticosteroids within 3 months and respiratory tract infection within the preceding 2 months were excluded.
A single centre, randomized, double blind, double-dummy, three-way crossover design was employed. The Tayside committee for medical research ethics fully approved the study protocol. Patients attended a screening visit at which suitability was assessed using strict inclusion and exclusion criteria and written informed consent was obtained. Participants who were on ≥ 400 μg BDP had their corticosteroid dose tapered by half at weekly intervals till a dose of ≤400 μg BDP and ≤200 μg intranasal BDP was attained. Thereafter participants who were clinically stable had their medications stopped and were provided intranasal fluticasone placebo nasal spray and inhaled HFA-fluticasone placebo via pMDI to enter the placebo run-in period for 2 weeks (double-dummy). Stability was assessed based on peak flow variability, symptom scores and reliever use. Participants on long acting beta-agonists, leukotriene receptor antagonists, or anti-histamines, had these discontinued prior to placebo run-in and HFA-salbutamol and intranasal sodium cromoglicate were provided as rescue therapy.
The screening visit was followed by six study visits (Fig. 1), each separated by a minimum of 2 weeks. The 2-week inhaled and intranasal corticosteroid free washout period between study treatment visits was designed to exceed the five elimination half lives of fluticasone (t1/2 : 14 h). The washout period was extended by a further week in participants whose pretreatment baseline methacholine PC20 were not within one doubling-dilution to obviate a carryover effect. Participants received the following treatments in a randomized assignment order: (i) One puff of inhaled fluticasone propionate Evohaler pMDI 50 μg twice a day (total fluticasone dose 100 μg) and one puff of inhaled placebo twice a day with placebo intranasal spray two squirts each nostril once a day. (ii). One puff of inhaled fluticasone propionate Evohaler pMDI 50 μg twice a day (total daily fluticasone dose 100 μg) and one puff of Placebo twice a day with intranasal fluticasone propionate (Flixonase®) 50 μg two squirts in each nostril once a day (total intranasal fluticasone daily dose 200 μg). (iii) One puff of inhaled fluticasone propionate Evohaler 250 μg twice a day (total daily fluticasone dose 500 μg) and one puff of inhaled placebo twice a day with placebo intranasal spray two squirts each nostril once a day. The above treatments were designed to reflect dosages used in current clinical practice and allowed us to construct a putative dose response relationship to demonstrate assay sensitivity between the three treatment arms using doubling dilution shift in methacholine PC20 as the primary outcome. Standard instructions were given for inhaler and spray use and technique was checked at each visit.
All measurements were made at post run-in baseline and after each 2 week treatment and washout period. All study visits took place at 8.30 a.m. at the research unit.
Airway inflammation, hyperresponsiveness and systemic absorption
Exhaled tidal NO, alveolar NO, bronchial flux and nasal NO measurements were recorded prior to any pulmonary or nasal function measurement by Niox® Nitric oxide analyser (Aerocrine AB, Solna, Sweden) using ATS criteria (14). Bronchial provocation testing using methacholine was performed using the five breath dosimeter technique in accordance with ATS recommendations to determine the PC20 threshold at baseline, after each treatment period, and after a corticosteroid free washout using log-linear interpolation (15). Estimation of PC20 for subjects who did not attain a reduction in FEV1 of 20% (i.e. those with a 15–19% fall in FEV1) was derived using single point log-linear interpolation formula (20 × concentration/FEV1 fall) (16). Adrenal suppression from systemic absorption of the inhaled and intranasal corticosteroid treatment was assessed by measuring overnight 10 h urinary free cortisol and creatinine ratios as described previously in literature (17).
Nasal and bronchial airway calibre
All patients received appropriate instructions at the initial screening visit and were required to demonstrate good technique in recording their peak nasal inspiratory flow (PNIF) using the In-Check® PNIF meter (Clement Clarke International Ltd, Harlow, UK). The morning PEF was recorded by using the Mini-Wright peak flow meter (Clement Clarke International Ltd). The technique was then further reassessed and reinforced, prior to each study visit. Peak nasal inspiratory flow and PEF measurements were recorded daily on a domiciliary basis. Spirometry was performed according to the ATS criteria.
Quality of life
Instructions for recording Juniper mini asthma and mini rhinoconjunctivitis quality of life questionnaires were provided at screening (18, 19).
All assays were performed in duplicate in a blinded fashion. The serum ECP was measured using an enzyme linked immunoassay technique (UniCAP; Sweden Diagnostics UK Ltd, Milton Keyes, UK) with an intra-assay co-efficient of variation of 3.3%. Blood Eosinophil count was analysed using the Sysmex XE 2100 Hematology auto analyser. The urinary cortisol was measured using a commercial radioimmunoassay kit (DiaSorin Ltd, Wokingham, Berkshire, UK), which has no cross reactivity with fluticasone. The intra assay coefficient of variation was 4% and the inter assay coefficient of variation was 8%. Urinary creatinine was measured on a Cobas-Bio auto analyser (Roche Products, Welwyn Garden City, UK). The intra assay and inter assay co-efficient of variation was 4.6% and 3% respectively.
spss version 14 (SPSS Inc., Chicago, IL, USA) was used to carry out the statistical analysis. A sample size of 24 completed patients was estimated to give 80% power to detect a one doubling-dilution shift (minimal important difference or MID) in methacholine PC20 value (primary endpoint) assuming a 0.84 within-patient SD for this outcome. The sample size estimations were supported by prior studies from our department and external data published to assist sample size calculations (20–22). Datasets were analysed for patients who completed the crossover study per protocol. Data were assessed for normality using visual inspection, Q–Q plots and with previous consideration for literature, and if appropriate non- Gaussian data was logarithmically transformed prior to analyses. Comparisons were made using an overall repeated measures anova (adjusted for baselines) with Bonferroni correction for multiple comparisons (P < 0.05, two tailed) and subject, treatment and sequence as factors.
Fifty participants were screened of which 37 were randomized into the trial. Twenty five subjects completed the study per protocol. Ten subjects withdrew because of personal reasons or illness unrelated to asthma and one subject had an exacerbation of asthma during washout prior to final randomized trial intervention and hence withdrawn. One additional subject was excluded from analysis as his methacholine PC20 was >4 mg/ml at baseline and hence should not have been permitted to continue in accordance with the trial protocol. Thus data from these 25 subjects were included in the final per-protocol analysis (Table 1). There was no significant difference in baseline and washout values for the primary outcome with respect to treatment or sequence (data not shown) thus ruling out a carryover effect. Similarly, there were no significant differences in the pretreatment baselines for all secondary outcomes except exhaled nitric oxide and rhinitis quality of life. Treatment effects are therefore presented as differences from respective pretreatment baselines.
|Subject||Age||Sex||ICS dose (μg)||SPT||FEV1% pred||FEF25–75% pred||PNIF l/min||FENO ppb||Nasal NO ppb||OUCC μm/mm||PC20 mg/ml|
|13||44||F||400||G, T, H,C||112||79||191.43||21.6||886.3||9.86||0.28|
|Mean (SEM)||*400 (0-800)||85.36 (3.04)||60.33 (3.93)||123.6 (9.81)||50.75† (1.14)||910 (1.04)||8.41† (1.11)||0.58† (0.16)|
Lower airway outcomes
For the primary outcome measure of methacholine PC20, all three treatments conferred significant improvements from respective pretreatment baseline (Table 2). Low dose fluticasone produced a 1.20 doubling-dilution shift (95% CI, 0.63, 1.77; P < 0.001); Combined fluticasone, a 1.79 doubling-dilutions (95% CI, 0.77, 2.80; P = 0.001) and high dose fluticasone, 2.01 doubling-dilutions (95% CI, 1.42, 2.61; P < 0.001) in methacholine PC20 from their respective baselines. There was a significant difference between high and low doses: 0.82 doubling-dilutions (95% CI, 0.12, 1.50; P = 0.02), showing there was room for further improvement. However there was no significant difference between combined and low dose treatments showing no additivity conferred by intranasal fluticasone: 0.58 doubling-dilutions (95% CI, –0.78, 1.95, P = 0.82) (Table 3).
|Lower airway outcomes||Baseline||High dose||Difference 95% CI, P||Baseline||Combined (LD+NFP)||Difference 95% CI, P||Baseline||Low dose||Difference 95% CI, P|
|Methacholine PC20*||0.64 (0.81)||2.57 (0.93)||2.01 (1.42, 2.61) P < 0.001||0.47 (0.98)||1.63 (0.95)||1.79 (0.77, 2.80) P = 0.001||0.76 (0.80)||1.75 (0.92)||1.20 (0.63, 1.77) P < 0.001|
|FENO† Ppb||51.16 (1.15)||28.58 (1.13)||0.56 (0.45, 0.69) P < 0.001||49.71 (1.16)||29.68 (1.13)||0.60 (0.51, 0.72) P < 0.001||39.07 (1.15)||25.04 (1.14)||0.65 (0.54, 0.78) P < 0.001|
|CANO† ppb||2.71 (1.31)||2.31 (1.15)||0.85 (0.44, 1.65) P = 0.61||1.84 (1.05)||2.42 (1.05)||1.31 (0.71, 2.44) P = 0.36||1.83 (1.31)||2.46 (1.26)||1.34 (0.68, 2.63) P = 0.37|
|Bronchial flux† (nl/s)||2.27 (1.17)||1.30 (1.15)||0.57 (0.42, 0.77) P = 0.001||2.19 (1.17)||1.42 (1.14)||0.64 (0.50, 0.83) P = 0.002||1.95 (1.15)||1.34 (1.15)||0.68 (0.57, 0.82) P < 0.001|
|PEF l/min||446.97 (23.98)||466.24 (24.72)||19.27 (8.48, 30.06) P = 0.001||432.02 (23.05)||457.35 (23.05)||25.32 (12.77, 37.87) P < 0.001||448.39 (23.83)||459.04 (24.06)||10.64 (–1.06, 22.35) P = 0.07|
|FEV1% predicted||85.70 (3.20)||91.16 (3.08)||5.45 (1.95, 8.96) P = 0.004||86.87 (3.21)||88.91 (3.01)||2.04 (0.009, 4.07) P = 0.049||86.5 (3.14)||90.83 (3.19)||4.33 (1.44, 7.22) P = 0.005|
|FEF25–75% predicted||60.12 (3.89)||66.41 (4.69)||6.29 (0.77, 11.80) P = 0.02||60.62 (4.31)||63.08 (4.20)||2.45 (–1.03, 5.94) P = 0.15||62.95 (4.30)||66.00 (4.54)||3.04 (–0.24, 6.32) P = 0.07|
|Asthma quality of life units||5.82 (0.21)||6.07 (0.20)||0.25 (0.02, 0.48) P = 0.03||5.78 (0.18)||6.11 (0.18)||0.33 (0.10, 0.55 P = 0.006||5.91 (0.20)||6.22 (0.14)||0.31 (0.05, 0.57) P = 0.01|
|Upper airway outcomes|
|NNO Ppb||1002.2 (54.27)||1043.54 (69.65)||41.33 (–24.88, 107.54) P = 0.72||923.33 (70.89)||800.33 (42.31)||–123.00 (–204.83,–41.17) P = 0.01||927.23 (57.50)||993.26 (88.65)||66.03 (–199.89, 67.82), P = 0.10|
|PNIF l/min||127.23 (8.78)||126.20 (10.18)||–1.02 (–9.81, 7.75), P = 0.81||123.57 (9.89)||146.99 (10.49)||23.41 (14.29, 32.54) P < 0.001||128.99 (9.52)||130.66 (11.55)||1.67 (–8.93, 12.28), P = 0.74|
|Rhinitis quality of life units||1.70 (0.20)||1.44 (0.20)||–0.25 (–0.70, 0.19) P = 0.25||1.02 (0.20)||1.60 (0.18)||0.57 (0.20, 0.94) P = 0.004||1.54 (0.21)||1.64 (0.25)||0.09 (–0.24, 0.47) P = 0.57|
|ECP† (μg/l)||16.21 (3.00)||14.12 (1.63)||0.87 (0.56, 1.34) P = 0.53||17.37 (2.81)||13.49 (2.18)||0.77 (0.63, 0.97) P = 0.02||15.49 (3.59)||15.85 (2.56)||1.02 (0.77, 1.38) P = 0.82|
|Peripheral‡ blood eosinophils (cells/μl)||351.90 (46.20)||326.66 (44.74)||–25.23, (–79.0, 28.6), P = 0.34||369.0 (52.62)||310 (45.74)||–59.04 (–113, –5.03) P = 0.03||321.42 (37.35)||341.42 (46.62)||20 (–17.2, 57.2) P = 0.41|
|OUCC† (nM/mM)||8.64 (1.09)||8.46 (0.84)||0.97 (0.79, 1.21) P = 0.84||8.07 (0.96)||7.55 (1.00)||0.93 (0.72, 1.20) P = 0.59||8.74 (1.06)||8.25 (0.91)||0.94, (0.71, 1.25) P = 0.67|
|EMUCC† (nM/mM)||12.76 (1.2)||14.44 (1.18)||1.13 (0.78, 1.62) P = 0.49||16.49 (1.17)||15.98 (1.23)||0.96 (0.69, 1.36) P = 0.85||15.32 (1.19)||15.47 (1.16)||1.01 (0.72, 1.41) P = 0.95|
|Variable lower airway outcomes||Combined vs high dose Difference, 95% CI P||Combined vs low dose Difference, 95% CI P||High dose vs low dose Difference, 95% CI P|
|Methacholine PC20||0.22 (–1.18, 1.62) P > 0.99||0.58 (–0.78,1.95) P = 0.82||0.82 (0.12, 1.50) P = 0.02|
|Tidal nitric oxide, ppb||0.92 (0.69, 1.21) P > 0.99||1.07 (0.86, 1.33) P > 0.99||1.16 (0.87, 1.56) P = 0.81|
|Alveolar nitric oxide, ppb||0.64 (0.21, 1.97) P = 0.95||1.01 (0.31, 3.31) P > 0.99||1.57 (0.42, 5.78) P > 0.99|
|Bronchial flux (nl/s)||0.88 (0.57, 1.34) P > 0.99||1.06 (0.75, 1.49) P > 0.99||1.20 (0.84, 1.71) P = 0.56|
|PEF (l/min)||–6.04 (–25.15, 13.06) P = 1||–14.68 (–32.46, 3.10) P = 0.13||–8.63 (–24.71, 7.44) P = 0.53|
|FEV1% predicted||3.41 (–0.97, 7.81) P = 0.17||2.29 (0.98, 5.56) P = 0.25||–1.12 (–4.9, 2.6) P > 0.99|
|FEF25–75% predicted||3.83 (–5.94, 13.61) P = 0.96||0.58 (–5.07, 6.24) P > 0.99||–3.25 (–9.85, 3.35) P = 0.64|
|Asthma quality of life units||0.07 (–0.29, 0.44) P > 0.99||0.01 (–0.37, 0.39) P > 0.99||–0.06 (–0.41, 0.28) P > 0.99|
|Upper airway outcomes|
|Nasal nitric oxide ppb||–164.33 (27.35, 301.32) P = 0.01||–189.03 (–317.73, –60.33) P = 0.002||–24.70 (–152.70,103.30) P = 0.14|
|PNIF (l/min)||–24.45 (–37.30, –11.58) P < 0.001||–21.75 (–37.61, –5.88) P = 0.005||2.70 (–10.35, 15.75) P > 0.99|
|Rhinitis quality of life units||0.83 (0.04, 1.62) P = 0.03||0.48 (–0.06, 1.02) P = 0.09||–34 (–1.18, 0.49) P = 0.89|
|ECP (μg/l)||0.89 (0.49, 1.61) P = 1||0.76 (0.47, 1.22) P = 0.46||0.85 (0.47, 1.53) P > 0.99|
|Peripheral eosinophil count (cells.μ/l)||33.81 (–67.68, 135.30) P > 0.99||79 (–13.68, 171.78) P = 0.11||45.23 (–14.08, 104.56) P = 0.18|
|Overnight urinary cortisol creatinine ratio (nM/mM)||1.04 (0.73, 1.48) P > 0.99||1.008 (0.62, 1.63) P > 0.99||0.96 (0.58, 1.58) P > 0.99|
|Early morning urinary cortisol creatinine ratio (nM/mM)||1.14 (0.61, 2.22) P > 0.99||1.02 (0.51, 2.15) P > 0.99||0.89 (0.48, 1.65) P > 0.99|
Combined treatment with low dose fluticasone and intranasal fluticasone was not significantly different to high dose treatment. Tidal NO was significantly suppressed by all three treatments to a similar degree compared to baseline. Low dose fluticasone led to a 0.65-fold change (35% reduction) 95% CI, 0.54–0.78; P < 0.001, Combined treatment a 0.60 geometric mean fold change (40% reduction) 95% CI, 0.51–0.72; P < 0.001 and high dose fluticasone a 0.56-fold change (44% reduction) 95% CI, 0.45–0.69; P < 0.001. Exhaled NO arising from the medium sized airways (bronchial flux) was also significantly suppressed by all three treatments to a similar degree (i.e. 32%, 36%, 43%, reductions by low, combined and high dose respectively). None of the treatments significantly reduced the alveolar component of exhaled NO.
All three treatments showed small but statistically significant improvements in asthma quality of life score, which were not clinically relevant as compared to the minimal important difference of 0.5 units. There were small, but statistically significant improvements in FEV1 or peak flow with all treatments, while FEF25-75 showed a small significant improvement with high dose fluticasone alone (Table 2).
Upper airway outcomes
Peak nasal inspiratory flow significantly improved from baseline only by combined treatment [mean difference 23.4 l/min, 95% CI (14.29–32.54); P < 0.001], and this was different from low or high dose inhaled fluticasone (Table 3). Nasal NO levels were significantly suppressed from baseline only by combined treatment (Mean –123 ppb, 95% CI–204 to –41.17; P = 0.01), which was significantly different compared to high dose or low dose (Table 3). The rhinitis quality of life questionnaire showed significant statistical and clinically relevant improvements from baseline (Table 2) in overall scores with combined treatment only [0.57 units, 95% CI (0.20–0.94), P = 0.004], which was significantly different from high dose inhaled fluticasone (Table 3) as compared to the MID of 0.5 units.
Systemic markers of inflammation and bioactivity
Serum eosinophil cationic protein and blood eosinophil count were significantly suppressed from baseline by combined treatment alone as: geometric mean fold change of 0.77, 95% CI (0.63 to 0.97), P = 0.02 and arithmetic mean change of –59 cells/μl, 95% CI (–113 to –5), P = 0.03, respectively. Systemic bioactivity as assessed by overnight and early morning urinary cortisol creatinine ratios were not suppressed from baseline by any of the three treatments (Tables 2 and 3).
For the primary outcome of methacholine PC20, our results showed that, while all three randomized treatments significantly improved airway hyperresponsiveness compared to baseline, there was no difference when comparing combined low dose inhaled fluticasone with intranasal fluticasone vs low dose inhaled fluticasone alone. Thus we were unable to demonstrate any steroid sparing effect per se on lower airway outcomes by adding intranasal steroid to low dose steroid. Crucially we also demonstrated a significant dose response effect between low and high dose fluticasone alone, showing there was room for further improvement by adding in intranasal fluticasone to low dose inhaled fluticasone.
This is to our knowledge the first randomized dose response study to prospectively investigate the putative sparing effects of intranasal corticosteroids on the dose of inhaled corticosteroid in patients with concomitant asthma and PAR. In this regard, it is pertinent to mention the study by Dahl et al. (23), who did not demonstrate superiority of combined intranasal (200 μg/day) and inhaled fluticasone (500 μg/day) over inhaled fluticasone alone (500 μg/day). Their patients had mild asthma and seasonal intermittent allergic rhinitis, with well preserved baseline lung function. They showed that treatment with 500 μg of inhaled fluticasone significantly improved morning PEF (primary outcome) by only 13 l/min, but pointedly did not improve methacholine PD20. Unsurprisingly the addition of intranasal fluticasone to this dose of inhaled fluticasone did not result in further improvements in PEF or methacholine PD20 or sputum eosinophils. This is because 500 μg/day of inhaled fluticasone would be on the top of the dose response curve for all of these outcomes in mild asthmatics. In other words there was clearly no room for further improvement in terms of being able to show potential additivity with intranasal fluticasone to 500 μg/day of inhaled fluticasone. Similarly, the study by Nathan et al. (24) once again in seasonal intermittent AR and asthma and also powered on PEF, showed no further improvement with intranasal fluticasone, as prior treatment with fluticasone propionate/salmeterol, 100/50 μg bid had fully optimized their already preserved lung function.
In this respect, our study is unique in that we looked at a lower dose of 100 μg/day of inhaled fluticasone. Importantly, we were able to show that there was room for further improvement between 100 μg vs 500 μg/day of inhaled fluticasone with a significant difference in the primary outcome of methacholine PC20 (Fig. 2). We did not demonstrate any significant difference between combined therapy vs low dose inhaled fluticasone (amounting to 0.58 doubling-dilutions) and thus were unable to conclude any significant additivity for the effect of intranasal fluticasone on airway hyper-responsiveness. It is, however, important to note that our study was powered to detect a one doubling-dilution difference in PC20 (within subject SD = 0.84) and as such may have missed differences smaller than the biological variability of this instrument. It could also be hypothesized that this might be a reflection of the type of challenge itself. Indeed, previous data have shown that sputum eosinophils are increased after adenosine monophosphate but not methacholine challenge (25). Hence, future studies looking for additivity with intranasal corticosteroid might be better advised to employ indirect challenges which reflect eosinophilic inflammation more closely in asthma.
As expected, only the combined treatment improved upper airway outcomes including PNIF, nasal NO and rhinitis quality of life (Fig. 3). From this we can infer that the apparent lack of additivity of intranasal corticosteroid on lower airway hyper-responsiveness cannot be attributed to a lack of topical efficacy on the upper airway.
The suppression of the blood eosinophil count and serum ECP by combined therapy but not high dose inhaled fluticasone is an interesting observation, given that combined therapy did not at the same time suppress urinary cortisol/creatinine. This systemic disconnect in turn suggests that the addition of intranasal fluticasone may have had a facilitatory effect on local eosinophil trafficking through the nose which was not associated with direct systemic absorption of fluticasone from the upper airway. This locally driven effect of intranasal fluticasone per se on systemic eosinophil markers is also supported by the lack of suppression by high dose inhaled fluticasone on the same systemic eosinophil makers.
We recognize the limitations of our study. Firstly, the dose response in PC20 when comparing 100 μg vs 500 μg/day of inhaled fluticasone was not mirrored by other markers such as exhaled NO, spirometry or asthma specific quality of life. Perhaps this is unsurprising given that we selected our patients a priori on the basis of being methacholine responsive, rather than having impaired lung function. Secondly, it is possible that a longer duration of treatment may have detected a further improvement in PC20. However, it has been shown previously with inhaled fluticasone that after 2 weeks treatment, the shift in PC20 is near maximal as compared to after 6 weeks (26). We chose to strike a pragmatic balance between the duration of treatment using a three-way crossover design, along with the risk of losing asthma control during the steroid free washout periods. Thirdly, it is important to consider any potential carryover effects when employing a crossover design, as in the present study. In terms of any potential carryover effect on the primary outcome between the three randomized treatment arms, our crossover analysis did not show any difference (P > 0.05) pretreatment baseline values, which is important when calculating the shift in PC20 from respective pretreatment baseline values. Finally, it is possible that one might show an additive effect of intranasal fluticasone on the lower airway in patients with more severe asthma. However, as this was an initial proof of concept study, we selected participants a priori on the degree of airway hyper-responsiveness. Indeed, we have clearly shown that our chosen primary end point had sufficient sensitivity in terms of there being room for improvement.
In summary, our results showed that for the primary outcome of methacholine PC20, the shift in methacholine PC20 with high dose fluticasone was significantly superior to low dose fluticasone demonstrating potential for improvement over and above low dose inhaled fluticasone alone. While combined treatment with intranasal and low dose inhaled fluticasone resulted in a significant improvement in methacholine PC20 from baseline, there was no difference when comparing with low dose inhaled steroid alone. Thus, we were unable to demonstrate any steroid sparing effect per se on lower airway outcomes by adding intranasal steroid to low dose steroid. However, as expected, combined treatment did improve upper airway outcomes and suppressed systemic eosinophilic inflammation which inhaled steroid alone did not. Further long-term studies are required to look at combined therapy on exacerbations.
This publication received unrestricted departmental research Grant from University of Dundee.