Bone implants modified with cyclodextrin: study of drug release in bulk fluid and into agarose gel.

The aim of this work was to better understand the importance of the type of experimental setup used to monitor antibiotic release from functionalized hydroxyapatite implants. Microporous hydroxyapatite discs were prepared by sintering and subsequently functionalized with hydroxypropyl-β-cyclodextrin (HPβCD) polymer crosslinked with butanetetracarboxylic acid. On one hand, polymerization was performed within the implant after its impregnation with the monomers (CD-HA-M implant). On the other hand, a pre-synthesized HPβCD polymer was loaded and fixed onto the HA discs (CD-HA-P implant). Both types of implants were soaked with ciprofloxacin hydrochloride or vancomycin hydrochloride solution and dried at 37°C. The DSC study highlighted that the cyclodextrin polymer could interfere with both drugs, due to the carboxylic groups carried by the crosslinks. Drug release was measured into phosphate buffered saline pH 7.4 in agitated vials, or into agarose gels to more realistically mimic in vivo conditions. Importantly, in all cases, drug release into agarose gels was much slower than into well-agitated phosphate buffer. Non-functionalized discs displayed faster drug release because no complex could be formed and/or due to the absence of the HPβCD polymer network hindering drug diffusion within the implant pores. In the case of ciprofloxacin hydrochloride, drug release from the CD-HA-M implants was faster than drug release from the CD-HA-P implants due to the different polymer structures resulting in different complexation strengths, whereas in the case of vancomycin hydrochloride the release patterns were similar because vancomycin hydrochloride was not included into the cyclodextrin. The agarose gel method seems more biorelevant and discriminatory than the vial method for drug release measurements from bone implants.