Background: Safety of long-acting beta agonists (LABA) has been questioned and recent evidence suggested a detrimental effect on asthma control as well as an increased risk of death.
Objective: To evaluate the safety of regular use of LABA compared with placebo or LABA added to inhaled corticosteroids (ICS) compared with ICS in persistent asthma.
Methods: Randomized studies from MEDLINE, EMBASE, and Cochrane Controlled Trials Register were identified. Additionally, AstraZeneca, GlaxoSmithKline, Novartis and FDA clinical trials databases were searched. Primary outcomes were asthma exacerbations (AE) requiring systemic corticosteroids or hospitalization, life-threatening exacerbations and asthma-related deaths.
Results: We identified 92 randomized clinical trials with 74,092 subjects. LABA (as monotherapy) reduced exacerbations requiring corticosteroids (Relative Risk [RR]=0.80; 95% CI, 0.73-0.88), without detrimental effects on hospitalizations or life-threatening episodes. Contrarily, LABA showed a significant increase in asthma-related deaths (Relative Risk=3.83; 95% CI, 1.21-12.14). Subgroup analysis suggests that children, patients receiving salmeterol, and a duration of treatment>12 weeks are associated with a higher risk of serious adverse effects; also there was a protective effect of concomitant use of ICS. On the other hand, combination of LABA/ICS reduced exacerbations (RR=0.73; 95% CI, 0.67-0.79), and hospitalizations (RR=0.58, 95% CI, 0.45-0.74). Combined therapy was also equivalent to ICS in terms of life-threatening episodes and asthma-related deaths. Again, children and use of salmeterol were associated with an increased risk of some severe outcomes as compared with adults and formoterol users, respectively.
Conclusions: This review reinforced the international recommendations in terms of the use of LABA remains the preferred add-on therapy to ICS for patients whose disease cannot adequately controlled with ICS, and that LABA cannot be prescribed as a monotherapy. Nevertheless, in spite of the protective effect of the ICS, children and salmeterol use still show an increased risk of non-fatal serious adverse events.