The bioavailability of diazepam from uncoated tablets in humans--Part I: correlation with the dissolution rates of the tablets.

Dissolution studies of 15 preparations of commercial uncoated tablets of diazepam (5 mg) were performed by six methods (beaker, rotating basket, oscillating basket, solubility simulator, rotating flask, and single basket). Diazepam dissolved rapidly at pH 1.2; the T50 (the time of 50% dissolution) values were less than 5 min. But at pH 4.6, T50 estimated by rotating basket method lasted 3-120 min. Four different tablets of diazepam were chosen for the bioavailability tests in humans. The bioavailabilities of the four tablet preparations were estimated by serum level measurements after a single dose to 12 adult male volunteers. Statistical analysis of the data showed significant differences in the rate of bioavailability (peak concentrations and serum concentrations at 1, 2, and 3 h after administration), but not in the amount of available (AUC). The mean peak concentration and serum concentration at 1 h showed significant correlation with T50 and T70 determined by the rotating flask method at pH 4.6 in log-log regression. The peak concentration and serum concentration at 1 h were also correlated with T70 determined by the rotating flask method at pH 4.6 and T70 determined by the rotating basket method at pH 4.6 on normal-normal regression. In contrast, the dissolution rates determined at pH 1.2 did not show a good correlation with in vitro parameters.