The purpose of this work was to describe the closed loop in situ perfusion method in rats and to compare the difficulties and advantages with other methods proposed by regulatory agencies for BCS classification and finally to illustrate its application to evaluate the permeability of digoxin at relevant clinical concentrations. Digoxin was evaluated at two concentration levels: 1.0 μg/ml (with and without sodium azide 65.0 μg/ml) and 6.0 μg/ml. These concentrations correspond to the ratio of the highest dose strength (0.25 mg) and the highest single dose administered (1.5 mg) and the 250 ml of water. In situ closed loop perfusion studies in rats were performed in the whole small intestine and also in duodenum, jejunum and ileum segments to evaluate the relevance of P-gp secretion in the overall permeability. A kinetic modelling approach involving passive permeation and efflux transport mechanism allowed the estimation of the passive diffusional component and the Michaelis-menten parameters. The estimated Km value demonstrated that at clinical luminal concentrations the efflux process is not saturated and then it could be inhibited by other drugs, excipients or food components leading to the already reported clinical drug-drug and drug-food interations. The present data confirms from a mechanistic point of view these interactions.