Current in vitro drug-release testing of the sustained-release parenterals represents the in vivo situation insufficiently. In this work, a thin agarose hydrogel layer surrounding the tested dosage form was proposed to mimic the tissue. The method was applied on implantable formulations of different geometries (films, microspheres, and cylindrical implants); prepared from various polymers (several Resomer® grades or ethyl cellulose) and loaded with different model drugs: flurbiprofen, lidocaine or risperidone. The hydrogel layer did not possess any retarding effect on the released drug and acted as a physical restriction to swelling and/or plastic deformation of the tested dosage forms. This led to a different surface area available for drug-release compared with testing in release medium alone and correspondingly to significantly different release profiles of the majority of the formulations obtained between the two methods (e.g. t50% = 18 days in pure release medium vs. t50% = 26 days in gel-setup for risperidone loaded Resomer® 503 H films or t50% = 7 days vs. t50% = 19 days for risperidone loaded Resomer® 503 H microspheres). The limited space for swelling and the rigidity of the agarose gel might mimic the tight encapsulation of the dosage form in the tissue better than the conventional liquid medium.

Department of Pharmaceutical Technology Faculty of Pharmacy Charles University Akademika Heyrovskeho 1203 8 500 05 Hradec Kralove Czech Republic

Department of Pharmaceutics Institute of Pharmacy University of Bonn Gerhard Domagk Straße 3 53121 Bonn Germany