The further paradoxes of asthma management: time for a new approach across the spectrum of asthma severity

Should short-acting β-agonist (SABA) reliever monotherapy be replaced by combination inhaled corticosteroid (ICS)/fast-onset β-agonist as reliever therapy in asthma patients at Global Initiative for Asthma step 1 [1]? This uncertainty is the latest chapter in the controversy relating to the risks and benefits of β-agonist therapy in asthma management [2]. This 70-year controversy began with the introduction of isoprenaline and the concerns that it caused severe refractory asthma and an increased risk of death [3–5].

The dilemma of evaluating the relative efficacy and risk of β-agonists has persisted since then. Two asthma mortality epidemics in the 1960s and 1970s/1980s were primarily caused by use of high-dose preparations of the potent, yet poorly β₂-selective agonists, isoprenaline [6] and fenoterol [7], respectively. The epidemics ended abruptly with regulatory restriction and withdrawal of these agents [2, 8]. In response to evidence that regular scheduled use of SABAs could increase asthma severity and the risk of exacerbations, the recommendation for their use was changed to “as required for relief of symptoms”. This therapeutic approach, together with the increasing use of ICSs, both as separate inhalers and combination ICS/long-acting β-agonist (LABA) inhalers, has been associated with a marked reduction in asthma mortality since the 1990s, although the global rates of asthma mortality have now plateaued [9]. There is clearly a need for alternative treatment strategies to achieve further reductions in global asthma mortality.

One strategy is to reduce the potential risks associated with starting SABA reliever monotherapy in intermittent and mild persistent asthma. This strategy was the focus of a recent review in the European Respiratory Journal [1], which identified five paradoxes of this therapeutic approach. The review summarised evidence that SABA reliever monotherapy exposes patients to significant avoidable risk and leads to learned behaviours, which result in difficulties for both patients and doctors (table 1). SABA reliever therapy not only reduces the potential to achieve optimal control at step 1, but also when patients progress to higher treatment steps in the stepwise approach to pharmacological therapy. In this editorial, we propose five related paradoxes that further develop the concepts introduced in this recent review [1] (table 2).

First paradox

The first paradox is that, although the recommendation is that all patients with asthma should be prescribed a SABA, regular use of this medication increases bronchial hyperresponsiveness (BHR). The magnitude of this effect is substantial, with a decreased doubling dose of −0.47 (95% CI −0.7 to −0.2, p<0.001) pooled across all three stimulant classes for BHR, derived from a systematic review and meta-analysis [10]. BHR is a physiological marker of asthma severity, and increasingly severe BHR will result in asthma patients becoming more sensitive to provoking stimuli, as well as increased treatment requirements, symptom burden and diurnal variation in lung function [11]. Although increased BHR has only been reported with regular SABA use, the effect of “as needed” SABA use on BHR may also be important, although this has not been studied [10]. With SABA reliever therapy, worsening asthma control results in an increased frequency of SABA use, which may be equivalent to regular prolonged use [12]. The effect of increased BHR occurs within 1–2 weeks of regular SABA treatment [13, 14], which is similar to the time course of an exacerbation associated with increased SABA reliever use [15].

ICSs improve BHR and the therapeutic approach of concomitant prescription of a maintenance ICS with SABA reliever therapy can arguably counteract the deleterious effects of frequent SABA use [16]. However, in clinical practice, this scenario does not apply to patients prescribed SABA as monotherapy, or to those who have prolonged nonadherence to maintenance ICS-based therapies.

The effects of SABA on BHR also have a major public health significance considered at a population level [17]. Modelling the effect of at least a half doubling dose worsening in BHR across a population, similar to the magnitude of the effect of SABA therapy [10], demonstrates that it will increase the prevalence of moderate and severe asthma (defined by the degree of BHR) from 48% to 63% [18].

Second paradox

The second paradox is that SABA monotherapy is prescribed when there is strong evidence that maintenance ICS with concomitant SABA reliever therapy results in superior efficacy [19]. The argument in defence of SABA monotherapy is that, although the efficacy of ICS is evident even in patients with only occasional symptoms and SABA use [20], patients are unlikely to take prescribed ICS daily when experiencing infrequent symptoms. This argument stands only if there is no alternative therapeutic approach to overcome low rates of ICS adherence. However, the alternative approach of using an ICS/fast-onset β-agonist as reliever monotherapy automatically ensures that ICS is titrated in response to symptom frequency and severity, through the vehicle of fast-onset β-agonist reliever use [21].

Third paradox

The third paradox is that prescribing SABA monotherapy as initial therapy may delay prescribing ICS, thereby reducing the potential efficacy of ICS therapy. It could be argued that patients initially prescribed SABA monotherapy should, if guidelines are followed, ultimately be prescribed an ICS if or when their asthma worsens. Unfortunately, patients with poorly controlled asthma that warrants ICS therapy often remain on SABA monotherapy [22]. There is evidence that such a delay in prescribing ICS reduces their potential efficacy [23, 24].

Fourth paradox

The fourth paradox is that ICS/SABA reliever monotherapy is more efficacious than SABA reliever monotherapy at step 1 and has similar efficacy to ICS maintenance and SABA reliever therapy at step 2 [25], and yet is not recommended in international guidelines at either step. More recently, ICS/fast-onset LABA reliever monotherapy has also been shown to be more efficacious than SABA reliever monotherapy at step 1 and has similar efficacy to ICS maintenance and SABA reliever therapy at step 2, in terms of severe exacerbation risk [26, 27]. Those patients who are over-reliant on SABA and nonadherent with ICS may gain particular benefit from combination ICS/fast-onset β-agonist reliever monotherapy, as ICS use will be driven by the use of such combination medications as reliever therapy. Furthermore, ICS/fast-onset LABA reliever monotherapy has similar efficacy in reducing severe exacerbation risk to maintenance ICS/fast-onset LABA with SABA reliever therapy at step 3 [28].

Fifth paradox

The fifth paradox is that, although ICS/fast-onset LABA as reliever therapy has greater efficacy than SABA reliever therapy across steps 3 and 4 [29], in clinical practice most patients receive SABA reliever therapy at these steps. Furthermore, even those prescribed ICS/fast-onset LABA maintenance and reliever therapy are commonly co-prescribed a SABA [30]. This entrenched practice persists, not only despite the greater efficacy of ICS/fast-onset LABA reliever therapy compared with SABA reliever therapy at steps 3 and 4, but also despite the evidence that ICS/fast-onset LABA reliever therapy reduces the excessive use of β-agonist therapy, the number of days of β-agonist overuse associated with delay in obtaining medical help and the number of days without self-administration of ICS [31]. These are all common and important factors associated with increased risk of asthma mortality [32].

Potential limitations of the proposed treatment approaches

The potential limitations of the proposed treatment approaches that arise from the consideration of these paradoxes also need consideration. These include, but are not restricted to, the possibility of overtreatment with both the ICS and LABA components of ICS/fast-onset LABA reliever therapy and the requirement for a stronger evidence base for ICS/SABA reliever therapy. However, such ICS “overtreatment” does not lead to an increased corticosteroid burden. In a randomised controlled trial, although ICS/fast-onset LABA reliever therapy resulted in greater ICS exposure than SABA reliever therapy, the subsequent reduction in severe exacerbations resulted in reduced oral corticosteroid exposure, leading to a similar overall systemic corticosteroid exposure [31]. In the same randomised controlled trial, the potential concern of LABA overuse with ICS/fast-onset LABA reliever therapy was addressed, as ICS/fast-onset LABA reliever therapy resulted in fewer β-agonist overuse episodes compared to SABA reliever therapy [31]. Additionally, there were fewer β-agonist overuse episodes that led to delays in seeking medical review, in high-risk patients receiving maintenance ICS/fast-onset LABA [31].

Prescription of ICS/fast-onset LABA reliever therapy with a different ICS/LABA or ICS/ultra-LABA for regular maintenance use is problematic, as the safety of concurrent use of two different LABAs has not been assessed. In these cases, ICS/SABA reliever therapy is likely to be the preferred approach, as there is an evidence base for the safety of concomitant use of SABA reliever and maintenance ICS/LABA therapy [33]. An alternative would be a change to ICS/fast-onset LABA as both maintenance and reliever therapy [29].

Clearly, further clinical research investigating the efficacy and safety of both ICS/SABA and ICS/fast-onset LABA reliever therapy regimens is required across the spectrum of asthma severity and its treatment. Comparisons between ICS/SABA and ICS/fast-onset LABA reliever therapy at the different steps will also be necessary to determine which patients may preferentially benefit from each regimen, to allow a personalised approach to asthma management.

The long-standing mantra of mandatory self-administration of SABAs for the acute relief of bronchoconstriction for all patients with asthma, and the tacit acceptance of the potential long-term adverse effects of this therapy, should be challenged. There are alternative options with superior efficacy and safety. These options include combination ICS/fast-onset β-agonist reliever therapy, based on supportive trial evidence for ICS/SABA reliever monotherapy at steps 1 and 2, ICS/fast-onset LABA reliever monotherapy at steps 1–3, and ICS/fast-onset LABA as part of the maintenance and reliever therapy regimen at steps 3 and 4. It is clearly time for a paradigm shift in the treatment of asthma that addresses the paradoxes of asthma management with respect to SABA reliever therapy. Strong consideration should be given to ICS/fast-onset β-agonist reliever therapy replacing the current advice for SABA reliever therapy across the range of asthma severity.