Control of gentamicin release from a calcium phosphate cement by admixed poly(acrylic acid)

The aim of this work was to develop a calcium phosphate cement (CPC) providing controlled release of the antibiotic gentamicin sulfate (GS) over at least 1 week. The CPC was made of beta-tricalcium phosphate [beta-TCP; beta-Ca(3)(PO(4))(2)], monocalcium phosphate monohydrate [MCPM; Ca(H(2)PO(4))(2). H(2)O] and water. Release of GS was controlled by admixture of poly(acrylic acid) (PAA). The effects on the GS release kinetics of the molecular weight of PAA, of the amount of admixed PAA, and of the pH of the release medium were investigated. A typical cement sample weighed 3.6 g and contained 100 mg of GS and between 0 and 150 mg of PAA. In the following, PAA content is expressed as the weight ratio, lambda, with respect to GS. At a low PAA content in the CPC (lambda < 0.7), GS was released over 1-2 days according to a square-root-of-time kinetics, but not all GS was released. The unreleased GS fraction increased from 0 to 58% with an increase of PAA content (up to lambda = 0.7). At high PAA content (lambda > 0.7), GS was released over a period of up to 8 days according to a combination of a square-root-of-time and a zero-order kinetics. The total GS fraction released increased again from 58 to 100% with an increase of the amount of PAA (up to lambda = 1.5). These observations were explained by molecular interaction between PAA and GS resulting in gel formation. The maximum fraction of GS released from the cement was indeed a function of the solubility of the PAA-GS (coacervate) complex in the release medium. Thus, GS release was controlled by two mechanisms: (1) diffusion of free GS molecules through the porous cement (square-root-of-time kinetics); and (2) dissociation of GS from the PAA-GS complex (zero-order kinetics). The first mechanism was predominant at low lambda, whereas the second mechanism became important at high lambda and later release times. As the solubility of the PAA-GS complex decreased with an increase in PAA molecular weight, the higher molecular weight PAA yielded more prolonged release periods of up to 8 days. Interestingly, the use of 450 kDa PAA at lambda = 1.00 provided an almost constant release profile over a period of 7 days. Gel formation between PAA and GS was explained in terms of hydrogen bonding of PAA carboxyl groups with GS amino groups. The molar ratio between carboxyl groups and amino groups in the gel was estimated to be approximately 1.9. In conclusion, admixture of PAA into calcium phosphate cement appeared to be a very elegant tool to control the release of the antibiotic over a period of 7 to 8 days.