Modification of in vitro drug release rate from oily parenteral depots using a formulation approach

Rate constants for transfer of naproxen and lidocaine from different oils and oil mixtures to aqueous buffer, pH 6.00, were determined using the rotating dialysis cell. Significantly different first-order rate constants related to attainment of equilibrium, k(obs), were derived depending on the type of oil/oil mixtures used in the release experiments. For the drugs a linear correlation was found between logk(obs) and the logarithm of the partition coefficient P(app): logk(obs)=-0.68 logP(app)-0.25 (k(obs) in h(-1), n=26). A linear relationship was observed between the calculated and experimentally determined P(app) values for the oil mixtures investigated. The specific rate constants, k(ow) and k(wo), related to the partition process were derived from the determined k(obs) and P(app) values. The rate constant k(ow) representing the rate of transfer of the solute from the oil phase to the aqueous buffer was shown to be strongly dependent on the partition coefficient according to the relationship: logk(ow)=-0.68 logP(app)-log(P(app)+1)-0.25 (k(ow) in h(-1), n=26). In particular, diminished release rates were seen for oil mixtures containing castor oil most likely afforded by hydrogen bonding between the solute and the hydroxy groups of the latter vegetable oil. In this study it has been possible to alter P(app) for a specific compound up to a factor of 10 by variation of the composition of the oil vehicle. Such a span of P(app) values results in in vitro release rates differing a factor of 37. Thus, by proper design of the oil vehicle composition it should be possible to modify the release rate for a specific compound within certain limits.