Rationale: To evaluate the association between asthma exacerbations and the decline in lung function, as well as the potential effects of an inhaled corticosteroid, budesonide, on exacerbation-related decline in patients with asthma.
Objectives: To determine whether severe asthma exacerbations are associated with a persistent decline in lung function.
Methods: The START (inhaled steroid treatment as regular therapy in early asthma) study was a 3-year, randomized, double-blind study of 7,165 patients (5–66 yr) with persistent asthma for less than 2 years, to determine whether early intervention with low-dose inhaled budesonide prevents severe asthma-related events (exacerbations requiring hospitalization or emergency treatment) and decline in lung function.
Measurements and Main Results: There were 315 patients who experienced at least one severe asthma exacerbation, of which 305 were analyzable, 190 in the placebo group and 115 in the budesonide group. In the placebo group, the change in post-bronchodilator FEV1 % predicted from baseline to the end of the study, in patients who did or did not experience a severe exacerbation was −6.44% and −2.43%, respectively (P < 0.001). A significant difference was seen in both children and in adults, but not in adolescents. In the budesonide group, the change in the post-bronchodilator FEV1 % predicted in patients who did or did not experience a severe exacerbation was −2.48% and −1.72%, respectively (P = 0.57). The difference in magnitude of reduction afforded by budesonide, in patients who experienced at least one severe asthma-related event compared with those who did not, was statistically significant (P = 0.042).
Conclusions: Severe asthma exacerbations are associated with a more rapid decline in lung function. Treatment with low doses of inhaled corticosteroid is associated with an attenuation of the decline.
Clinical trial registered with www.clinicaltrials.gov (NCT00641914).
Patients with asthma can lose lung function more rapidly than patients without asthma. The reason why this occurs in some patients and not others is unknown.
This study identified that severe asthma exacerbations are associated with a more rapid loss in lung function, and that this did not occur in patients taking inhaled corticosteroids.
Inhaled corticosteroids (ICS) are considered the mainstay of asthma treatment (8). There has, however, been controversy about their ability to alter the course of asthma; some studies demonstrating that early intervention with ICS provides long term benefit (9, 10), whereas others have failed to show any benefit of asthma treatment with ICS on changes in lung function over time (4), or on the persistence of symptoms once the medication is discontinued (11). However, the potential association between exacerbations and decline in lung function in patients with asthma and the influence of ICSs on exacerbation-induced accelerated decline have not yet been established in a prospective, controlled trial.
The START (inhaled steroid treatment as regular therapy in early asthma) study is a large, worldwide, long-term, double-blind, placebo-controlled trial, which was undertaken to determine whether early intervention with low doses of the ICS, budesonide, in patients with newly diagnosed, persistent asthma would prevent severe asthma-related events (SAREs) and modify the decline in lung function (12). The main results of the START trial have been reported previously (13). Both the occurrence of severe asthma exacerbations and the decline in post-bronchodilator FEV1 were significantly reduced by early intervention with budesonide. The START trial also provided a unique opportunity to evaluate the association between asthma exacerbations and the decline in lung function, and the potential effects of budesonide on exacerbation-related accelerated decline in patients with asthma in a longitudinal, prospective study.
The study design, methods, and inclusion and exclusion criteria of the START trial were previously described in detail (12). The study enrolled 7,241 patients aged 5–66 years from 32 countries. Patients had asthma symptoms weekly, but not daily, during the 3 months prior to study. These symptoms had to have been present for less than 2 years (ideally <1 yr). Patients had to demonstrate reversible airway obstruction by an increase in FEV1 of more than 12% after a β2-agonist, a fall in FEV1 of more than 15% after exercise, or a variation of more than 15% in peak expiratory flow rates over 14 days. The study was approved by all the participating institution's research ethics boards prior to initiation, and all subjects provided informed consent before being enrolled in the study.
The patients were randomly assigned to receive either once-daily budesonide (Pulmicort; AstraZeneca, Lund, Sweden) or placebo, delivered from the dry powder inhaler Turbuhaler (AstraZeneca, Sweden), during a 3-year double-blind study. The daily dose of budesonide was 200 μg in children aged younger than 11 years at randomization, and 400 μg in the others. Placebo was lactose. Changes in concurrent medication, including introducing inhaled or systemic corticosteroids, could be made during the study, at the investigator's discretion, to achieve asthma control.
The primary outcome was time to the first SARE, defined as events requiring hospitalization or emergency treatment due to worsening of asthma, or death due to asthma. Emergency treatment was defined as treatment of acute airway obstruction with systemic corticosteroids and nebulized or parenteral bronchodilators given at a healthcare institution. This most stringent definition of a severe asthma exacerbation was selected to ensure uniformity in the 32 countries enrolling patients in the trial. Information about asthma-related events was collected at each study visit. Patients recorded details of asthma-related events and asthma control between scheduled visits in a diary. This information included the need for courses of systemic corticosteroids (oral or parenteral) for asthma exacerbations that did not require a hospitalization or emergency room visit. At each visit, all asthma-related events were recorded by the investigators from the patient's diaries. Patients were not discontinued from the study if they experienced a SARE. Spirometry was performed using a MicroLoop II spirometer (Micro Medical, Rochester, UK). The post-bronchodilator FEV1 was measured at randomization after 6 and 12 weeks and then quarterly, 30 minutes after inhaling terbutaline, 0.5 mg via Turbuhaler (Bricanyl; AstraZeneca, Sweden) or 1 mg via pMDI (Breathaire, Novartis, Basel, Switzerland). The prebronchodilator FEV1 was measured at randomization and then at each yearly visit. Patients were asked to refrain from short-acting inhaled β2-agonists for at least 6 hours, and oral or long-acting inhaled β2-agonists for at least 24 hours before these measurements.
Predicted normal values of FEV1 were calculated based on gender, age, and height at each visit. For males 16.0 years or younger, and for females 15.0 years or younger, the prediction formulas of Quanjer and colleagues (14) were used, and race correction factors of 1.00, 0.91, 0.87, and 0.88 were applied for white, Asian, black, and other, respectively. For males older than 18.0 years, and females older than 17.0 years, predicted values were calculated from the official statement of the European Respiratory Society (15), and the race correction factors of 1.00, 0.90, 0.87, and 0.85 were applied for white, Oriental, black, and other, respectively. In the age range of 16.0–18.0 years for males and 15.0–17.0 years for females, predicted normal values were computed by linear (in age) interpolation between the two formulae.
Patients were divided into two strata depending on the occurrence of SAREs during the treatment phase: those with no SAREs, and those with at least one SARE. The 3-year mean change from baseline in post-bronchodilator FEV1 % predicted in the two treatment groups was estimated, separately within each stratum, using a mixed effects model (16), including as covariates the randomized treatment, baseline post-bronchodilator FEV1 % predicted, geographical region, a second degree polynomial in time since baseline, and a number of interaction terms. Association between the occurrence of exacerbations and the decline in lung function was studied using a test of the difference in mean level of the change in lung function between patients with or without SAREs, with the null hypothesis that the distribution of the change in lung function is independent of the occurrence of SAREs. A statistically significant outcome of the test falsifies the null hypothesis, indicating that there is an association.
For a patient to be analyzable, it was required that there was a baseline assessment and at least one on-treatment assessment of post-bronchodilator FEV1. Measurements of post-bronchodilator FEV1 made between 14 days before and 14 days after the start of a SARE were excluded from the analysis. Otherwise, all available assessments of post-bronchodilator FEV1 were used for the analysis, without the requirement for a complete set of data for a patient. Subgroup analyses were carried out separately for each of the age groups—children, adolescents, and adults—using similar statistical models. The statistical software SAS version 8.2 (SAS Institute, Cary, NC) was used for computations.
The START study enrolled 7,241 patients, of whom 7,165 (budesonide, 3,597; placebo, 3,568) were available for analysis. From this cohort, 2,010 patients did not complete the 3 years of double-blind treatment. The dropout rate (27.5% in the budesonide group and 28.6% in the placebo group) and the mean time in study (2.47 yr in the budesonide group and 2.44 yr in the placebo group) were comparable for the two treatment arms. The analysis of the influence of asthma exacerbations on the decline in lung function presented here is based on 6,767 analyzable patients, who had a baseline assessment and at least one on-treatment assessment of post-bronchodilator FEV1 (budesonide, 3,399; placebo, 3,368) that was not made within a time interval of ±14 days around a SARE.
The total number of patients having at least one SARE during the double-blind treatment period was 315 (budesonide group, 117; placebo group, 198). This gives an odds ratio (budesonide relative to placebo) of 0.57 (95% confidence interval, 0.45–0.72; P < 0.001). However, two patients with a SARE in the budesonide group and eight patients with a SARE in the placebo group dropped out before the first on-treatment visit and measurement of post-bronchodilator FEV1, and were not analyzable. This left 115 patients in the budesonide group and 190 patients in the placebo group with at least one SARE for analysis, the corresponding odds ratio being 0.59 (95% confidence interval, 0.46–0.74; P < 0.001). No statistically significant differences in demographics or lung function at baseline were seen between patients who did or did not develop a SARE during the double-blind treatment period (Table 1). There were no asthma-related deaths during the study.
Budesonide | Placebo | |||||
---|---|---|---|---|---|---|
No SARE* | SARE† | No SARE | SARE | |||
(n = 3,284) | (n = 115) | (n = 3,178) | (n = 190) | |||
Age group, n | ||||||
Children: 5–10 yr | 926 | 38 | 873 | 60 | ||
Adolescents: 11–17 yr | 579 | 16 | 529 | 24 | ||
Adults: 18–66 yr | 1,779 | 61 | 1,776 | 106 | ||
Ethnic origin, % | ||||||
White | 65.4 | 52.2 | 65.4 | 51.0 | ||
Black | 1.3 | 1.7 | 1.6 | 1.6 | ||
Asian | 28.0 | 36.5 | 27.6 | 40.0 | ||
Other | 5.3 | 9.6 | 5.4 | 7.4 | ||
Gender, % | ||||||
Male | 45.7 | 40.0 | 46.3 | 42.6 | ||
Female | 54.3 | 60.0 | 53.7 | 57.4 | ||
Duration of asthma, % | ||||||
<3 mo | 37.1 | 29.6 | 36.1 | 24.2 | ||
3 to <6 mo | 14.4 | 14.8 | 13.7 | 10.0 | ||
6 mo to <1 yr | 15.7 | 13.9 | 17.1 | 16.8 | ||
1–2 yr | 32.9 | 41.7 | 33.1 | 49.0 | ||
No. of symptomatic daysin past 2 wk, % | ||||||
None | 8.6 | 8.7 | 8.7 | 5.3 | ||
1–3 d | 36.3 | 30.4 | 34.8 | 33.2 | ||
4–7 d | 33.6 | 33.0 | 35.9 | 38.4 | ||
>7 d | 21.5 | 27.8 | 20.6 | 23.2 | ||
Smoking status, % | ||||||
Current or previous smoker | 20.3 | 20.9 | 20.2 | 14.2 | ||
Passive smoker | 28.0 | 33.9 | 29.9 | 40.0 | ||
Nonsmoker | 51.7 | 45.2 | 50.0 | 45.8 | ||
Mean (SD) FEV1 % | ||||||
Prebronchodilator | 86.6 (13.6)‡ | 85.5 (14.2) | 86.9 (13.6) | 85.6 (13.4) | ||
Post-bronchodilator | 96.4 (12.8) | 97.1 (14.1) | 96.7 (13.0) | 96.9 (12.2) |
The mean post-bronchodilator FEV1 % predicted, visit by visit, in the various groups is shown in Figure 1. In the placebo group, the mean change in post-bronchodilator FEV1 % predicted at 3 years, in patients with at least one SARE, was −6.44% (SE, 1.03%), as compared with −2.43% (SE, 0.18%) in patients who did not experience a SARE (P < 0.001; Table 2 and Figure 2). In the budesonide group, the mean change in post-bronchodilator FEV1 % predicted in patients with at least one SARE was −2.48% (SE, 1.31%), as compared with −1.72% (SE, 0.18%) in patients who did not experience a SARE (P = 0.57; Table 2 and Figure 2). The difference in magnitude of reduction afforded by budesonide in patients who experienced at least one SARE compared with those who did not was statistically significant (P = 0.042).
No SARE* | SARE† | Difference‡ | |
---|---|---|---|
All patients | |||
Budesonide | −1.72 (0.18) | −2.48 (1.31) | +0.76; P = 0.57 |
Placebo | −2.43 (0.18) | −6.44 (1.03) | +4.01; P < 0.001 |
Children: 5–10 yr | |||
Budesonide | −2.00 (0.33) | −2.38 (1.95) | +0.38; P = 0.85 |
Placebo | −2.28 (0.34) | −6.29 (1.52) | +4.01; P = 0.010 |
Adolescents: 11–17 yr | |||
Budesonide | −0.24 (0.38) | −0.99 (2.97) | +0.75; P = 0.80 |
Placebo | +0.13 (0.40) | −2.99 (2.24) | +3.12; P = 0.17 |
Adults: 18–66 yr | |||
Budesonide | −2.03 (0.24) | −2.92 (2.02) | +0.89; P = 0.66 |
Placebo | −3.33 (0.24) | −6.78 (1.63) | +3.45; P = 0.037 |
The association between exacerbations and decline in lung function was statistically significant in placebo-treated children and adults when considered separately, but not in placebo-treated adolescents (Table 2), nor was the association statistically significant in budesonide-treated patients of any age group (Table 2).
In adult patients in the placebo group, with or without a SARE, the mean 3-year decline in post-bronchodilator FEV1 % predicted corresponds to a mean yearly absolute decline of 66 and 34 ml, respectively, whereas, in the budesonide group, with or without a SARE, the mean 3-year decline in post-bronchodilator FEV1 % predicted corresponds to a mean yearly absolute decline of 27 ad 21 ml respectively.
There were 1,363 patients who required at least one course of systemic (oral or parenteral) corticosteroids (818 in the placebo group and 545 in the budesonide group), whereas 5,404 patients did not require a course of systemic corticosteroids. The decline in the post-bronchodilator FEV1 at 3 years in placebo-treated patients requiring systemic corticosteroids was −4.66% (SE, 0.41%), whereas, for placebo-treated patients not requiring systemic corticosteroids, the decline was −1.91% (SE, 0.20%; P < 0.001; Figure 3). The decline in the post-bronchodilator FEV1 at 3 years in budesonide-treated patients requiring systemic corticosteroids was −2.44% (SE, 0.50%), whereas, for budesonide-treated patients not requiring systemic corticosteroids, the decline was −1.60% (SE, 0.19%; P = 0.12; Figure 3). There was a statistically significant difference in the magnitude of reduction afforded by budesonide in patients who required systemic corticosteroids compared with those who did not (P = 0.006).
This post hoc analysis of data from a large, randomized, controlled, long-term study with repeated measures of post-bronchodilator FEV1, and accurate recording of SAREs, provides insights into the relationship between asthma exacerbations, decline in post-bronchodilator FEV1 over time, and the possible effects of ICSs on this relationship in a large group of patients with recently diagnosed, persistent asthma. The results demonstrate that severe asthma exacerbations are associated with a decline in post-bronchodilator FEV1, and this effect was seen in adults and children, but not in adolescents. In addition, the decline in FEV1 associated with severe exacerbations was not seen in patients treated with low doses of inhaled budesonide.
Severe exacerbations and decline in FEV1 may be both manifestations of a more severe asthma phenotype; thus, the association of these two manifestations of asthma may not be causal. It is important to note, however, that all patients recruited to the START study were thought to suffer from mild, persistent asthma. In addition, patients who experienced exacerbations had identical baseline pre- and post-bronchodilator FEV1 to those without any exacerbations. In addition, although there were slight differences in the populations that experienced SAREs (slightly longer disease duration, higher exposure to passive smoking, slightly higher number of symptom days), there were no significant lung function or demographic characteristics that suggested that patients with exacerbations would have more severe asthma before randomization than patients without exacerbations. Finally, similar trends were seen in patients with SAREs and those (many more) patients who had exacerbations requiring oral corticosteroids. However, in the absence of another very large and long-term prospective study examining the possible causal relationship between severe asthma exacerbations and decline in lung function, this must remain a hypothesis.
Several cohort studies and cross-sectional studies in children and adults have suggested that, over time, groups of patients with asthma lose lung function at a greater rate than subjects without asthma, and that increased loss of lung function is not seen in all patients (1–4). The effects of asthma exacerbations and ICSs on the rate of decline have been less well studied. A recent retrospective analysis of the data from 93 adult patients with asthma who had been followed at regular intervals for several years without treatment with ICSs found a 17-ml increase in annual decline in patients with more than 0.1 severe asthma exacerbations per year (6). Compared with the patients in the START study, these patients appeared to have somewhat more severe asthma, which had persisted for a much longer duration. Different patient populations and different definitions of a severe asthma exacerbation in the two studies make a direct comparison of the effect on lung function difficult. The annual declines in both the current and previous studies (6) were lower than the 51 ml per year reported by Lange and colleagues (1) in untreated adult patients with asthma. The latter study did not report the frequency or severity of exacerbations.
In patients who received placebo during the START study, a decline in post-bronchodilator FEV1 was measured at the first (6-wk) visit after randomization. This effect was greater in those patients who experienced SAREs. The most reasonable explanation of this effect is regression to the mean following the spirometric measurements made at baseline to enroll patients in the study. This effect has also been seen in another long-term study comparing placebo with budesonide in patients with chronic obstructive pulmonary disease (18). However, a similar pattern was not seen in those patients randomized to receive budesonide.
In the whole START population, the decline in post-bronchodilator FEV1 % predicted in children in the placebo group was less—at 0.8% per year—than in adults on placebo, who had a decline of 1.2% per year (13). Furthermore, the effect of budesonide on the annual decline was greater in adults, who received 400 μg/day, than in children, who received 200 μg/day (19). In contrast, the annual declines in placebo-treated patients with exacerbations were very similar in children and adults. Furthermore, the effects of acute exacerbations on lung function persisted for the duration of the study in both children and adults. This is in agreement with the findings of some smaller studies on children, which did not report exacerbations (20–22), as well as the findings in birth cohort studies, which have demonstrated loss of lung function in children soon after the onset of asthma (3). The effect of exacerbations on lung function decline was much less striking in adolescents, where the decline in post-bronchodilator FEV1 in placebo-treated patients with severe exacerbations was −2.99%, as compared with −6.29% and −6.78% in children and adults, respectively. This is consistent with observations made in the Childhood Asthma Management Program (CAMP) trial, which did not find any appreciable decline in FEV1 in a population of children and adolescents over time (17); howevever, a subsequent analysis identified that a substantial proportion (25.7%) did have a significant reduction in post-bronchodilator FEV1, with significant predictors being male sex and younger age (4). In addition, this analysis (4) did not find any benefit of ICS treatment on the decline in lung function in the population of children with an accelerated decline in the CAMP study. It is not clear why there are differences between that analysis and the current results with regard to the benefits of ICS, other than the facts that most of the patients in the CAMP study were adolescents, and were not selected by the presence or absence of a severe exacerbation. In addition, the reasons that adolescents with asthma appear to behave differently from children and adults in this regard in the present study have not yet been elucidated, but might include less precision in the predicted normal values for FEV1 in adolescents, whose lung volumes are changing rapidly as they grow.
The definition of a SARE used in this study was more stringent than definitions used in most studies that measure severe asthma exacerbations. This definition required events requiring hospitalization or emergency treatment due to worsening of asthma or death due to asthma, but did not include a predetermined decline in lung function. Fortunately, there was no asthma-related death during the study. Furthermore, in the analysis of the influence of SAREs on lung function, we excluded assessments of FEV1 made around (±14 d) the start of a SARE. In a stability analysis, we defined an alternative, wider censoring interval around the start of a SARE (−14 to + 84 d). In the SARE stratum, the 3-year change in post-bronchodilator FEV1 % predicted was −6.74% (SE, 1.04%) in placebo-treated patients and −1.93% (SE, 1.32%) in budesonide-treated patients, results almost identical to those from our main analysis (Table 2).
The post-bronchodilator FEV1 values were selected to reflect lung function loss in the START study as in other studies following lung function in patients with asthma over time (3, 4, 17). This measurement was made at every study visit, whereas reversibility measurements (prebronchodilator FEV1 and post-bronchodilator FEV1 after an inhaled β2-agonist) were taken only at the yearly visits. Because of the size of the study and the large number of study centers (>500) where standardization would need to be undertaken, more complicated measurements of lung function or measurements of airway responsiveness were not made.
The use of oral corticosteroids, or a fixed decline in peak expiratory flow, which are added to the definition of severe asthma exacerbations in many studies (23, 24), were not included in the definition in START. As a result of this, the number of SAREs was much lower, and the events more severe, than in studies that include additional components to the definition. However, information about exacerbations requiring the use of systemic corticosteroids (oral or parenteral) was also collected in an asthma diary. The fact that these presumably less severe exacerbations were also associated with an accelerated decline in lung function supports the association between SAREs and accelerated decline in lung function. Therefore, asthma exacerbations in general may be an important contributor to the accelerated decline in lung function that occurs over time in groups of both children and adults with asthma.
The design of the START study allowed for additional medications to be added to the study medications at any time during the study. This was done to avoid the ethical concern of keeping patients on placebo for the 3 years in the study if their asthma control was worsening, and to maintain the highest possible enrollment in the study. One disadvantage of this decision is that ICSs were added to a relatively high percentage of the placebo-treated patients (23.6%) when compared with the budesonide-treated patients (12.5%). This means that any treatment effect of the study medication is likely to diminish as the study progresses.
This post hoc analysis of the START study provides two insights: that asthma exacerbations may contribute to the accelerated decline in lung function that occurs over time in both children and adults with asthma; and that treatment with ICSs, early after diagnosis, not only reduces the risks of a severe exacerbation, but is associated with an attenuation of the decline in lung function associated with an exacerbation.
1. | Lange P, Parner J, Vestbo J, Schnohr P, Jensen G. A 15-year follow-up study of ventilatory function in adults with asthma. N Engl J Med 1998;339:1194–1200. |
2. | Peat JK, Woolcock AJ, Cullen K. Rate of decline of lung function in subjects with asthma. Eur J Respir Dis 1987;70:171–179. |
3. | Sears MR, Greene JM, Willan AR, Wiecek EM, Taylor DR, Flannery EM, Cowan JO, Herbison GP, Silva PA, Poulton R. A longitudinal, population-based, cohort study of childhood asthma followed to adulthood. N Engl J Med 2003;349:1414–1422. |
4. | Covar RA, Spahn JD, Murphy JR, Szefler SJ. Progression of asthma measured by lung function in the childhood asthma management program. Am J Respir Crit Care Med 2004;170:234–241. |
5. | Lee JH, Haselkorn T, Borish L, Rasouliyan L, Chipps BE, Wenzel SE. Risk factors associated with persistent airflow limitation in severe or difficult-to-treat asthma: insights from the TENOR study. Chest 2007;132:1882–1889. |
6. | Bai TR, Vonk JM, Postma DS, Boezen HM. Severe exacerbations predict excess lung function decline in asthma. Eur Respir J 2007;30:452–456. |
7. | Donaldson GC, Seemungal TA, Bhowmik A, Wedzicha JA. Relationship between exacerbation frequency and lung function decline in chronic obstructive pulmonary disease. Thorax 2002;57:847–852. |
8. | Bateman ED, Hurd SS, Barnes PJ, Bousquet J, Drazen JM, FitzGerald M, Gibson P, Ohta K, O'Byrne P, Pedersen SE, et al. Global strategy for asthma management and prevention: GINA executive summary. Eur Respir J 2008;31:143–178. |
9. | Agertoft L, Pedersen S. Effects of long-term treatment with an inhaled corticosteroid on growth and pulmonary function in asthmatic children. Respir Med 1994;88:373–381. |
10. | Haahtela T, Järvinen M, Kava T, Kiviranta K, Koskinen S, Lehtonen K, Nikander K, Persson T, Selroos O, Sovijärvi A, et al. Effects of reducing or discontinuing inhaled budesonide in patients with mild asthma. N Engl J Med 1994;331:700–705. |
11. | Guilbert TW, Morgan WJ, Zeiger RS, Mauger DT, Boehmer SJ, Szefler SJ, Bacharier LB, Lemanske RF Jr, Strunk RC, Allen DB, et al. Long-term inhaled corticosteroids in preschool children at high risk for asthma. N Engl J Med 2006;354:1985–1997. |
12. | Pauwels RA, Busse WW, O'Byrne PM, Pedersen S, Tan WC, Chen YZ, Ohlsson SV, Ullman A. The inhaled Steroid Treatment as Regular Therapy in early asthma (START) study: rationale and design. Control Clin Trials 2001;22:405–419. |
13. | Pauwels RA, Pedersen S, Busse WW, Tan WC, Chen YZ, Ohlsson SV, Ullman A, Lamm CJ, O'Byrne PM; START Investigators Group. Early intervention with budesonide in mild persistent asthma: a randomised, double-blind trial. Lancet 2003;361:1071–1076. |
14. | Quanjer PH, Stocks J, Polgar G, Wise M, Karlberg J, Borsboom G. Compilation of reference values for lung function measurements in children. Eur Respir J Suppl 1989;4:184S–261S. |
15. | Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventilatory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. Eur Respir J Suppl 1993;16:5–40. |
16. | Brown H, Prescott R. Applied mixed models in medicine. London: John Wiley and Sons Ltd.; 2001. |
17. | The Childhood Asthma Management Program Research Group. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med 2000;343:1054–1063. |
18. | Pauwels RA, Löfdahl CG, Laitinen LA, Schouten JP, Postma DS, Pride NB, Ohlsson SV. Long-term treatment with inhaled budesonide in persons with mild chronic obstructive pulmonary disease who continue smoking. European Respiratory Society Study on Chronic Obstructive Pulmonary Disease. N Engl J Med 1999;340:1948–1953. |
19. | O'Byrne PM, Pedersen S, Busse WW, Tan WC, Chen YZ, Ohlsson SV, Ullman A, Lamm CJ, Pauwels RA. Effects of early intervention with inhaled budesonide on lung function in newly diagnosed asthma. Chest 2006;129:1478–1485. |
20. | Verberne AA, Frost C, Roorda RJ, van der Laag H, Kerrebijn KF. One year treatment with salmeterol compared with beclomethasone in children with asthma. The Dutch Paediatric Asthma Study Group. Am J Respir Crit Care Med 1997;156:688–695. |
21. | Merkus PJ, van Pelt W, van Houwelingen JC, van Essen-Zandvliet LE, Duiverman EJ, Kerrebijn KF, Quanjer PH. Inhaled corticosteroids and growth of airway function in asthmatic children. Eur Respir J 2004;23:861–868. |
22. | Nuijsink M, Hop WC, Sterk PJ, Duiverman EJ, De Jongste JC. Long-term asthma treatment guided by airway hyperresponsiveness in children: a randomised controlled trial. Eur Respir J 2007;30:457–466. |
23. | O'Byrne PM, Bisgaard H, Godard PP, Pistolesi M, Palmqvist M, Zhu Y, Ekström T, Bateman ED. Budesonide/formoterol combination therapy as both maintenance and reliever medication in asthma. Am J Respir Crit Care Med 2005;171:129–136. |
24. | Pauwels RA, Löfdahl CG, Postma DS, Tattersfield AE, O'Byrne P, Barnes PJ, Ullman A. Effect of inhaled formoterol and budesonide on exacerbations of asthma. Formoterol and Corticosteroids Establishing Therapy (FACET) International Study Group. N Engl J Med 1997;337:1405–1411. |