A possible role of methylxanthines in the high incidence of gastroesophageal reflux (GER) in patients with asthma has been suggested. Therefore, we used a randomized, double-blind, crossover design to compare the effects of a 1-week conventional theophylline treatment and a 1-week placebo treatment in 16 adult patients with asthma. No oral or parenteral glucocorticoids were administered, but seven patients were taking inhaled corticoids. All patients needed inhaled adrenergic drugs. At the end of each period of theophylline or placebo treatment, the patients were carefully questioned with respect to respiratory and digestive symptoms, forced expiratory flows were measured, and GER was assessed by prolonged nocturnal intraesophageal pH monitoring. Peak expiratory flow was measured three times a day throughout the study. No significant increase in GER was found with theophylline compared to placebo, and forced expiratory flows improved with theophylline (p less than 0.05 for FEV1 and p less than 0.01 for peak expiratory flow rate). There was no correlation between GER, the duration of asthma, and forced expiratory flows. Thus, our study failed to demonstrate any adverse effect of a slow-release theophylline preparation on GER in patients with asthma. These results further suggest that the presence of GER is not a contraindication to the use of a slow-release theophylline in subjects with asthma.