Prediction of intravenous theophylline dosage based on a single, nonsteady-state concentration: a clinical study of childhood status asthmaticus.

A pharmacokinetic model was applied to achieve therapeutic serum theophylline concentrations rapidly in 25 children with status asthmaticus. A sustained release theophylline preparation had been taken within 36 hours by 12 children; within 14 hours, seven had taken an immediate release preparation; for six children, no theophylline was taken before hospital admission. Single serum theophylline concentrations were determined at nonsteady-state conditions within 13.5 hours of admission (median 6.75 hours). An iterative program was applied to predict the steady-state theophylline concentration as well as necessary adjustments in dosage. Measured steady-state concentrations were then compared with the predicted values. The median measured steady-state concentration was 15 mg/L, and the median predicted steady-state level was 13 mg/L. The least squares regression line was: Measured = 0.738 predicted + 4.77; r = .721, P less than .01. No patient experienced symptoms of toxicity. This technique affords the possibility of accurate prediction of steady-state theophylline concentrations and dosing requirements with a minimum number of serum concentration determinations in children with status asthmaticus.