The pharmacokinetic interaction between betamethasone (5 mg/day for 5 days) and theophylline (3.35 mg/kg orally before and after betamethasone treatment) was studied in 13 female patients evaluating saliva and serum theophylline concentrations by means of an enzyme multiplied immunoassay technique. Betamethasone treatment did not significantly change saliva and serum theophylline levels. As regards the potential usefulness of saliva theophylline levels in predicting the concomitant serum concentrations of the bronchodilator, in spite of the significant correlation found in our study (r = 0.758; P less than 0.001), we think that, using individual saliva values, there is a poor predictive value for serum theophylline concentrations.