The clinical effect of bag-plasmapheresis was investigated in 60 patients with peripheral arterial occlusive disease stage II according to Fontaine. The initial number of patients was subdivided in three groups of 20 individuals using a randomised double-blind placebo-controlled design. Each patient gave 300 ml of blood twice a week for a 6 week duration. Blood plasma was separated in two groups and replaced with Hydroxyethyl-starch (200/0.5 10%) in group 1 and with Laevulose 5% in group 2. Patients in group 3 received their whole blood without any processing. All patients had to undergo a physical training of 45 minutes three times a week. The group who received Hydroxyethylstarch presented a 20% increase in walking distance whereas the increase in the Laevulose group was 5% and approximately 1% in the group receiving whole blood. The increase in walking distance in the Hydroxyethylstarch-group was significant on the 0.1%-level and significantly better than the improvement in walking distance of the other groups. Additionally in this group plasma viscosity showed a 3% decrease, erythrocyte aggregation was reduced by 10%. Results in the Laevulose group were only half as good as in the Hydroxyethylstarch group while parameters remained unchanged in the whole-blood-group. Bag plasmapheresis with Hydroxyethylstarch as substitute leads to an improvement in the walking capacity and blood fluidity thus offering a promising therapy for peripheral vascular occlusive disease.