Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Methylxanthines for COPD

Systematic review
Adverse events

Use of methylxanthines (oral theophylline, intravenous aminophylline or doxofylline) is recommended for consideration in addition to bronchodilators for severe exacerbations of COPD. An answer to the simple question of whether they work comes from a systematic review and meta-analysis [1].

Systematic review

Reviewers searched four electronic databases, including the Cochrane Library, as well as hand-searching 20 respiratory journals. Included were randomised studies comparing methylxanthine with placebo for exacerbations of COPD. Patients had to have known COPD with an exacerbation requiring admission to hospital or emergency care. Treatment had to occur in the emergency department or immediately on admission to hospital. Outcomes of interest were lung function tests, admission to hospital or readmission, and adverse events.


There were four studies included in the review, with information on 169 patients. Three studies were published in full and one was available as an abstract. For two of the studies, authors provided additional information. The studies were of high reporting quality, minimising the possibility of bias. Three trials tested intravenous aminophylline and one oral theophylline, all added to standard treatments.


There were no substantial or convincing benefits for any clinical outcomes, including lung function tests, symptom scores, or emergency department return within one week.

Adverse events

Results for the adverse events of nausea and vomiting, tremor, and palpitations or arrhythmias are shown in Table 1. For each event the incidence was higher with methylxanthines than for placebo, but was statistically significant only for nausea and vomiting, where 37% of patients had this adverse event with methylxanthines, and 13% with placebo (Figure 1). The number needed to harm was 4.6 (95%CI 2.6 to 12), meaning that one additional patient would suffer nausea and vomiting for every five treated.

Table 1: Results for adverse events of methylxanthines compared with placebo as adjunct treatment for COPD exacerbation

adverse event/total (%)
Adverse event
Number of trials
Relative risk
2.6 (1.4 to 5.1)
4.6 (2.6 to 12)
1.3 (0.9 to 2.0)
not calculated
3.6 (0.9 to 15)
not calculated

Figure 1: Nausea and vomiting


The simple conclusion is that, on the limited evidence available, methylxanthines produce no convincing benefit for treating exacerbations of COPD, but they do produce additional harm. Clearly there are limitations here, the obvious one being that with so few patients in trials there might be a small benefit that larger studies might uncover.

Yes, but. That but is the big one. The evidence we have suggests that any limited benefit there may be would be at the risk of significant harm. The ethics of additional research in this area would make for an interesting discussion.

There are two interesting points from this meta-analysis of small studies. The first is that several important guidelines for treating COPD include methylxanthines, while at least one in the USA does not. The review should thus help guideline makers amend their advice.

The second is apparently trivial, but actually quite important. There are several simple numerical errors in this paper. They do not affect the result, although there is an interesting forensic exercise in determining that to be the case. It would make it a useful teaching paper in critical appraisal to demonstrate that we need to be hawk-eyed when reading papers even in the most prestigious journals.


  1. RG Barr et al. Methylxanthines for exacerbations of chronic obstructive pulmonary disease: meta-analysis of randomised trials. BMJ 2003 327: 646-651.

previous or next story