INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

High doses of inhaled glucocorticosteroids are recommended in the treatment of moderate to severe persistent asthma (1). It is well accepted that inhaled glucocorticosteroids have fewer systemic side effects than oral or parenteral glucocorticosteroids (2), but there is still some concern about the long-term safety of high doses of inhaled glucocorticosteroids (3). Long-term systemic side effects of high doses of inhaled glucocorticosteroids include thinning of the skin and easy bruising (4). An association between the chronic use of high doses of inhaled glucocorticosteroids and the development of cataract (5) or the occurrence of open-angle glaucoma has also been reported (6). However, long-term prospective studies on systemic effects of inhaled glucocorticosteroids are difficult to perform. For that reason, the potential for systemic effects of inhaled glucocorticosteroids is often evaluated using measurements of adrenal function. The two most sensitive methods involve the measurement of plasma cortisol concentrations at regular time intervals over 20 to 24 h or of free urinary cortisol over 24 h (7).

Budesonide and fluticasone propionate ("fluticasone") are the most potent inhaled glucocorticosteroids on the market. Their potential for systemic effects has been compared in healthy volunteers (8, 9, 12), but patients with asthma might differ in this respect. We therefore compared the effects of two different, approximately equi-efficacious doses of both budesonide and fluticasone on adrenal function in patients with asthma. At the same time, we also examined the efficacy of these treatments.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Patients of either sex, between 18 and 60 yr of age, were included. They all presented symptoms of asthma according to the criteria proposed by the American Thoracic Society (ATS) (15). All patients had a baseline FEV₁  40% of the expected value (16). Moreover, they exhibited either a postbronchodilator increase in FEV₁ of at least 200 ml and  12% of baseline, 15 min after inhalation of 0.5 mg of terbutaline sulfate, or a diurnal variation in peak expiratory flow (PEF)  15% on at least 2 d during the 2-wk run-in period. Patients were excluded if they had experienced an exacerbation 4 wk before inclusion, had used oral glucocorticosteroids within 6 mo, or had been treated with inhaled, rectal, ocular, intraarticular, or other glucocorticosteroids during the 4 wk before inclusion. The use of ipratropium bromide, xanthines, or sodium cromoglycate was permitted provided that dosages were kept at a constant level 4 wk before and throughout the study. All participants were carefully instructed in the correct use of the inhalers. The study was performed in accordance with the principles stated in the Declaration of Helsinki and approved by the Ethics Committees of the University Hospitals of Ghent and Brussels (Belgium). All participants were fully informed about the purpose of the study and gave written informed consent before inclusion.

Design

The study was of a randomized, double-blind, double-dummy, placebo-controlled, and five-period cross-over design. Placebo and four active treatments were compared: fluticasone, 200 and 1,000 µg, twice daily, both inhaled from a Diskhaler (Flixotide Diskhaler, Glaxo Wellcome, Greenford, UK), and budesonide, 200 and 800 µg, twice daily, both inhaled from a Turbuhaler (Pulmicort Turbuhaler, Astra AB, Lund, Sweden). The duration of the treatment period was of 1 wk, as the available studies suggest that maximal suppression occurs after 5 d of treatment (9, 12). The five treatment periods were separated by washout periods of at least 2 wk (maximum, 3 wk). The first dose was taken on the evening of Day 1, at home, and the last (fifteenth) dose was always taken on Day 8 in the clinic at 10:00 P.M. All patients were instructed to inhale the study medication at 10:00 A.M. and at 10:00 P.M. and to report the exact time of inhalation in a diary. They were urged to rinse their mouths with water after each inhalation. They were allowed to use terbutaline (by Turbuhaler, 0.25 µg per dose) as rescue medication throughout the study, and the long-acting ₂-agonist salmeterol (50 µg, twice daily) during the washout periods.

Outcome Measurements

On Day 8 of each treatment period, patients came to the hospital at 8:00 P.M., and remained under close supervision for the next 22 h. A catheter was inserted into a forearm vein. At 10:00 P.M., blood samples for analysis of serum cortisol (5 ml) and total and differential white blood cell count (WBCs, 5 ml) were taken. Thereafter, the last dose of the study drug was administered. Further blood samples for analysis of cortisol and WBCs were obtained at 2 h intervals until 6:00 P.M. of the following day (Day 9). The blood samples for cortisol analysis were collected in tubes without anticoagulant and centrifuged at room temperature for 10 min at 3,000 rpm, the applied g force being 800 g (Hettich-Zentrifugen; Universal, Tuttingen, Germany). The serum was then transferred to new tubes and stored at 20° C until analysis. The blood samples for WBCs were stored in EDTA tubes at room temperature and analyzed within 8 h.

Lung function measurements (FEV₁ and FVC) were performed using a water-sealed spirograph (Expirograph; Goddard, Bilthoven, The Netherlands) on Day 9 at 7:00, 8:00, 9:00, and 11:00 A.M. The highest of three FEV₁ measurements at each time point were systematically retained.

Morning and evening PEFs were measured daily with a Vitalograph (Goddard) peak flow meter throughout the treatment period before inhalation of the study medication, and noted in a diary. Inhalers with study medication and diaries were returned to the supervising physician at each visit. Empty "blisters" that had contained fluticasone or placebo and the number of doses remaining in the Turbuhaler were counted, allowing an estimation of therapeutic compliance.

Bioanalytical Methods

The analysis of serum samples for cortisol was performed in a blinded fashion at the Laboratory of Hormonology of the University Hospital of Ghent. A radioimmunoassay technique was used (GammaCoat ¹²⁵I-labeled cortisol radioimmunoassay kit; Incstar, Stillwater, MN). The lowest measurable concentration was 11 nmol/L. The interassay coefficient of variation was 9.4 and 4.6% at 68 and 371 nmol/L, respectively. The intraassay coefficient of variation was 4.5, 5.7, and 6.8% at 59, 378, and 742 nmol/L, respectively. There was no cross-reactivity with the antiserum at 50% of maximum binding for budesonide, 6--hydroxybudesonide, and fluticasone-17-propionate.

White blood cell total and differential counts were done at the Laboratory of Clinical Biochemistry of the University Hospital of Ghent, with an automatic five differential cell counter (STKS; Coulter Electronics, Hialeah, FL).

Data Analysis

Summary statistics (including arithmetic means, medians, and standard deviation) were calculated. The 20-h area under the curve (AUC_0-20) and the AUC_0-12 (representing the first 12 h of sampling) were derived from serum cortisol concentrations, using the trapezoidal method and actual sampling times. The AUC_0-20 for WBC total and differential counts was calculated in a similar way.

The FEV₁ values measured between 7:00 and 11:00 A.M. on Day 9 of each treatment period were compared by analysis of variance for repeated measurements, with treatment and time as variables. Differences between treatments at given time points were assessed, using a Friedman test. Morning and evening PEFs during each treatment period were analyzed using an additive model. Pairwise comparisons were then used to compare the two drugs at each dose level. In addition, all active treatments were compared with placebo and the two dose levels were compared for each drug. All tests were two-sided and p values less than 5% were considered significant.

In addition, analysis of variance with previous treatment as an additional factor was performed for all of the examined variables, in which placebo was assumed to have no effect on the next period.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Characteristics

Twenty-three patients (8 women) with mild asthma entered the study. Mean age was 33 yr (range, 19-57 yr), mean weight was 74 kg (range, 50-96 kg), and mean height was 172 cm (range, 156-183 cm). Compliance, estimated from the data reported in the diaries, was judged to be excellent. One patient withdrew during the washout period that followed the second treatment period (placebo) and another patient withdrew after the third treatment period (fluticasone, 1,000 µg twice daily) because of asthma exacerbations, for which they needed a course of oral glucocorticosteroids. Their data have been excluded from the analysis.

Baseline Spirometry

Mean FEV₁ at inclusion was 2.95 ± 0.83 L (range, 1.45-4.76 L), corresponding to 80.0 ± 21.4% of predicted values. Mean FVC was 4.42 ± 0.94 L (range, 2.91-6.46 L). Mean reversibility (range, 3-63%) of FEV₁ was 21.0% of baseline. One patient, who exhibited a reversibility of only 3% of baseline FEV₁, was included because he exhibited a diurnal variation in PEF of 29 and 38% during the run-in period.

Serum Cortisol

One week of treatment with 200 µg of fluticasone or budesonide twice daily did not produce any significant suppression of cortisol secretion, the shape of the two curves being similar to the curve obtained after placebo treatment (Figure 1). The AUC_0-20 was similar to that obtained after placebo treatment (Figure 2). However, inhalation of high doses of fluticasone or budesonide significantly modified the serum cortisol-time curve. Twice-daily inhalation of 800 µg of budesonide and 1,000 µg of fluticasone decreased the AUC_0-20 by 16% (95% confidence intrval [CI], 3-27%, p = 0.0167) and by 34% (95% CI, 25-48%, p < 0.0001), respectively. Differences between treatments with high and low doses of either drug as well as between the high doses of fluticasone and budesonide reached significance, 1,000 µg of fluticasone twice daily being more suppressive than 800 µg of budesonide twice daily (p = 0.0006). Two patients exhibited serum cortisol concentrations below the limit of detection (200 nmol/ml) at 8:00 P.M. after twice-daily treatment with 800 µg of budesonide, as did four patients after twice daily treatment with 1,000 µg of fluticasone.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Mean serum cortisol concentrations over a 20-h period after 1 wk of treatment with placebo, budesonide (200 µg, twice daily), budesonide (800 µg, twice daily) fluticasone (200 µg, twice daily), and fluticasone (1,000 µg, twice daily). Samples were obtained every 2 h, starting at 10:00 P.M., after inhalation of the last dose.

View larger version (15K): [in this window] [in a new window]   Figure 2.   Mean serum cortisol suppression, after 1 wk of treatment with budesonide (200 µg, twice daily), budesonide (800 µg, twice daily), fluticasone (200 µg, twice daily), and fluticasone (1,000 µg, twice daily), measured as AUC0-20 and expressed as a percentage fall relative to placebo for the various treatments.

The cortisol concentrations during the first 12 h after the last dose were not significantly different from placebo for the two low dose treatments, whereas the AUC_0-12 for serum cortisol decreased by 28% (95% CI, 13-40%) after budesonide (800 µg, twice daily) and by 45% (95% CI, 34-54%) after fluticasone (1,000 µg, twice daily). The suppression of cortisol secretion induced by fluticasone (1,000 µg, twice daily) during the first 12 h exceeded that resulting from budesonide (800 µg, twice daily) (p = 0.0037).

WBC Total and Differential Counts

Total white blood cell count, as well as neutrophils, basophils, lymphocytes, and monocytes, were not affected by any of the active treatments. Both fluticasone and budesonide, however, resulted in a dose-dependent decrease in eosinophils (Figure 3), which reached significance for the higher dose. Budesonide (800 µg, twice daily) thus caused a 40% (CI, 27-50%) decrease in eosinophil AUC_0-20, compared with placebo (p < 0.0001), whereas fluticasone (1,000 µg, twice daily) decreased the eosinophil AUC_0-20 by 29% (CI, 15-42%) (p = 0.0005). Differences between the two drugs were not significant. However, differences between the higher and the lower doses of a given drug reached significance (p = 0.0004 for budesonide [200 µg twice daily versus 800 µg twice daily]; p = 0.047 for fluticasone [200 µg twice daily versus 1,000 µg twice daily]).

View larger version (25K): [in this window] [in a new window]   Figure 3.   Mean cell count for total eosinophils over 20 h, after 1 wk of treatment with placebo, budesonide (200 µg, twice daily), budesonide (800 µg, twice daily), fluticasone (200 µg, twice daily), and fluticasone (1,000 µg, twice daily), starting at 10:00 P.M., after inhalation of the last dose.

Lung Function

The overall FEV₁ after all four active treatments significantly exceeded the FEV₁ after placebo (p = 0.012), significance being reached at 7:00 A.M. (p < 0.0001) and 8:00 A.M. (p = 0.006). No significant differences were observed between the four active treatments (Figure 4).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Mean FEV1 measured between 7:00 A.M. and 11:00 A.M., after 1 wk of treatment with placebo, budesonide (200 µg, twice daily), budesonide (800 µg, twice daily), fluticasone (200 µg, twice daily), and fluticasone (1,000 µg, twice daily).

On average, active treatments resulted in a significant increase in morning PEF (Figure 5), the increase ranging between 24 and 30 L/min (p < 0.006), compared with placebo. The increase in evening PEF ranged between 15 and 24 L/min (p < 0.03), compared with placebo. Differences in PEF between any of the active treatments were small and not significant.

View larger version (26K): [in this window] [in a new window]   Figure 5.   Morning PEF measured between the first day and the last day of treatment, consisting of placebo, budesonide (200 µg, twice daily), budesonide (800 µg, twice daily), fluticasone (200 µg, twice daily), and fluticasone (1,000 µg, twice daily).

Carryover Effect

Analysis of variance did not reveal the existence of any carryover effect.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study showed that a 1-wk treatment with low doses of inhaled budesonide or fluticasone did not suppress endogenous cortisol secretion in adult patients with mild asthma. Conversely, inhalation of their highest recommended doses resulted in a measurable suppression of cortisol secretion, which was greater for fluticasone than for budesonide. Both doses of both drugs caused a significant, but similar, increase in pulmonary function.

Adrenal suppression has been previously assessed in patients with asthma, using less sensitive methods, such as single morning cortisol serum levels (17). The present study, which is the first to provide convincing evidence that low doses of inhaled glucocorticosteroids do not exert any measurable systemic effect in asthmatic patients, has its limitations. First, some suppression of adrenal function might have been missed by omitting the inhalation on the morning of the ninth day of treatment. Restricting the analysis to the first 12 h, however, did not show any suppression with the low doses. Second, it remains difficult to predict whether the magnitude of adrenal suppression measured after 1 wk was representative of that after one or several years of treatment. Interestingly, the decreases in cortisol secretion over 12 h in children after 2 wk and 1 yr of treatment with inhaled glucocorticosteroids are in the same range (21).

Cortisol secretion in our patients after a 1-wk treatment with twice-daily doses of 1,000 µg of fluticasone inhaled via Diskhaler and twice-daily doses of 800 µg of budesonide inhaled via Turbuhaler was suppressed by 34 and 16%, respectively (Figure 2). This contrasts with previously reported data obtained in healthy volunteers, in whom treatments with high doses of fluticasone and budesonide reduced the AUC_0-20 and AUC_0-24 of the cortisol-versus-time curve by 55 and 34% after 3.5 d (13), by 84 and 27% after 4 d (9), and by 93 and 49% after 5 d (14), respectively. Our data are rather in line with most studies performed in patients with asthma, in which suppression of the hypothalamic-pituitary-adrenal (HPA) axis did not occur as long as doses did not exceed 800 µg of budesonide twice daily via pMDI (2, 17, 19, 22) or 750 µg of fluticasone twice daily via pMDI or Diskhaler (17, 18, 20, 22), although some suppression has been occasionally reported with lower doses of fluticasone (23).

Several factors may have contributed to the differences in suppression between patients and healthy volunteers, such as disease-induced differences in pulmonary deposition and absorption of inhaled drugs and the conditions under which the present study was performed. The spirometric measurements early in the morning might also have influenced cortisol secretion. The intake of doses 1 to 14 was not supervised, although patient compliance was checked by dose counting and found to be excellent.

A 1-wk treatment with 1,000 µg of fluticasone twice daily via Diskhaler suppressed endogenous cortisol secretion more than did a 1-wk treatment with 800 µg of budesonide twice daily via Turbuhaler. This is in line with the differences in suppressive effect between the two drugs reported in healthy volunteers (9, 12) and patients with mild asthma (20, 23). As patients systematically rinsed their mouths after inhalation, and inactivation through the first-pass metabolism is estimated to be 99% for fluticasone (24) and 89% for budesonide (25), the systemic effect is almost exclusively derived from the drug deposited in, and absorbed from, the lung (26). The observed difference in suppressive effect between the two treatments cannot be explained by a difference in pulmonary deposition between the inhalers because, if anything, pulmonary deposition of budesonide inhaled via Turbuhaler is likely to exceed that of fluticasone inhaled via Diskhaler (27, 28). However, differences in volume of distribution, clearance, plasma elimination half-life, lipophilicity, receptor-binding affinity (29), or tissue retention between the two drugs may exist. The presently reported difference in suppression of cortisol secretion may be a consequence of differences in terminal half-life between fluticasone, which ranges between 7.8 and 14.4 h (30, 31) and budesonide, the half-life which is only 2.8 h (27). Indeed, it is generally assumed that the half-life of an inhaled glucocorticosteroid is one of the prominent factors that determines the magnitude of systemic effects, in particular when repeated doses are administered (32).

Fluticasone and budesonide did not alter total WBC counts, but produced a decrease in circulating eosinophils, reaching significance with the higher dose. Similar decreases with inhaled glucocorticosteroids have been reported (33). It remains unclear, however, to what extent the reduction in eosinophils resulted from a direct systemic effect on the bone marrow rather than from an inhibitory action on the generation and release of various cytokines produced in the airways (34, 35).

Both the low doses and high doses of budesonide and fluticasone improved lung function. Significant differences between the different doses and drugs were not observed, because most patients included in the present study had mild asthma. Indeed, they exhibited a mean FEV₁ of 80% of their predicted value before any treatment was given. Moreover, a 1-wk treatment might have been too short to obtain the maximal improvement with the higher doses, as has been shown (21).

The study could have been performed in a population with more severe asthma, as these patients are more obvious candidates for a treatment with high doses of inhaled glucocorticosteroids. Although this would have provided the opportunity to better assess the comparison between efficacy and systemic effects, such patients are so dependent on inhaled glucocorticosteroids that they could not have sustained a withdrawal of inhaled glucocorticosteroids during run-in and washout periods.

In conclusion, we have found no evidence of systemic activity of low doses of budesonide or fluticasone in patients with asthma. At the maximum recommended therapeutic doses, both budesonide and fluticasone (particularly the latter) suppressed adrenal function, but apparently to a lower degree than was reported in healthy volunteers. Whether the degree of adrenal suppression and the difference in suppression between the two drugs are important in terms of development of clinically relevant long-term systemic side effects, such as osteoporosis, glaucoma, cataract, or easy bruising of the skin, remain to be investigated.