In vitro estimation of drug loss during the implantation procedure of drug-eluting stents

Abstract

Drug-eluting stents are drug/device combinatory products designed to physically re-establish the blood flow in arteriosclerotic blood vessels and deliver anti-proliferative drugs to the stented vessel section to prevent neointimal hyperplasia. Drug that is released during the implantation procedure is washed away with the blood flow. It was our aim to estimate this drug loss using stents coated with fluorescent model substances in an in vitro setup. Stents mounted on balloon catheters were introduced into a polymethacrylate coronary artery model (adapted from ASTM standard F 2394-07 intended for measurement of securement of stents on delivery systems) via a guiding catheter. The system was perfused by phosphate buffered saline pH 7.4 at a flow-rate adapted to the blood flow-rate in coronary arteries. The perfusate was collected in fractions and model substance content was determined fluorimetrically. The hydrophilic model substance fluorescein sodium was released very fast resulting in a loss of approximately 64 % of the drug load within the first minute and up to 82 % after 5 minutes. The hydrophobic model substance triamterene was released much slower from the coating. A total loss of 5 % was detected after 5 minutes. Furthermore, commercially available sirolimus-eluting Cypher Select ®+ stents were tested. Release into the media was not detected. Obviously, the drug-free top coating effectively prevented drug release during the simulated passage to the site of implantation. An in vitro test system to estimate drug loss during the implantation procedure of drug-eluting stents was successfully established. First results using fluorescent model substances indicate that vast losses have to be expected when using fast releasing delivery systems. Therefore, the model may also prove valuable for the in vitro testing of drug-eluting balloons. Further adaptations, such as the inclusion of biorelevant media, may be suitable to further improve the predictability of in vivo performance. © 2012 by Walter de Gruyter. Berlin. Boston.