Displacement by plasma of radiolabeled thrombin and radiolabeled thrombin-antithrombin III inactive complex from a heparinized surface (heparin-PVA) was measured and found to be significant. For example, 63% of the thrombin and 90% of the complex that could not be removed by PBS alone was displaced by heat defibrinated plasma. Preliminary characterization (molecular weight, antithrombin III content) suggests that the eluting product consists largely of thrombin-antithrombin III complex and post complex antithrombin III. Heparin polyvinyl alcohol (PVA) gel beads were prepared by acetal coupling of the heparin to PVA using glutaraldehyde with MgCl2 catalysis. Although permanently bound to the PVA (heparin removal rate was 1.67 × 10-2 mg/g gel⋅min), the heparin retained at least part of its activity in thrombin time, recalcification time, chromogenic substrate and AV shunt assays. Thus, heparin need not be lost from a surface to impart thromboresistance. Our results further suggest that the resulting surface bound thrombin-antithrombin III complex can be displaced from the surface by a component or components of plasma (antithrombin III?) indicating that the surface bound heparin does not become saturated with inactive complex. These results support the argument that heparinization can be an important means of preparing the materials needed for the development of improved cardio-circulatory assist devices and blood handling procedures.