We examined whether or not any difference in erythrocyte aggregability (RBC-A) exists between stroke patients with diabetes mellitus and those without diabetes mellitus. The subjects comprised 117 patients at the chronic phase of cerebral infarction (at least 2 months after onset). The patients were divided into two groups: group A, those with cerebral infarction and diabetes mellitus (N=35), aged 60+/-10 YO (mean+/-SD); and group B, those with cerebral infarction but without diabetes mellitus (N=82), aged 61+/-7 YO. For comparison with the erythrocyte aggregability in the stroke patients, we also undertook measurements in 52 age-matched healthy human volunteers (59+/-9 YO: the control group) and 44 age-matched patients with diabetes mellitus without macroangiopathy (60+/-7 YO; the DM group). The erythrocyte aggregability was determined using the whole blood erythrocyte aggregometer developed by us previously (Am J Physiol 251:H1205-H1210, 1986) with concomitant estimation of blood factors such as the hematocrit, albumin:globulin ratio and concentration of fibrinogen. The RBC-A values were 0.164+/-0.024/s in group A, 0.148+/-0.024/s in group B, 0.122+/-0.027/s in the control group and 0.142+/-0.023/s in the DM group, respectively. The RBC-A values in group A, group B and the DM group were significantly (p<0.01, p<0.01, p<0.01) higher than that in the control group, and the RBC-A value in group A was significantly (p<0.01) higher than that in the DM group. A significant difference (p<0.01) in RBC-A values was noted between groups A and B. Although the values of the hematocrit and albumin:globulin ratio did not differ between groups A and B, the concentration of fibrinogen in group A (350+/-79 mg/dl) was significantly (p<0.01) hi-her than that in group B (313+/-81 mg/dl). The above results suggest that diabetes mellitus exerted a deleterious effect on the patients with cerebral infarction from the hemorheological standpoint.