O2. Oral bronchodilators

O2.1 Methylxanthines

Theophylline is rarely used for COPD in Australia.  A small randomised placebo controlled trial in China demonstrated that doses of 100mg twice daily reduced exacerbations compared with placebo (Zhou 2006).  In this study, patients were not on inhaled corticosteroids or long-acting bronchodilators which limits the generalisability of the study findings.  Devereux et al randomised 1,567 UK-based COPD patients with a history of exacerbations to theophylline or placebo.  All patients were receiving inhaled corticosteroids and 80% of patients were on ‘triple-therapy’ (Devereux 2018). An RCT of low dose theophylline plus low dose oral prednisone, theophylline or placebo in 1,670 patients with COPD in China found no statistically significant differences in exacerbation rates, hospitalisations, FEV₁, SGRQ and CAT scores at 48 weeks (Jenkins 2020) [evidence level II].

A meta-analysis of 4 RCTs and 3 cohort studies (n=47,556) examined the addition of theophylline to inhaled corticosteroids.  Of the 7 studies reviewed, 4 used an ICS/LABA combination, 2 used ICS alone and 1 trial did not specify. Theophylline was associated with a higher hospitalization rate (HR 1.12, 95% CI 1.10-1.15), and mortality (HR 1.19, 95% CI 1.14-1.25) (Shuai 2021) [evidence level I].

Based on the available evidence, theophylline cannot be recommended for patients with COPD.

O2.2 Phosphodiesterase type-4 inhibitors

Phosphodiesterase type-4 (PDE-4) inhibitors act by increasing intracellular concentrations of cyclic adenosine monophosphate (cAMP) to suppress inflammation and bronchoconstriction. A Cochrane Review analysed results from RCTs of roflumilast (20 trials, 17,627 patients) and cilomilast (14 trials, 6,457 patients) (Chong 2017) [evidence level I]. Compared to placebo, PDE-4 inhibitors improved FEV₁ (mean difference 51 ml, 95% CI 43 to 60, moderate quality evidence) and reduced exacerbation rates (OR 0.78, 95% CI 0.73 to 0.83, high quality evidence; NNTB 20, 95% CI 16 to 26), but had relatively small effects on quality of life and symptoms. Gastrointestinal adverse effects were more frequent with the PDE-4 inhibitors, and psychiatric adverse events such as insomnia and depressive mood symptoms were more frequent with roflumilast (OR 2.13, 95% CI 1.79 to 2.54). These oral agents are not currently available in Australia.