Equilibrium dialysis for determination of protein binding or imipramine--evaluation of a method.

A range of methodological problems associated with equilibrium dialysis for serum or plasma protein binding assay of a weakly basic drug imipramine, in man has been carried out in order to establish a reliable procedure for large scale studies. Freezing of serum samples did not change the binding significantly, but a slight tendency towards increased binding to samples stored for 30 days was found. Protein binding was significantly higher in serum than in heparinized or citrated plasma, whereas there was no significant difference between plasma obtained by use of heparin or citrate as anticoagulant. There was a linear correlation between temperature (25-42 degrees) and binding such that binding decreased 0.1% per centigrade increase in temperature. Increase in pH resulted in increase in binding (0.6% per 0.1 unit pH). Equilibrium between serum and buffer phase was obtained within 5-6 hrs. The changes following osmotic equilibrium necessitated volume corrections of about 4-5% after 5-6 hrs, and 20-25% after 16-20 hrs (overnight experiments). With this correction the free fraction estimate remained constant for 26 hrs. Application of small volumes in the dialysis chamber may make the volume correction difficult and inaccurate. The use of Tris-Ringer buffer gave free fraction values about 20% higher (absolute values of free fraction) than that with phosphate buffer. With different imipramine concentrations a constant degree of binding was found at "therapeutic" levels, whereas a significantly decreased binding was found at a concentration of 1000 ng/ml. Control of these methodological factors is a prerequisite for obtaining reproducible and valid results, and authors should give more details on methodology in this field.