It is disturbing to realise that, although inhaled corticosteroids have been prescribed for over three decades in the treatment of asthma, there are still major therapeutic issues associated with their use that have yet to be clearly determined. One of these issues is the optimal starting dose. This uncertainty is illustrated by the different approaches that have been proposed for starting inhaled corticosteroids in patients with asthma ranging from “start with a high dose then step down once control has been achieved”,^1,2 “start with a dose of 400 µg³ or 200–500 µg per day beclomethasone dipropionate (BDP) or equivalent and then step up if required”,⁴ or “start at the dose considered to be appropriate for the severity of disease, usually 400 µg of BDP or equivalent”.⁵

To address this uncertainty and determine the optimal starting dose of inhaled corticosteroids in adults with asthma, Powell and Gibson have undertaken a systematic review and meta-analysis of randomised controlled trials that have investigated this issue. Their findings were reported in a recent issue of Thorax⁶ and indicate that, for patients with asthma who require inhaled corticosteroids, there is no evidence of a difference in efficacy between commencing with a moderate dose (400 µg per day BDP or equivalent) compared with a high dose (⩾800 µg per day) and stepping down. An initial moderate dose of inhaled corticosteroids was more effective than an initial low dose (200 µg per day).

These findings are reassuring in that they do not support the “start high, step down” approach which has the potential to lead to inappropriately high doses of inhaled corticosteroids being used long term. This may occur if the patient does not attend for regular review so the initial dose becomes the long term maintenance dose, or if on medical review there is a reluctance to reduce the high dose in case it leads to worsening asthma control. The preferred approach demonstrated in this meta-analysis of initiating treatment with moderate doses of inhaled corticosteroids has advantages in terms of a better risk/benefit ratio and reduced economic cost. As discussed by the authors, the main limitations of their meta-analysis were the small number of studies which could be included in the subgroup analyses and the inability to examine severe exacerbations as an outcome measure as it was not reported in most of the included studies. However, despite these limitations, the findings from this meta-analysis represent the best integration of current scientific evidence and can be incorporated in consensus guidelines. In this regard, the current British guidelines represent the ideal recommendation—to start at a dose considered to be appropriate for the severity of disease, usually 400 µg BDP or equivalent.⁵

The findings are also consistent with recent meta-analyses of the dose-response relationship of inhaled corticosteroids in adult asthma.^7,8,9,10,11 The initial meta-analysis showed that, for different outcome measures including lung function, symptoms, β agonist use and exacerbations, at least 90% of the maximum efficacy can be achieved with a dose of fluticasone propionate (FP) of around 200 μg per day.⁷ In the moderate to severe adult asthmatic patients included in the studies of the meta-analysis, the maximum therapeutic effect was achieved with a dose of FP of around 500 μg per day. This meta-analysis challenged the dogma that existed at the time—that higher doses were required to achieve the maximal obtainable effect and that there were marked differences in the dose-response relationship for different clinical outcome measures. In particular, the dose of FP required to reduce exacerbations was similar to that required to reduce symptoms and improve lung function.

Similarly, in another meta-analysis of placebo controlled dose-response studies of FP, no significant differences were noted in the magnitude of change in morning peak flow in patients receiving high (500 or 1000 µg per day) or low (⩽200 µg per day) doses of FP.¹¹ The time taken to reach either 50% or 100% of the best observed effect was no longer in the low dose group, once again demonstrating no reduction in different parameters of efficacy.

The major limitation of these meta-analyses was the small number of studies that included FP doses of 1000 µg per day or more, due to the requirement for the studies to be placebo controlled. This led to a subsequent meta-analysis which specifically focused on comparisons between a dose of 200 µg per day and higher doses to determine whether the 200 µg per day dose regime provided most of the therapeutic benefit as suggested in the original study.⁸ It was confirmed that most of the therapeutic effect was achieved with a dose of 200 µg per day, and that an increase in dose to ⩾500 µg per day resulted in minimal additional benefit.

Another meta-analysis with inhaled budesonide has shown that most of the clinical efficacy for the same outcome measures is achieved with a dose of around 400 µg per day.⁹ Likewise, a large dose-response study of BDP showed a plateau in response between 400 and 800 μg per day, depending on the clinical outcome variable.¹² These findings are comparable with those of FP when their relative potencies are considered (FP v budesonide or BDP around 2:1).

Another consideration is whether a different dose-response relationship exists for inhaled corticosteroids in terms of modifying the underlying airways inflammation. Only a few studies have addressed this issue because of the requirement for repeated bronchoscopic examinations to obtain bronchial biopsy and lavage samples. These studies have reported no additional effect on either the nature or magnitude of the airways inflammation with increasing the dose of FP from 400–500 µg per day to 1000–2000 µg per day.^13,14 As a result, the dose-response relationship of inhaled corticosteroids for anti-inflammatory effects in the airways appears to be similar to that for the clinical outcome measures.

The major clinical outcome measure which could not be assessed in dose-response studies is mortality because of its rare occurrence even in patients with severe asthma. However, it is possible to obtain an indication of the dose-response effect of inhaled corticosteroids for reducing the risk of mortality from the epidemiological study reported by Suissa et al.¹⁵ In this study there was a progressive reduction in risk of mortality with increasing use of inhaled corticosteroids. Consistent with the major clinical outcome measures, at least 80% of the maximum obtainable benefit (reduction in mortality) was achieved at around 200 µg BDP per day (adjusted odds ratio 0.15). As a result, available evidence suggests that low doses of inhaled corticosteroids are effective in reducing the risk of death from asthma, with a dose response similar to that of other major outcome variables such as symptoms, lung function, and severe exacerbations.

Based on these studies, it can be recommended that the standard dose of inhaled corticosteroid for routine prescribing in adult asthma is around 400 μg per day BDP or budesonide, or 200 μg per day FP. This recommendation should be qualified by the recognition that there is considerable variability between individuals in their response to inhaled corticosteroids in asthma, which means that some patients may obtain a greater clinical benefit at higher doses, just as some patients may obtain the maximum efficacy at lower doses.¹⁶

So what is the priority now in terms of research into the use of inhaled corticosteroid therapy in asthma? Currently, the major dilemma for clinicians is to know at what dose of inhaled corticosteroids to start concomitant long acting β agonist (LABA) therapy in a patient inadequately controlled on inhaled corticosteroids alone. This uncertainty is reflected by the different guideline recommendations to add a LABA over a fivefold inhaled corticosteroid dose range (200–1000 µg per day BDP or equivalent).^2–5 Determination of the dose of inhaled corticosteroids at which to add a LABA as first line add-on therapy at step 3 of the guidelines will probably also require a meta-analytical approach. Elucidation of the optimal starting dose and the therapeutic dose range of inhaled corticosteroids has provided important information for the planning of such a meta-analysis.