The effects of oral xanthines (e.g. theophylline) for chronic asthma in children

Xanthines (e.g. theophylline) are a group of drugs thought to have helpful preventative and reliever properties in the treatment of asthma in children. This review of studies has established that there is evidence for useful effects of these drugs in terms of symptom relief and lung function, but also some evidence of side-effects. As a primary preventative therapy, whilst there is evidence that xanthines are effective, this review suggests that more effective alternative treatment options (inhaled steroids) are available. In children with more severe asthma, the role of xanthines as an add-on therapy has only been assessed in a small number of trials, which report mixed effects. More studies in this area would help to generate a more reliable overview of the effects of treatment in these children. Xanthines are an effective preventative treatment in childhood asthma, but less effective than inhaled steroids, and with a less favourable side-effect profile. There is insufficient evidence at present to assess their role as "add-on" preventer treatment versus newer alternatives. Some of the trials exposed the children they recruited to a pre-trial phase of xanthine in order to maintain effective dosing during the trial. This could have made the trial participants less representative of the general population by making them more inclined to tolerate the study drug. This exposure also may have meant that they could recognise what drug they were taking during the blinded phase of the study.

Authors' conclusions: 

Xanthines as first-line preventer alleviate symptoms and reduce requirement for rescue medication in children with mild to moderate asthma. When compared with ICS they were less effective in preventing exacerbations. Xanthines had similar efficacy as single preventative agent compared with regular SABA and SCG. Evidence on AEs (adverse effects) was equivocal: there was evidence for increased AEs overall, but no evidence that any specific AE (including effects on behaviour and attention) occurred more frequently than with placebo. There is insufficient evidence from available studies to make firm conclusions about the effectiveness of xanthines as add-on preventative treatment to ICS, and there are no published paediatric studies comparing xanthines with alternatives in this role. Our data suggest that xanthines are only suitable as first-line preventative asthma therapy in children when ICS are not available. Pre-trial exposure to the agents assessed may have pre-disposed the trial populations to tolerate the drug, and may have threatened blinding. They may have a role as add-on therapy in more severe asthma not controlled by ICS, but further studies are needed to examine this, and to define the risk-benefit ratio compared with other agents.

Read the full abstract...

Xanthines have been used in the treatment of asthma as a bronchodilator, though they may also have anti-inflammatory effects. The current role of xanthines in the long-term treatment of childhood asthma needs to be reassessed.


To determine the efficacy of xanthines (e.g. theophylline) in the maintenance treatment of paediatric asthma.

Search strategy: 

A search of the Cochrane Airways Group Specialised Register was undertaken with predefined search terms. Searches are current to May 2008.

Selection criteria: 

Randomised controlled trials, lasting at least four weeks comparing a xanthine with placebo, regular short-acting beta-agonist (SABA), inhaled corticosteroids (ICS), cromoglycate (SCG), ketotifen (KET) or leukotriene antagonist, in children with diagnosed with chronic asthma between 18 months and 18 years old.

Data collection and analysis: 

Two reviewers independently selected each study for inclusion in the review and extracted data. Primary outcome was percentage of symptom-free days.

Main results: 

Thirty-six studies (2838 participants) were included. Xanthine versus placebo (18 studies): The proportion of symptom free days was larger with xanthine compared with placebo (7.97% [95% CI 3.41, 12.53]). Rescue medication usage was lower with xanthine, with no significant difference in symptom scores or hospitalisations. FEV1, and PEF were better with xanthine. Xanthine was associated with non-specific side-effects. Data from behavioural scores were inconclusive. Xanthine versus ICS (four studies) : Exacerbations were less frequent with ICS, but no significant difference on lung function was observed. Individual studies reported significant improvements in symptom measures in favour of steroids, and one study reported a difference in growth rate in favour of xanthine. No difference was observed for study withdrawal or tremor. Xanthine was associated with more frequent headache and nausea. Xanthine versus regular SABA (10 studies): No significant difference in symptoms, rescue medication usage and spirometry. Individual studies reported improvement in PEF with beta-agonist. Beta-agonist treatment led to fewer hospitalisations and headaches. Xanthine was associated with less tremor. Xanthine versus SCG (six studies ): No significant difference in symptoms, exacerbations and rescue medication. Sodium cromoglycate was associated with fewer gastro-intestinal side-effects than xanthine. Xanthine versus KET (one study): No statistical tests of significance between xanthine and ketotifen were reported. : Xanthine + ICS versus placebo + same dose ICS (three studies) : Results were conflicting due to clinical/methodological differences, and could not be aggregated. Xanthine + ICS versus ICS + leukotriene (one study): Results from one trial One small parallel study did not measure the primary outcome of symptoms; differences between treatments in end of treatment values were not statistically significant.