Drug-coated balloons are medical devices designed to locally deliver drug to diseased segments of the vessel wall. For these dosage forms, drug transfer to the vessel wall needs to be examined in detail, since drug released into the blood is cleared from the site. In order to examine drug transfer, a new in vitro setup was developed combining the estimation of drug loss during advancement to the site of application in a model coronary artery pathway with a hydrogel compartment representing, as a very simplified model, the vessel wall. The transfer of fluorescent model substances as well as the drug paclitaxel from coated balloons to the simulated vessel wall was evaluated using this method. The model was suitable to quantify the fractions transferred to the hydrogel and also to qualitatively assess distribution patterns in the hydrogel film. In the case of fluorescein sodium, rhodamin b and paclitaxel, vast amounts of the coated substance were lost during the simulated passage and only very small fractions of about 1% of the total load were transferred to the gel. This must be attributed to good water solubility of the fluorescent substances and the mechanical instability of the paclitaxel coating. Transfer of the hydrophobic model substance triamterene was however nearly unaffected by the preliminary tracking procedure with transferred fractions ranging from 8% to 14%. Analysis of model substance distribution yielded inhomogeneous distributions indicating that the coating was not evenly distributed on the balloon surface and that a great fraction of the coating liquid did not penetrate the folds of the balloon. This finding is contradictory to the generally accepted assumption of a drug depot inside the folds and emphasizes the necessity to thoroughly characterize in vitro performance of drug-coated balloons to support the very promising clinical data.