Theophylline absorption from three sustained-release formulations was evaluated in 18 healthy men. First, a 600-mg oral dose of aminophylline, a rapidly absorbed formulation, was studied to confirm that theophylline clearance was in the expected range (mean +/- SD, 0.710 +/- 0.096 mL/min/kg), t1/2 = 6.9 +/- 1.1 hr. Then each of the three sustained-release formulations was given at one hour before breakfast for eight days to achieve steady state. Multiple blood samples were drawn on days 7 and 8 over two complete dosing intervals. After a six-day drug-free period each subject was crossed over to the next product. At steady state, mean fluctuations of serum theophylline levels were more than 200% with Theo-Dur (Key Pharmaceuticals, Miami, FL), more than 150% with Uniphyl (Purdue-Frederick, Norwalk, CT), and about 75% with Theo-24 (Searle Laboratories, Chicago, IL). Theophylline Cmax levels averaged 16.5 micrograms/mL after 900 mg Theo-Dur, with 8% exceeding 20 micrograms/mL. Despite these high peak levels, nearly half of the Cmin levels were below 5 micrograms/mL. Uniphyl administration resulted in three-fourths of trough levels to be at 5 micrograms/mL or lower. Peak levels from 800-mg daily doses of Uniphyl were also low, with 42% never reaching 10 micrograms/mL. With Theo-24 only 17% of trough levels were below 5 micrograms/mL, as compared to 47% with Theo-Dur and 72% with Uniphyl. The three products differed in the areas under the serum concentration-time curves (AUC), which reflect bioavailability. Theo-Dur AUC was nearly equivalent to that of rapidly absorbed aminophylline (when adjusted for dosage).(ABSTRACT TRUNCATED AT 250 WORDS)