In vivo-in vitro correlations with sustained-release theophylline preparations

The in vitro dissolution of four commercially available sustained-release theophylline products was studied. The dissolution of theophylline was evaluated under four conditions: USP-method I or Resotest apparatus, and artificial gastric (pH 3) or artificial intestinal (pH) fluids. The resulting data were evaluated by graphic, first-order transform, dissolution efficiency, non-linear fitting, mean time of dissolution and Weibull analysis methods. The four theophylline preparations and a peroral solution were administered to four male beagle dogs in a cross-over fashion. Parameters were obtained from graphed concentration vs. time data, a one-compartment open model fit, Wagner-Nelson absorption plots, ratios of half values of duration and maximum concentrations, mean times of absorption and analog computer evaluation. Correlations were made between all possible combinations of in vitro and in vivo data by quadrant analysis, Pearson product moment, r, and Spearman rank order correlation methods. A total of 8280 correlation coefficients were calculated. The optimal set of studied conditions for in vivo-in vitro correlations with sustained-release theophylline preparations is the time to reach maximum concentration in the blood (from the fitted data) vs. the time for 15% of the drug to be released into solution from the dosage form (from graphed data) using either the USP I apparatus or the Resotest.