BACKGROUND Recurrent apnea is common in preterm infants, particularly at very early gestational ages. These episodes of loss of effective breathing can lead to hypoxemia and bradycardia which may be severe enough to require resuscitation including use of positive pressure ventilation. Doxapram and methylxanthine drugs have been used to stimulate breathing and so prevent apnea and its consequences. OBJECTIVE In preterm infants with recurrent apnea, how does treatment with doxapram compare with treatment with theophylline in leading to a clinically important reduction in apnea and use of mechanical ventilation, without clinically important side effects. METHODS The standard search strategy of the Neonatal Review Group, as outlined in the Cochrane Library, was used. METHODS All trials utilising random or quasi-random patient allocation, in which doxapram was compared with methylxanthine (e.g. theophylline) for the treatment of apnea, were eligible. There must have been an effort to exclude specific causes of apnea. METHODS The standard methods of the Cochrane Collaboration and its Neonatal Review Group were used to select trials, assess quality and to extract and synthesize data. The methodological quality of each trial was reviewed by the second author blinded to trial authors and institution(s). Additional information was requested from authors to clarify methodology. Each author extracted the data separately, then they were compared and differences resolved. Meta-analysis was carried out with use of relative risk and risk difference. RESULTS In these trials involving a relatively small number of preterm infants with apnea of prematurity, there is no apparent difference between the effect of intravenous treatment with doxapram or methylxanthine on the incidence of apnea within 48 hours. There were no infants reported to have been given mechanical ventilation on either treatment. No adverse effects were reported. CONCLUSIONS Implications for practice. The overall results of these small trials suggest that intravenous doxapram and intravenous methylxanthine are not different in their effectiveness in the short term in the treatment of apnea of prematurity. Caution is warranted as the number of patients in these trials is too small to exclude an important difference between these two treatments or to exclude the possibility of less common side effects. Longer term outcome of infants treated in these trials has not been reported. Implications for research. Further studies would require a large number of infants, stratified by gestation, to clarify which infants are likely to benefit and whether there might be differences in responses or side effects with these two drugs at different ages. It would be valuable to include important clinical outcomes such as use of mechanical ventilation as well as subsequent growth and development in future studies. Responses to treatment would have to take account of co-interventions, such as nasal continuous airway pressure which is frequently used post-intubation.