We performed a randomized, double-blind, pilot study on magnesium sulfate (MgSO4) infusion for aneurysmal subarachnoid hemorrhage (SAH).Sixty patients with SAH were randomly allocated to receive either MgSO4 80 mmol/day or saline infusion for 14 days. Patients also received intravenous nimodipine. Episodes of vasospasm were treated with hypertensive and hypervolemic therapy. Neurologic status was assessed 6 months after hemorrhage using the Barthel index and Glasgow Outcome Scale. Incidences of cardiac and pulmonary complications were also recorded. Patient characteristics, severity of SAH, and surgical treatment did not differ between groups. The incidence of symptomatic vasospasm decreased from 43% in the saline group to 23% in patients receiving MgSO4 infusion, but it did not reach statistical significance, P=0.06. For patients who had transcranial Doppler-detected vasospasm, defined as mean flow velocity >120 cm/s and a Lindegaard index >3, the duration was shorter in the magnesium group compared with controls (P<0.01). There was, however, no difference between groups in functional recovery or Glasgow Outcome Scale score. The incidence of adverse events such as brain swelling, hydrocephalus, and nosocomial infection was also similar in patients receiving MgSO4 or saline. In this small pilot study, MgSO4 infusion for aneurysmal SAH is feasible. On the basis of the preliminary data, a larger study recruiting approximately 800 patients is required to test for a possible neuroprotective effect of magnesium after SAH.