A randomized controlled trial of hemodilution therapy in acute ischemic stroke.

Rapid hemodilution in the early phase of ischemic stroke by the combination of venesection (250-650 ml during the first 2 days) and administration of low-molecular weight dextran was evaluated in a prospective controlled trial. Fifty-two patients were randomized to hemodilution therapy and 50 to a control group; the two groups were comparable in important prognostic variables. Mean hemoglobin was reduced from 147 to 127 g/l, hematocrit from 43 to 37% and, in a subsample of patients, whole-blood viscosity at a shear rate of 23 sec-1 from 7.0 to 4.3 cps over the first 2 days. Hemodilution was then maintained by repeated dextran infusions. Of the hemodiluted patients, 85% improved in neurological scoring over the first 10 days as compared to 64% of the control patients (p less than 0.025). The case fatality rate during the first 3 months was little affected by hemodilution. Among the survivors, 8% of the hemodiluted and 31% of the non-hemodiluted patients were unable to walk at 3 months. The proportion of surviving patients still hospitalized at the 3-month follow-up was 13% in the hemodilution group and 39% in the control group (p less than 0.01). The combination of venesection and dextran 40 administration is thus an unsophisticated but effective way to achieve rapid hemodilution in patients with acute cerebral infarction, and it improves the overall clinical outcome over the first 3 months.