Theophylline is one of those medications that has intrigued and possibly confused clinicians and pharmacologists alike since the mid 19th century. Related agents caffeine and theobromine were used in the 1930s as bronchodilators, and theophylline developed an established place in the management of acute airflow obstruction associated with asthma and chronic obstructive pulmonary disease (COPD) during the mid 1900s.¹

Pharmacologically, theophylline is characterised principally as a phosphodiesterase (PDE) inhibitor. Its main biological action is to block the inactivation of cyclic AMP and cyclic GMP giving rise to bronchodilation, increased ciliary beat frequency, and reduced inflammatory cell numbers in the airways. However, it has been observed that other PDE inhibitors do not exert such effects, which suggests that other activities may be relevant. Theophylline also acts as an adenosine receptor antagonist, which may be relevant to its desirable effects and probably gives rise to a number of the side effects seen with use of this agent including arrhythmias, mental agitation, and diuresis. As well as bronchodilation, theophylline has a number of other potentially useful actions including improved gas exchange, respiratory stimulation, increased diaphragmatic performance, and improved exercise tolerance.¹

The pharmacokinetics of theophylline are important because of its narrow therapeutic index. In the acute setting most authorities recommend a loading dose in naïve patients followed by an infusion to maintain a serum concentration within the therapeutic range of 55–110 μmol/l. Because of the many factors which can affect theophylline metabolism including a number of potential drug interactions, therapeutic drug monitoring is recommended to avoid serious or even fatal toxicity. In elderly patients cardiac monitoring is advised to check for the emergence of serious tachyarrhythmias. Seizures are another potentially fatal complication.¹

Current evidence in support of the use of theophylline in acute severe asthma or COPD is weak at best. With regard to COPD, a recent meta-analysis published by Barr and colleagues identified four small clinical trials of oral or intravenous theophylline (aminophylline) suitable for inclusion, comprising a total of 169 subjects.² Summary data showed no benefit in terms of lung function, symptom scores or length of hospital stay, but a significant problem with side effects such as nausea and vomiting together with a non-significant increase in palpitations compared with placebo. They concluded that there was no clinical benefit, and this research informed the statements in both the UK and Australasian management guidelines on the subject that “it should only be used when there is an inadequate response to nebulised bronchodilators” (grade D, NICE guideline)³ and “the routine use of aminophylline is not recommended for acute exacerbations” (grade D, Thoracic Society of Australia and New Zealand guideline).⁴

A notable feature of published papers, including the meta-analysis by Barr et al,² is the small sample size of previous studies. In this issue of Thorax an adequately powered study is published which more or less clarifies the picture regarding the use of aminophylline in acute exacerbations of COPD. Duffy and co-workers⁵ studied the effects of adding intravenous aminophylline to usual care in subjects admitted to hospital with an acute exacerbation of COPD. Eighty well matched, non-acidotic subjects were randomised in a double blind fashion to receive intravenous aminophylline 0.5 mg mg/kg/hour or placebo and were then followed for 5 days. All subjects received nebulised bronchodilator and oral corticosteroids. The study showed no difference in spirometric values between the two groups, and there was no difference in symptom scores or length of hospital stay. The mean theophylline level achieved in the intervention group was 73.4 μmol/l (range 62.9–83.9). A significant proportion of the participants receiving active treatment experienced adverse effects including nausea in 46%. In six subjects the drug was stopped due to possible side effects including one subject who experienced a seizure during the second hour of treatment (the drug level was in the therapeutic range at the time of the seizure (53 µmol/l) and the seizure was later attributed on clinical grounds to alcohol withdrawal).

As might be expected, the investigators did not include subjects with very severe exacerbations. Those with a pH of 7.32 or less were excluded from entry to the study. The subjects on active treatment sustained a small but significant change in gas exchange parameters with a fall in arterial carbon dioxide tension and a small rise in pH over the initial 2 hours of treatment. This finding would be in keeping with the drug’s known positive influence on respiratory drive.

Thus, in considering the generalisablility of the study, aminophylline might still be considered in the management of life threatening episodes of COPD by an experienced doctor in selected cases, together with other measures such as non-invasive ventilation. In such circumstances the benefits of respiratory stimulation and any effect on respiratory muscles may be more important than bronchodilation per se. However, for most clinical situations involving mild to moderate COPD exacerbations, we now have a clear answer to the question whether aminophylline should be used—and it is “no”.