Purpose:

To compare the efficacy and safety of inhaled corticosteroids (ICSs), long-acting beta-2 agonists (LABAs), leukotriene modifiers (LMs), anti-IgE therapy, combination products, and tiotropium for people with persistent asthma.

Data Sources:

To identify published studies, we searched MEDLINE, The Cochrane Library, Embase, International Pharmaceutical Abstracts, and reference lists of included studies through September 2010. We also requested dossiers of information from pharmaceutical manufacturers.

Review Methods:

Study selection, data abstraction, validity assessment, grading the strength of the evidence, and data synthesis were all carried out according to standard Drug Effectiveness Review Project methods.

Results:

Efficacy studies provide moderate strength of evidence (SOE) that equipotent doses of ICSs administered through comparable delivery devices do not differ in their ability to control asthma symptoms, prevent exacerbations, reduce the need for additional rescue medication, or in their overall incidence of adverse events or withdrawals due to adverse events. Evidence does not support a difference between montelukast and zafirlukast in their ability to decrease rescue medicine use or improve quality of life (low SOE for ≥ 12 years of age, insufficient <12), between formoterol and salmeterol in their ability to control symptoms, prevent exacerbations, improve quality of life, or cause harms among patients not controlled on ICSs alone (moderate SOE), or between budesonide/formoterol and fluticasone/salmeterol for efficacy or harms when each combination is administered via a single inhaler (moderate SOE for ≥ 12, insufficient <12). Meta-analyses and efficacy studies provide consistent evidence favoring omalizumab over placebo for most included outcomes. Omalizumab-treated patients have an increased incidence of injection site reactions and anaphylaxis compared to placebo-treated patients.

We found consistent evidence of greater benefit for subjects treated with ICS monotherapy compared with those treated with LM monotherapy (high SOE). Direct evidence suggests no difference in tolerability or overall adverse events between ICSs and LMs (moderate SOE). Specific adverse events reported with ICSs, such as cataracts and decreased growth velocity, were not found among patients taking LMs. The best longer-term evidence on growth (avg 4.3 years) for ICSs compared with placebo found that very small differences (1.1 cm) occurred primarily during the first year of treatment, suggesting that the effect on growth velocity occurs early in treatment and is not progressive. Evidence is insufficient to determine if long-term treatment with ICSs leads to a reduction in final adult height. Overall evidence indicates that ICSs and leukotriene receptor antagonists (LTRAs) are safer than LABAs for use as monotherapy (high SOE). Indirect evidence suggests that the potential increased risk of asthma-related death for those taking LABAs may be confined to patients not taking ICSs at baseline.

We did not find sufficient evidence to support the routine use of combination therapy rather than an ICS alone as first line therapy (moderate SOE for ≥ 12, insufficient <12). Results from large trials support greater efficacy with the addition of a LABA to an ICS than with a higher dose ICS (high SOE for ≥ 12, low <12) and greater efficacy with the addition of a LABA to an ICS over continuing the current dose of ICS alone for poorly controlled persistent asthma (high SOE). The addition of LMs to ICSs compared to continuing the same dose of ICSs resulted in improvement in rescue medicine use and no statistically significant differences in other health outcomes (low SOE for ≥ 12, insufficient <12). There is no apparent difference in symptoms, exacerbations, rescue medicine use, overall adverse events, or withdrawals due to adverse events between those treated with ICSs plus LTRAs compared to those treated with increasing the dose of ICSs (moderate SOE for ≥ 12, low <12). Results provide strong evidence that the addition of a LABA to ICS therapy (ICS+LABA) is more efficacious than the addition of an LTRA to ICS therapy (ICS+LTRA) (high SOE for ≥ 12, low <12). We found no difference in overall adverse events or withdrawals due to adverse events between ICS+LABA and ICS+LTRA (moderate SOE for ≥ 12, insufficient <12).

Conclusion:

Overall findings do not suggest that one medication within any of the classes evaluated is significantly more effective or harmful than the other medications within the same class, with the exception of zileuton being more harmful than the other LMs. Our results support the general clinical practice of starting initial treatment for persistent asthma with an ICS. For people with poorly controlled persistent asthma taking an ICS, our findings suggest that the addition of a LABA is most likely to provide the greatest benefit as the next step in treatment.