A NOVEL SKELETAL DRUG-DELIVERY SYSTEM USING SELF-SETTING CALCIUM-PHOSPHATE CEMENT .7. EFFECT OF BIOLOGICAL FACTORS ON INDOMETHACIN RELEASE FROM THE CEMENT LOADED ON BOVINE BONE

The use of self-setting bioactive calcium phosphate cement containing indomethacin as a model drug in bovine bone was investigated by means of an in vitro drug release test, mercury porosimetry, and scanning el ectron microscopy (SEM). Calcium phosphate cements containing 2 and 5% indomethacin after being mixed with dilute phosphoric acid were appli ed to defect sites and the medullary cavity of bovine bone and transfo rmed into hydroxyapatite. The in vitro drug release from the cement lo aded on the defect site into a simulated body fluid (SBF) containing 2 .5 mM Ca2+ and 1.0 mM HPO42+ or 0.1 M phosphate buffer at pH 7.25 and 37 degrees C continued for more than 3 weeks. The release profiles of the drug-loaded cements in phosphate buffer were linear using the Higu chi plot; however, that was not the case for SBF. The drug release in SBF was much lower than that in phosphate buffer. The total pore volum e of the cement after the drug release test in SBF was lower than its initial value. However, the pore size of 0.1-0.01 mu m after drug rele ase in phosphate buffer was higher than that seen in SBF. The micropor e distribution results suggested that hydroxyapatite crystallized from SBF and the pore volume in the cement decreased after drug release. H owever, in phosphate buffer it appeared to dissolve. The SEM observati ons for cements loaded on the bone after drug release in phosphate buf fer suggested that there was a boundary layer between the cement and n atural bone, but this was not the case in SBF, where the cement bonded with the natural bone. The drug release rates from the cement-loaded bone were significantly higher than those from cement loaded on the di ssolution holder. The results suggested that cement formation and drug release were affected by the presence of protein from natural bone. T he drug release rates from the cement loaded on the defective bone wer e slower than those from the medullary cavity.