UK Users

Access Prodigy topics and clinical content at the NICE website.

Visit the NICE website

Free Trial

To arrange a free trial of the Prodigy application, get in touch.

Contact us to arrange a trial

International Users

The Prodigy application is available on a subscription basis.

Subscription types

Sample Topic: Asthma

Asthma : Last revised in December 2013


Evidence on long-acting beta2-agonists

  • LABA versus placebo:
    • Four randomized controlled trials (RCTs) (n = 2063) in people with persistent asthma not controlled with beclometasone, 200 to 2000 micrograms/day, showed that the addition of salmeterol or formoterol improved peak expiratory flow rate, forced expiratory volume in 1 second (FEV1) and reduced night awakening compared with placebo. The studies involving salmeterol also showed improvements in quality-of-life scores compared with placebo. In the formoterol trials, symptoms were improved significantly at 6 months. However, exacerbation rates did not differ between groups in any RCT [Rodolfo et al, 2005].
    • A systematic review (search date June 2004, 26 RCTs [eight on children, 18 on adults]) randomized people to an ICS alone (beclometasone, 200 to 400 g/day or equivalent) or an ICS with LABAs (formoterol or salmeterol). Most trials lasted 4 months or less. The addition of LABA reduced risk of exacerbations requiring systemic steroids (NNT 18, 95% CI 13 to 33), and increased the proportion of symptom-free days by 17% (95% CI 12 to 22) in 6 trials, increased the proportion of rescue-free days by 19% (95% CI 12 to 26) in two trials, and reduced use of rescue short-acting beta2-agonists by 0.7 puff/day (95% CI –1.2 to –0.2 puff/day) [Ni Chroinin et al, 2005]. A later Cochrane review supports the benefits of LABA therapy in terms of lung function, fewer symptoms, less use of rescue medication, and improved quality-of-life score compared with placebo. However, the risk of exacerbations in children may have been increased; this finding needs further confirmation [Walters et al, 2007].
    • In a systematic review (search date July 2005) of ten parallel-group randomized trials, two studies showed that in adults, the addition of a LABA to ICS allows reduction of the ICS dose. The groups did not significantly differ in the rates of severe exacerbations requiring oral corticosteroids and withdrawal due to worsening asthma, and LABA therapy was associated with significant improvements in FEV1, peak expiratory flow rate, and percentage of rescue-free days [Gibson et al, 2005].
  • LABA versus increasing ICS:
    • A systematic review (search date July 2005, 30 RCTs, n = 9509, children > 2 years of age [three trials] and adults [27 trials]) compared a combination of LABA plus ICS with a higher dose of ICS. The combination regimen significantly increased symptom-free days by 12% (eight trials), reduced daytime use of rescue beta2-agonists by 1 puff/day (four trials), and reduced the rate of withdrawals owing to poor asthma control (20 trials). The groups did not significantly differ in the rate of exacerbations requiring systemic steroids (15 trials) or overall adverse events (15 trials). Tremor was the only significant adverse effect that was increased with LABA therapy (10 trials) [Greenstone et al, 2005].
    • One systematic review (search date 1999, nine RCTs) and eight additional RCTs found that adding LABA (salmeterol or formoterol) to ICS improved lung function and symptoms compared with increasing the ICS dose. Addition of LABA reduced exacerbation rates. However, one RCT found that increasing ICS compared with adding formoterol significantly reduced severe exacerbations at 1 year [Rodolfo et al, 2005].
    • An RCT (n = 1272) showed that formoterol was more effective than doubling the dose of budesonide at 1 year for reducing the risk of severe exacerbations and poorly controlled asthma days [O’Byrne et al, 2001].
  • LABA versus leukotriene receptor antagonists:
    • Seven RCTs (total n = 3943) compared salmeterol or formoterol with leukotriene receptor antagonists (montelukast or zafirlukast). All studies had slightly different results, but overall, LABA improved lung function and symptoms more than leukotriene receptor antagonists. One of the larger RCTs (n = 1490) showed no significant difference in asthma exacerbations over 48 weeks between fluticasone plus salmeterol and fluticasone plus montelukast (19.1% vs. 20.1%; difference 1%, 95% CI –3.1 to +5) [Rodolfo et al, 2005]. A recent Cochrane review (15 RCTs, 11 RCTs included in the meta-analysis, n = 6030) supports the findings of early RCTs and adds that LABA lower the risk of exacerbations requiring systemic corticosteroids (NNT 38 to prevent one exacerbation over 48 weeks, 95% CI 23 to 247) compared with leukotriene receptor antagonists [Ducharme et al, 2006].
  • LABA versus theophylline:
    • A systematic review (search date April 2003, 12 RCTs) showed LABA to be at least as effective as theophylline in reducing symptoms (including night awakening) and improving lung function. However, LABA were associated with less adverse effects than theophylline. A later systematic review (search date November 2006, 13 RCTs, n = 1344) confirmed these findings and suggested that LABA were more effective than theophylline in improving morning and evening peak expiratory flow rate. Salmeterol decreased the need for short-acting beta2-agonist compared with theophylline [Tee et al, 2007].
  • LABA adverse effects:
    • One RCT (Salmeterol Multicenter Asthma Research Trial [SMART], n = 26,355 people > 12 years of age) showed that salmeterol significantly increased the number of respiratory-related deaths (RR 2.16, 95% CI 1.06 to 4.41), asthma-related deaths (RR 4.37, 95% CI 1.25 to 15.3; NNH 1315), and combined asthma-related deaths or life-threatening experiences (RR 1.71, 95% CI 1.01 to 2.89; NNH 909). Mortality in the RCT was very low: the number needed to harm (one asthma-related death) was 1318 (95% CI 746 to 6173) after a mean of 7 months. Whether this increased risk represents lack of appropriate use of ICS, a treatment effect, or genetic factors is not yet clear [Nelson et al, 2006]. A Cochrane review (search date October 2005, 67 RCTs, n = 42,333) highlights that these observations were drawn from a post hoc analysis and lack the validity of pre-defined distinctions. The review also states that LABA used in appropriate candidates, along with ICS, are very safe [Walters et al, 2007].
    • A systematic review (19 RCTs, n = 33,836) including the SMART trial showed that use of LABA (salmeterol and formoterol) for at least 3 months (range 3–12 months) compared with placebo resulted in hospitalization for asthma exacerbation in 1.72% versus 0.6% (NNH 89) based on meta-analysis of 12 trials (n = 5091), and life-threatening asthma exacerbations in 0.32% versus 0.17% (NNH 666) based on meta-analysis of seven trials (n = 29,981). There were 15 versus 3 asthma-related deaths in 14 trials with reported asthma-related deaths (NNH 1428, 95% CI 1000 to 10,000). In sensitivity analysis that assumed no deaths in 28 additional trials, the absolute increase in risk was 0.06% (NNH 1666) [Salpeter et al, 2006].