Two experimental formulations of theophylline with a hydrophilic starch matrix were evaluated for their sustained-release characteristics after single administration in healthy human volunteers. Theo-dur was chosen as a reference sustained-release formulation. In a first study, the extent of absorption was similar for a syrup, for Theo-dur, and for the experimental formulation of theophylline with 70% drum-dried corn starch as the sustained-release agent (DDCS-70). The maximal plasma concentration (Cmax) was significantly lower, and the time to reach Cmax as well as the time span during which the plasma concentration was at least 75% of the Cmax were significantly higher for Theo-dur than for the DDCS-70 formulation. A sustained-release profile, as for Theo-dur, was not reached for DDCS-70. In a second study the influence of the starch:drug ratio on the bioavailability was investigated. The decrease in starch content from 70 to 50% of the formulation did not improve the plasma concentration-time profile towards a sustained-release profile.