Multicenter Trial of Hemodilution in Acute Ischemic Stroke

Hemodilution by the combination of venesection and dextran 40 administration has previously been reported to enhance neurologic recovery in the acute phase of ischemic stroke. To study this therapeutic principle in its "natural habitat," a stratified and randomized multicenter trial involving 15 large and small hospitals was performed. Patients with acute ischemic stroke of <48 hours' duration and with hematocrits of 38-50% on admission were randomized to a hemodilution (183 patients, mean age 72.0 years) or a control group (190 patients, mean age 71.6 years). The two groups did not differ in sex distribution or medical history. Hematocrit, blood pressure, and neurologic score were closely similar at entry. By graded venesection (250-1000 ml) during the first 2 days and dextran 40 infusions (500 ml daily) during 5 days, the mean hematocrit was reduced from 44.2 to 37.1%. Three-month survival expressed as life table product was 0.84 in hemodilution and 0.88 in control patients. In survivors, neurologic score and activities of daily living performance during 3 months of follow-up were not improved by hemodilution. Length of stay in an acute-care hospital and the need for long-term institutional care was not reduced among patients in the hemodilution group. Major cardiovascular events occurred somewhat more often and there was an apparent increase in mortality during the first few days of hemodilution therapy. However, the differences were not significant. We conclude that the present standardized treatment with moderate hemodilution has no overall beneficial effects in genera] patients with acute ischemic stroke. (Stroke 1987; 18:691-699)