Systematic Review and Meta-Analysis of Budesonide/Formoterol in a Single Inhaler

By Edwards, Steven J Gruffydd-Jones, Kevin; Ryan, Dermot P

Key words: Asthma – Budesonide – Formoterol – Meta-analysis – Systematic review ABSTRACT

Objective: To compare the effectiveness of budesonide/formoterol using fixed dosing (BUD/FORM) with inhaled corticosteroid (ICS) alone or alternative ICS and long-acting beta^sub 2^-agonist (LABA) regimens for adults with moderate/severe asthma.

Methods: BIOSIS, CENTRAL, EMBASE and MEDLINE were searched for abstracts and papers. All searching was completed in July 2006. No restriction was placed on language. Meta-analysis of randomised controlled trials (RCTs) using a fixed effects model. RCTs were included if the comparator with BUD/FORM had an equivalent daily dose of ICS at the start of the trial. The primary outcome measure was, ‘treatment failure’, defined as: asthma-related serious adverse event, oral glucocorticosteroid treatment, A&E visit and/or admission to hospital, withdrawal due to a need for additional asthma therapy.

Results: Of the 330 papers identified in the literature search, 15 met the inclusion criteria. The following alternative treatments were found: ICS alone (BUD), BUD/FORM adjustable maintenance dose (BUD/FORM-AMD), and salmeterol/fluticasone in a single inhaler (SALM/ FP). Meta-analysis of treatment failure demonstrated a 50% increase with BUD versus BUD/FORM (Relative Risk [RR] 1.50, 95% confidence interval [95% CI]: 1.12-2.02, p = 0.007; 2 RCTs); a trend in favour of a reduction with BUD/FORM-AMD versus BUD/FORM (RR 0.88, 95% CI: 0.77-1 .02, p = 0.09; 11 RCTs); and no evidence of a difference with SALM/FP versus BUD/FORM (RR 0.99, 95% CI: 0.831.16, p= 0.86; 3 RCTs). Significant heterogeneity was not detected in the primary analyses. Secondary analyses demonstrated the following significant differences: hospitalisations/A&E visits (49% increased risk with SALM/FP vs. BUD/FORM, RR 1.49, 95% Cl: 1 .07-2.08, p = 0.02, and 28% reduced risk with BUD/FORM-AMD vs. BUD/FORM, RR 0.72, 95% Cl: 0.52- 0.99, p= 0.04); and use of oral steroids (51% increase in risk with BUD vs. BUD/FORM, RR 1.51, 95% CI: 1.10-2.09, p= 0.01, and 19% reduced risk with BUD/FORM-AMD vs. BUD/FORM, RR 0.81 , 95% CI: 0.70- 0.95, p= 0.01).

Conclusions: Fixed-dose BUD/FORM is an effective treatment option for adult patients with moderate/ severe asthma when compared to BUD and SALM/FP, with adjustable maintenance dosing demonstrating important advantages over fixed dosing in relation to exacerbation prevention and reduced treatment load.

Introduction

The joint British Thoracic Society (BTSjVScottish Intercollegiate Guidelines Network (SIGN) British Guideline for the Management of Asthma categorises asthma in a step-wise manner, with a higher step suggesting more severe asthma. Appropriate treatments are recommended at each step1.

Various outcome measures are collected as part of randomised controlled trials comparing different treatments for asthma. However, exacerbations have a profound effect on patients with asthma compared to outcomes such as changes in lung function measures and nocturnal awakenings, leading to urgent or emergency contact with clinicians (whether in primary or secondary care) and may cause hospitalisation or even death2.

While all patients with asthma are at risk of an exacerbation, it has been shown that the risk increases as the severity of asthma increases, as indicated by which ‘step’ of the BTS/SIGN guideline applies to the asthmatic3. For patients with moderate/severe asthma (step 3 in the guideline) the addition of a long-acting beta^sub 2^- agonist (LABA) to an inhaled corticosteroid (ICS) in a combination inhaler is one option for patients uncontrolled on low-dose ICS alone1.

In the UK there are currently two different combination inhalers available: Seretide/Advair (GlaxoSmithKline) – salmeterol/ fluticasone (SALM/FP) in a single inhaler and Symbicort (AstraZeneca) – budesonide/formoterol (BUD/FORM) in a single inhaler.

Traditional use of inhaled treatments for asthma involves a fixed dosing regimen, where a patient takes a fixed number of inhalations per day. However, the combination of BUD/FORM is approved in a dosing range that enables a flexible dosing regimen (termed adjustable maintenance dosing or AMD) as part of an asthma management plan. This allows patients to step up treatment during periods where their asthma has deteriorated and step down to the lowest recommended dose of 2 inhalations/day to maintain adequate control4.

The aim of this systematic review is to quantify the benefit of BUD/FORM used in a traditional fixed dosing regimen compared to the other preventative inhaled treatments for adults with moderate/ severe asthma.

Methods

Searches

BIOSIS, CENTRAL, EMBASE and MEDLINE were searched for abstracts and papers. All searching was completed in July 2006. The exact search strategy had to be modified to comply with the functionality of the individual databases but the following search terms were common: budesonide (and the brand name – Pulmicort [AstraZeneca]); formoterol or eformoterol, (and the brand names Foradil [Novartis Pharmaceuticals]) and Oxis [AstraZeneca]); the combination device brand name of BUD/FORM, Symbicort; and clinical trial (or clinical trial, phase III or clinical trial, phase IV or controlled clinical trial or randomized controlled trial).

No restriction was placed on language in the search and it was planned that, where possible, full translations of non-English papers would be sought.

Trial selection

The systematic review was developed to compare BUD/FORM (as 200/ 6 [mu]g or 100/6 [mu]g) used in a traditional fixed dosing regimen with the other preventative treatments for adults with moderate/ severe asthma. These alternative treatments were defined as follows:

(i) ICS – the daily dose of ICS in the trial had to be equivalent to the daily dose of ICS in BUD/FORM;

(ii) SALM/FP – the daily dose of ICS in the trial had to be equivalent to the daily dose of ICS in BUD/ FORM (e.g. 500 [mu]g FP would be considered equal to 800 [mu]g BUD in the BTS/SIGN guideline1);

(iii) BUD/FORM – BUD/FORM can also be used with an adjustable maintenance dosing (BUD/ FORM-AMD) regimen as part of an asthma plan (where the number of inhalations could vary from 2 to 8 inhalations per day); the two regimens could be compared providing the same strengths of BUD/FORM were used in the trial.

The BTS/SIGN guidelines state that the monocomponents (BUD and FORM) when used in separate devices are considered to offer similar effectiveness to the combination device1. As such, any trial using the monocomponents at a daily dose that was achievable with the combination device was considered potentially appropriate for inclusion in the review.

The other selection criteria for inclusion in the review were: randomised controlled trial; in adult patients (i.e. aged 16 years or above); with moderate/severe asthma; with any additional treatment being given equally in the two randomised groups.

Quality assessment

The methodological quality of trials was assessed using the criteria devised by Jadad et al.5 (called the ‘Jadad Score’ hereinafter). The elements appraised include: trial design, level of blinding, method of randomisation and an adequate account of patients lost to follow up. A Jadad Score of >/= 3 was required for a trial to be considered of ‘good’ quality and included in the analysis.

Data extraction

One reviewer extracted data before being independently validated by another reviewer on the following outcomes:

* treatment failure – defined as one or more of the following – serious asthma exacerbation leading to use of non-trial medication (excluding a short course of oral steroids); hospitalisation due to deterioration in asthma; emergency treatment (e.g. nebulised beta^sub 2^-agonist, glucocorticosteroid injection) due to deterioration in asthma; need for an oral course of steroids due to a deterioration in asthma; lack of efficacy necessitating a change in asthma medication and withdrawal from the trial;

* hospitalisations and/or accident and emergency (A&E) room visits (due to deterioration in asthma);

* use of oral corticosteroids (i.e. need for an oral course of steroids lasting at least 5 days due to a deterioration in asthma);

* withdrawal;

* withdrawal due to adverse events.

Where treatment failure was not reported in a trial, the outcome of ‘serious exacerbation’ was used as a surrogate measure as this is a composite end point with a similar definition to treatment failure. Data were also sought from clinical trial websites, e.g. additional data on hospitalisations for Dahl et al.6 were obtained from www.ctr.gsk.co.uk.

For the BUD/FORM and BUD/FORM-AMD comparison the weighted mean number of BUD/ FORM inhalations was also calculated as the adjustable dosing regimen allows for both lower and higher maintenance doses to be delivered to patients than with the fixed dosing regimen. To ensure that the use of lower doses of BUD/FORM, in periods without asthma worsenings, when using the AMD strategy, were not associated with a loss of daily asthma control, we also calculated the weighted mean number of short-acting beta^sub 2^- agonist (SABA) inhalations/day (terbutaline 500 pg or equivalent) for the two treatment groups. This acted as a surrogate measure of daily symptom control when using the two alternative BUD/FORM regimens.

Figure 1. Results of a search of BIOSIS, CENTRAL, EMBASE and MEDLINE, for randomised controlled triab comparing BUD/FORM with any other asthma treatment Quantitative data synthesis

The meta-analyses were based on the full analysis set, as defined in ICH E9 Guidelines7. This means that all patients with data after randomisation were included in the analyses. Each patient was assigned to treatment according to the randomisation schedule.

All analyses were done using RevMan version 4.2.4(8). The meta- analyses used a fixed effects model using the Mantel-Haenszel method9. A fixed effects model was used rather than a random effects model in the primary analysis because, in a fixed effects model, the weight given to individual trials favours larger trials rather than giving a more equal weighting to smaller trials as in a random effects model10.

Assessment of heterogeneity

A chi-square test was carried out to investigate possible heterogeneity in the individual analyses.

Sensitivity analyses

As a sensitivity analysis the effect of using a random effects model using the DerSimonian and Laird method11 for the meta- analyses rather than a fixed effects model was performed.

Results

Trial flow

The implementation of the search strategy produced the results presented in Figure 1.

Trial characteristics

Of the 15 randomised controlled trials identified (BUD/FORM compared to: ICS – 12,13; SALM/FP – 3(6,14,15); BUD/FORM-AMD – 11(14,16-25), all had a Jadad Score >/= 3. The trials identified and a breakdown of the Jadad Score assessment can be found in Table 1 . The trials comparing BUD/FORM with ICS12,13 and two of the trials comparing BUD/FORM with SALM/FP6,15 were double blind, double dummy. The remaining trials were all open in design14,16-25. For trials comparing BUD/FORM-AMD with BUD/FORM, an open design was unavoidable as the trials were comparing different management strategies (a self- management strategy and fixed dosing, respectively).

Quantitative data synthesis

Primary analysis results

Two RCTs identified for inclusion into the analysis compared BUD/ FORM with BUD. BUD increased the risk of treatment failure compared to BUD/FORM by 50% (RR 1.50; 95% confidence interval [CI]: 1.122.02, p = 0.007). There is no evidence of a difference in treatment failure with fixed-dose BUD/FORM compared to SALM/FP (RR 0.99; 95% CI: 0.83-1.16, ? = 0.86). A small trend in favour of BUD/FORMAMD for reducing treatment failure compared to fixed-dose BUD/FORM (RR 0.88; 95% CI: 0.77-1.02, ? = 0.09) was observed. The trend would appear to be most affected by trial AD-039-000124. This is the only trial where excluding it shifts the positive trend into statistical significance (RR 0.86, 95% CI: 0.74-0.99, p= 0.04). Removing any other single trial does not change the direction of the trend. This meta-analysis is presented in Figure 2.

Secondary analysis

Hospitalisations/A&E visits were used as an indicator of a severe exacerbation requiring Secondary Care intervention. The comparison of BUD/FORM to BUD showed no significant difference between the groups; however, there was a numerical increase seen in patients using BUD alone (RR 1.12; 95% CI: 0.57-2.22, p= 0.74). Use of SALM/ FP compared with fixed-dose BUD/FORM increased the risk of exacerbations requiring hospitalisations/A&E visits by 49% (RR 1.49; 95% CI: 1.07-2.08, p= 0.02). BUD/FORM-AMD was also associated with a 28% reduction in the risk of hospitalisations/A&E visits compared with fixed-dose BUD/FORM (RR 0.72; 95% CI: 0.52-0.99, ? = 0.04). This meta-analysis is presented in Figure 3.

Oral steroids were used as an indicator of Primary Care intervention for patients experiencing a severe exacerbation. Patients taking BUD were significantly more likely to require oral steroids than patients using BUD/FORM (RR 1.51; 95% CI: 1.10-1.09, ? = 0.01). There is no evidence of a difference in the number of exacerbations requiring oral steroid use with BUD/ FORM compared to SALM/FP (RR 0.94; 95% CI: 0.79-1.1 1, p = 0.46). However, patients taking BUD/ FORM-AMD required oral steroids less than patients using BUD/FORM (RR 0.81; 95% CI: 0.70-0.95, ? = 0.01). This meta-analysis is presented in Figure 4.

In terms of withdrawals from treatment, there is no significant difference with BUD/FORM compared to any of the alternative asthma treatments, see the meta-analysis presented in Figure 5. Similarly, in the meta-analysis of withdrawals from treatment due to adverse events, there is no significant difference with BUD/FORM compared to any of the alternative asthma treatments, see the meta-analysis presented in Figure 6.

A summary of the results from the primary and secondary analyses is presented in Table 2.

Publication bias

Publication bias is potentially a serious problem in a systematic review, since clinical trials that produce significant results are more likely to be published than trials that show no difference in treatment26-30. The possibility of publication bias was assessed visually by creating funnel plots for the primary outcome measure of treatment failure. The limited number of trials in each of the comparisons makes assessment of publication bias difficult, however, there was no apparent asymmetry in the funnel plots.

Assessment of heterogeneity

There was no evidence of statistically significant heterogeneity in the primary analysis of treatment failure (Figure 2, all p > 0.05). The only secondary analysis that failed the test for homogeneity was the use of oral steroids in the BUD/FORM comparison with BUD/FORM-AMD (p = 0.02).

Sensitivity analysis

The effect of using a random effects model using the DerSimonian and Laird method for the primary analyses made small numerical differences to the overall estimates but did not change their direction (Table 3). The only significant result that was made non- significant by this change in analysis was the use of oral steroids in the BUD/FORM comparison with BUD/FORM-AMD.

Table 1. Randomised coniivlled trials comparing budesonide/ fornwteivl (BUD/FORM) in a single inhaler compared to preventative treatments for adults with moderate/severe asthma

Table 2. Summary of all meta-analyses comparing budesonide/ formoterol (BUD/FORM) in a single inhaler compared to preventative treatments for adults with moderate/severe asthma using a fixed effects model

Figure 2. Meta-analyses of treatment failure in alternative therapies compared to BUD/FORM in the treatment of patients with asthma

As an additional sensitivity analysis, the effect of excluding the two unpublished trials that were identified was explored. This only affected the BUD/ FORM comparison with BUD/FORM-AMD. Similarly, this made small numerical differences to the overall effect estimates and did not change the significance or non-significance of any outcomes except treatment failure (Table 4). Here BUD-FORM-AMD was associated with a significant reduction in treatment failure compared to BUD-FORM (RR 0.85, 95% CI: 0.73-0.99; ? = 0.03).

Figure 3. Meta-analyses of hospitalisations /A&E visits in alternative therapies compared to BUD/FORM in the treatment of patients with asthma

Figure 4. Meta-analyses of use of oral steroids in alternative therapies compared to BUD/FORM in the treatment of patients with asthma

Figure 5. Meta-analysis of withdrawals in alternative therapies compared to BUD/FORM in the treatment of patients with asthma

Figure 6. Meta-analyses of withdrawals due to adverse events in alternative therapies compared to BUD/FORM in the treatment of patients with asthma

Table 3. Summary of all meta-analyses comparing budesonide/ formoterol (BUD/FORM) in a single inhaler compared to preventative treatments for adults with moderate/severe asthma using a random effects model

Table 4. Summary of all meta-analyses comparing budesonide/ formoterol (BUD/FORM) fixed dosing compared to adjustable maintenance dosing (BUD/FORM-AMD) for adults with moderate/severe asthma based only on published trials

Treatment use

For the BUD/FORM and BUD/FORM-AMD comparison the available data for SABA inhalations per day and for BUD/FORM inhalations per day were also evaluated. These are detailed in Tables 5 and 6, respectively. The weighted mean number of BUD/FORM inhalations per day in the fixed-dose and adjustable dose comparison were: fixed- dose BUD/FORM 3.85 (SD 0.20) and BUD/FORM-AMD 2.78 (SD 0.34) producing a weighted mean difference in favour of BUD/FORM-AMD (WMD – 1.07; 95% CI: -1.08 to -1.06, p < 0.00001). This is equivalent to an overall reduction of 214/6.5 pg per day of metered dose budesonide and formoterol with BUD/FORM-AMD. The variability captures intertrial variability as individual patient-level data were unavailable.

The weighted mean number of SABA inhalations per day in the fixed dose and adjustable dose comparison were similar: BUD/FORM 0.55 (SD 0.55) and BUD/ FORM-AMD 0.49 (SD 0.49) producing a small weighted mean difference of 1 SABA inhalation per 20 treatment days in favour of BUD/FORM-AMD (WMD -0.06; 95% CI: -0.08 to -0.04, p < 0.00001). The variability captures inter-trial variability as individual patient-level data were unavailable.

Discussion

A long-term objective of treating patients with asthma is to prevent exacerbations utilising a minimum of medications, particularly the lowest ICS load’ . In the UK a recent 6-month survey of prescription records in GP practices confirmed that, 14- 20% of patients prescribed ICS/LABA therapy as either a single or separate inhalers experience an exacerbation requiring oral corticosteroids in a 6-month treatment period31. This meta-analysis confirms that patients taking BUD/ FORM are at significantly less risk of treatment failure and of using oral steroids than patients taking BUD alone at a similar daily dose.

Table 5. Mean number of inhalations of SABA (terbutaline 500 [mu]g or equivalent) in the trials comparing BUD/FORM with BUD/FORM- AMD

Table 6. Mean number ofinhahtions of BUD/FORM in the trials comparing fixed dosing and adjustable dosing

In contrast to recent safety concerns associated with the use of LABA as monotherapy in the USA32,33, asthma patients using BUD/FORM did not demonstrate an increased risk of requiring hospitalisations/ A&E visits confirming the safety of formoterol when used in combination with budesonide. Moreover, the increased efficacy of BUD/ FORM versus BUD was not accompanied by any significant difference in withdrawals due to adverse events. These results are in broad agreement with several recent meta-analyses performed to show adding LABA to ICS as separate or single inhaler combinations reduces exacerbations and improves asthma control compared to ICS alone34,35,36. This analysis showed no evidence of a significant difference between fixed-dose BUD/FORM and SALM/FP on treatment failure and oral corticosteroid use based on the outcome of 3 separate randomised controlled trials. However, those asthmatics taking BUD/FORM who experience severe exacerbations were at significantly less risk of requiring hospitalisations than patients using a dose of SALM/FP providing similar daily symptom control6,14,15. The increase in risk seen with SALM/FP was apparent in all three trials. This dataset in over 4000 patients suggests that for the most severe forms of asthma exacerbation, clinically important differences may exist between the two fixed-dose ICS/LABA regimens.

Despite showing advantages for BUD/FORM over ICS alone or SALM/ FP, we found evidence to support the conclusion that asthmatics taking BUD/FORM-AMD, using 25% less maintenance medication than fixed-dose BUD/FORM, had 28% less risk of hospitalisations/A&E visits and 19% less risk of requiring a course of oral steroids for exacerbations, than patients treated with fixed-dose BUD/FORM. These benefits on asthma control with BUD/FORM-AMD were not accompanied by any loss of daily asthma control as assessed by the use of SABA rescue therapy or any significant difference in withdrawals due to adverse events.

Overall, BUD/FORM-AMD would appear to be clinically more effective, in patients with moderate/ severe asthma, than fixed- dose BUD/FORM given at a higher dose, suggesting that a treatment approach allowing adjustments in the dose of both LABA and ICS in response to symptoms is superior to existing management approaches that rely solely on fixed-doses of either BUD/FORM or SALM/FP irrespective of the variation in symptoms patients experience over time14.

The BUD/FORM comparison with BUD/FORMAMD provides an example of the potential dangers of publication bias when attempts to publish data are unsuccessful and so unpublished trials fail to be included in a meta-analysis. When the two unpublished trials24,25 are excluded from the primary analysis of treatment failure, an adjustable maintenance dosing strategy is significantly more effective than a fixed dosing strategy in reducing treatment failures (p = 0.03).

There are several limitations to the current analyses. Principally the fact that the primary outcome measured was not universally available in all trials compared, and an assumption had to be made to include all of the trials identified (i.e. using severe exacerbation as a surrogate for treatment failure).

While the lack of blinding in several of the trials included in the research14,16-23 could be considered a limitation of the results, it should be highlighted that it would have been inappropriate to blind patients to treatment where different styles of patient management were under investigation (e.g. in the BUD/ FORM and BUD/FORM-AMD comparisons). Blinding has also been shown to have less influence on clinical trials where the outcomes are objective rather than subjective (e.g. death or asthma exacerbation compared to pain relief or withdrawal from treatment)37. In the current research no significant differences were identified in the comparisons of withdrawal or withdrawal due to adverse events and so any bias introduced into these outcomes due to the absence of blinding is unlikely to have any effect on the conclusions drawn from the results.

In the analysis of use of oral steroids there is the potential confounding factor of double counting, i.e. the data tended to be reported as the total number of events rather than events per patient. This may be the source of the heterogeneity identified in the metaanalysis comparing BUD/FORM and BUD/FORMAMD in the use of oral steroids. Unfortunately, as individual patient-level data were unavailable this is also a potential limitation of the analysis.

The comparison of BUD/FORM to ICS alone relies on the use of trials using the monocomponents compared to BUD. Two trials have been undertaken comparing BUD/FORM with budesonide38 and fluticasone39 that would have met the inclusion criteria. However, the primary outcome measure was lung function (i.e. morning peak expiratory flow, mPEF) in both trials and treatment failure (or severe exacerbation) was not reported.

Use of monocomponents (i.e. BUD and FORM in separate inhalers) were not compared to BUD/ FORM in this analysis, as the trials available have shown similar benefits between the two treatment options3840. However, in real-life there may be issues with compliance with taking BUD alone as it does not provide patients with the same immediate relief of symptoms that FORM does and so there may be preferential use of FORM when BUD/FORM are provided as separate monocomponents. This would not necessarily be apparent within a double blind, double dummy, randomised controlled trial comparing BUD/ FORM with the monocomponents38.

Since this review was completed, BUD/FORM has been granted a European license for use as maintenance and reliever therapy for patients with moderate/severe asthma. This treatment strategy is not evaluated within this paper and further research is currently underway to assess the overall benefits of this treatment option based on its use within randomised controlled trials15,41-45.

Conclusions

Fixed-dose BUD/FORM is an effective treatment option for adult patients with moderate/severe asthma when compared to BUD and SALM/ FP, with adjustable maintenance dosing demonstrating important advantages over fixed dosing in relation to exacerbation prevention and reduced treatment load.

Acknowledgement

Declarations of interest: This research was supported by AstraZeneca. SJE is a full-time employee of AstraZeneca UK Ltd, the manufacturer of Oxis, Pulmicort and Symbicort.

The authors would like to thank Ian Naya for his help in preparing the manuscript for publication.

References

1. BTS, SIGN (British Thoracic Society/Scottish Intercollegiate Guideline Network/British Thoracic Society). British guideline on the management of asthma, revised ed. April 2004. Available at: http://www.enterpriseportal2.co.uk/filestore/bts/asthmafull. pdf [last downloaded September 2006]

2. Rodrigo GJ, Rodrigoi C, Hall JB. Acute asthma in adults: a review. Chest 2004; 125; 1081-102

3. Hoskins G, McCowan C, Neville RG, et al. Risk factors and costs associated with an asthma attack. Thorax 2000;55:19-24

4. Canonica GW, Vignola AM. Adjustable maintenance dosing: suitability of budesonide/formoterol in a single inhaler and overview of a clinical study programme. Int J Clin Pract 2004;58 (Suppl. 141):9-17

5. Jadad A, Moore RA, Carroll D, et al. Assessing the quality of reports of randomised controlled trials: is blinding necessary? Control Clin Trials 1996;17:1-12

6. Dahl R, Chuchalin A, Gor D, et al. EXCEL: a randomised trial comparing salmeterol/fluticasone propionate and formoterol/ budesonide combinations in adults with persistent asthma. Respir Med 2006; 100:1 152-62

7. International conference on harmonisation of technical requirements for registration of pharmaceuticals for human use (ICH). E9: statistical principles for clinical trials, February 1998. Available at: http://www.ich.org/cache/compo/475-272-l.html [last downloaded September 2006].

8. Review Manager (RevMan) [Computer program]. Version 4.1 for Windows. Oxford, England: The Cochrane Collaboration; 2000 [CD-ROM and Internet]

9. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719-48

10. Petitti DB. Meta-analysis, decision analysis, and cost- effectiveness analysis: methods for quantitative synthesis in medicine, 2nd ed. Oxford University Press: Oxford; 2000

11. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177-88

12. Pauwels RA, Lofdahl C-G, Postma DS, et al. Effect of inhaled formoterol and budesonide on exacerbations of asthma. New Eng J Med 1997;337:1405-11

13. Price D, Dutchman D, Mawson A, et al. Early asthma control and maintenance with eformoterol following inhaled corticosteroid dose reduction. Thorax 2002;57:791-8

14. Aalbers R, Backer V, Kava TTK, et al. Adjustable maintenance dosing with budesonide/formoterol compared with fixed-dose salmeterol/fluticasone in moderate to severe asthma. Curr Med Res Opin 2004;20:225-40

15. Kuna P, Peters MJ, Manjira AI, et al. Effect of budesonide/ formoterol maintenance and reliever therapy on asthma exacerbations. Int J Clin Pract 2007;61:725-36

16. FitzGerald JM, Sears MR, Boulet L-P, et al. Adjustable maintenance dosing with budesonide/formoterol reduces asthma exacerbations compared with traditional fixed dosing: a five-month multicentre Canadian study. Can Respir J 2003,10: 427-34

17. Leuppi JD, Salzberg M, Meyer L, et al. An individualized, adjustable maintenance regimen of budesonide/formoterol provides effective astrima symptom control at a lower overall dose than fixed dosing. Swiss Med Wkly 2003;133:302-9

18. Michilis A, Peche RVJ, Verbraecken JA, et al. SURF study: real-life effectiveness of budesonide/formoterol (B/F) adjustable maintenance dosing. Allergy Clin Immunol Int 2003;15(Suppl 1):56 [Abs P-2-39].

19. Stallberg B, Olsson P, Jorgensen LA, et al. Budesonide/ formoterol adjustable maintenance dosing reduces asthma exacerbations versus fixed dosing. Int J Clin Pract 2003;57: 656-61 20. Buhl R, Kardos P, Richter K, et al. The effect of adjustable dosing with budesonide/formoterol on health-related quality of life and asthma control compared with fixed dosing. Curr Med Res Opin 2004;20:1209-20

21. Canonica GW, Castellani P, Cazzola M, et al. Adjustable maintenance dosing with budesonide/formoterol in a single inhaler provides effective asthma symptom control at a lower dose than fixed maintenance dosing. PuIm Pharmacol Ther 2004;17:239-47

22. lnd PW, Haughney J, Price D, et al. Adjustable and fixed dosing with budesonide/formoterol via a single inhaler in asthma patients: the ASSURE study. Respir Med 2004;98:464-75

23. Smozik Y, Iankova Z. Effectiveness of adjustable maintenance dosing and fixed dosing with budesonide/formoterol single inhaler in a multi-ethnic asthma population. Abstract and poster presented at the 1 4th annual congress of the European Respiratory Society, 4-8 September 2004, Glasgow, UK

24. AstraZeneca Data on File SYM/045/MAR2007

25. AstraZeneca Data on File SYM/044/MAR2007

26. Gotzsche PC. Multiple publication of reports of drug trials. Eur J Clin Pharmacol 1989;36:42932

27. Gotzsche PC. Reference bias in reports of drug trials. Br Med J 1987;295:6546

28. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet 1991;337:86772

29. Dickersin K, Min YI, Meinert CL. Factors influencing publication of research results: follow-up of applications submitted to two institutional review boards. J Am Med Assoc 1992;263:3748

30. Stern JM, Simes RJ. Publication bias: evidence of delayed publication in a cohort study of clinical research projects. Br Med J 1997;3 15:6405

31 . Angus R, Reagon R, Cheesbrough A. Short-acting beta 2- agonist and oral corticosteroid use in asthma patients prescribed either concurrent beclomethasone and long-acting beta 2-agonist or salmeterol/fluticasone propionate combination. Int J Clin Pract 2005;59:156-62

32. Mann M, Chowdhury B, Sullivan E, et al. Serious asthma exacerbations in asthmatics treated with high-dose formoterol. Chest 2003;124:70-4

33. Nelson HS, Weiss ST, Bleecker ER, et al. The salmeterol multicenter asthma research trial: a comparison of usual pharmacotherapy for asthma or usual pharmacotherapy plus salmeterol. Chest 2006; 129: 1 5-26

34. Sin DD, Man J, Sharpe H, et al. Pharmacological management to reduce exacerbations in adults with asthma: a systematic review and meta-analysis. J Am Med Assoc 2004;292:367-76

35. Gibson PG, Powell H, Ducharme FM. Differential effects of maintenance long-acting beta-agonist and inhaled corticosteroid on asthma control and asthma exacerbations. J Allergy Clin Immunol 2007;119:344-50

36. Ernst P, Mclvor A, Ducharme FM, et al. Safety and effectiveness of long-acting inhaled beta-agonist bronchodilators when taken with inhaled corticosteroids. Ann Intern Med 2006;145:692- 4

37. Day SJ, Altman DG. Blinding in clinical trials and other studies. BrMedJ2000;321:504

38. Zetterstrom O, Buhl R, Mellem H, et al. Improved asthma control with budesonide/formoterol in a single inhaler, compared with budesonide alone. Eur Respir J 2001;18:262-8

39. Bateman ED, Bantje TA, Joao Gomes M, et al. Combination therapy with single inhaler budesonide/formoterol compared with high dose of fluticasone propionate alone in patients with moderate persistent asthma. Am J Respir Med 2003;2:275-81

40. Rosenhall L, Heinig JH, Lindqvist A, et al. Budesonide/- formoterol (Symbicort) is well tolerated and effective in patients with moderate persistent asthma. Int J Clin Pract 2002;56: 427-33

41. Scicchitano R, Aalbers R, Ukena D, et al. Efficacy and safety of budesonide/formoterol single inhaler therapy versus a higher dose of budesonide in moderate to severe asthma. Curr Med Res Opin 2004;20:1403-18

42. O’Byrne PM, Bisgaard H, Godard PP, et al. Budesonide/ formoterol combination therapy as both maintenance and reliever medication in asthma. Am J Respir Crit Care Med 2005;171:129-36

43. Vogelmeier C, D’Urzo A, Pauwels R, et al. Budesonide/ formoterol maintenance and reliever therapy: an effective asthma treatment option? Eur Respir J 2005;26:819-28

44. Rabe KF, Atienza T, Magyar P, et al. Effect of budesonide in combination with formoterol for reliever in asthma exacerbations: a randomised controlled, double-blind study. Lancet 2006;368:744-53

45. Rabe KF, Pizzichini E, Stallberg B, et al. Budesonide/ formoterol in a single inhaler for maintenance and relief in mild- to-moderate asthma. Chest 2006;129:246-56

CrossRef links are available in the online published version of this paper: http://www.cmrojournal.com

Paper CMRO-3979_4, Accepted for publication: 15 May 2007

Published Online: 26 June 2007

doi:10.1185/030079907X210697

Steven J. Edwards(a), Kevin Gruffydd-Jones (b), and Dermot P. Ryan(c)

a Outcomes Research, AstraZeneca UK Ltd, Luton Bedfordshire, UK

b Box Surgery, Corsham, UK

c Woodbrook Medical Centre, Loughborough, UK

Address for correspondence: Steven J. Edwards MSc, Outcomes Research, AstraZeneca UK Ltd, Horizon Place, 600 Capability Green, Luton, Bedfordshire LUI 3LU, UK. Tel: +44 1582 836371; Fax: +44 1582 837372; email: steven.j.edwards@astrazeneca.com

Copyright Librapharm Aug 2007

(c) 2007 Current Medical Research and Opinion. Provided by ProQuest Information and Learning. All rights Reserved.