Evaluation of the in vitro and in vivo performance of two sustained-release lithium carbonate matrix tablets. Effect of different diets on the bioavailability.

ABSTRACT

Two sustained-release (SR) lithium carbonate (Li) matrix tablets, which use a hydrophilic (HP) matrix of hydroxypropylmethylcellulose (Methocel 4K MP) and a lipid (L) matrix of hydrogenated castor oil (Cutina HR) as sustaining agents, have been studied. In vitro performance through dissolution tests in different media was established. The L and HP formulations were affected by the composition of the dissolution media, and liberation was complete in 8 hr using a variable-pH medium that simulates the gastrointestinal (Gl) pH. Liberation was better described by the diffusional model of the square root of time for the L matrix and by zero-order kinetics for the HP matrix. Absolute bioavailability (BA) and food-induced changes on BA of both formulations were studied. The in vivo study design was a 4 x 4 Latin square involving 12 subjects who received two tablets of a 300-mg dose of SR formulations while fasting or with a standardized normal, high-fat, or high-fat/high-protein meal. The results for both formulations showed no differences in the disposition parameters and mean residence time when the tablets were administered with any type of diet. Changes in rate of absorption were found when both types of tablets were administered with any class of diet. The analysis of the ratio Cmax/AUC (area under the curve) evidenced that changes in Cmax were attributable to a higher rate of absorption for the HP matrix and to a higher amount absorbed for the L matrix. In the last, high-fat and high-fat/high-protein diets produced higher AUCs than under fasting condition. The SR Li tablets formulated with hydrogenated castor oil were affected more by high-fat food, probably because of the increase of pancreatic and biliary secretions promoted by the meal, which would affect the matrix itself. The HP matrix was also affected, but to a lesser extent. The magnitude of the change in Cmax observed with this matrix probably is not important from a clinical point of view. Absolute BA was very low for the lipid matrix; in addition, since it is more seriously affected by food, probably it is not a good choice for a drug such as lithium. The in vivo behavior of the HP matrix makes it advisable to invest in efforts to achieve increased BA. Comparing in vitro and in vivo results, the focus should be achieving sustained, but complete, in vitro liberation in not more than 3 hr, with simulation of the transit time through the stomach and small bowel since lithium ion is only absorbed to this point.