We evaluated the use of a standard protocol utilizing premixed theophylline containers in comparison to traditional intravenous aminophylline loading doses in 19 critically ill adults. Blood samples were obtained immediately before and after 30-minute infusions of aminophylline and theophylline bags. Aminophylline loading doses were calculated to the exact mg/kg and were delivered as such. Calculation of the theophylline loading dose used a simplified protocol that uses doses rounded to the nearest 100 mg. Comparison of the relative performance of the two loading dose methods was evaluated by computing mean squared prediction errors and root mean squared errors. The two methods were not statistically different when evaluated by a matched pair t-test. We conclude that use of a standard protocol that rounds theophylline doses to the nearest 100mg results in peak theophylline concentrations not significantly different from those predicted.