OBJECTIVE Symptomatic cerebral vasospasm is a major complication in patients with subarachnoid hemorrhage (SAH). Symptomatic cerebral vasospasm has been reported to be related to the patient's blood volume which is influenced by cerebral salt wasting syndrome (CSWS). We undertook a prospective study to assess whether the onset of symptomatic cerebral vasospasm was predictable or not, by observing the phenomena of CSWS (natriuresis and osmotic diuresis). METHODS Sixty-seven consecutive aneurysmal SAH patients were analysed. After surgery, all patients underwent hypervolemic therapy in order to keep central venous pressure (CVP) within 8-12 cmH(2)O, serum sodium level above 140 mEq/l and a positive water balance. Patients were classified into two groups: those without symptomatic cerebral vasospasm (n=55) and those with symptomatic cerebral vasospasm (n=12). To estimate natriuresis and osmotic diuresis, sodium in/out, water in/out, CVP and other parameters were measured and compared between the two groups. RESULTS One day before symptomatic cerebral vasospasm, three factors reached statistical difference in the group that experienced symptomatic cerebral vasospasm: sodium balance, urine volume and water balance. On the day of symptomatic cerebral vasospasm, two factors reached statistical difference: sodium excretion and urine volume. No factor was significantly different 2 days before symptomatic cerebral vasospasm. CONCLUSIONS Symptomatic cerebral vasospasm has a strong relationship with CSWS. Negative sodium and water balance and increased urine volume indicate a predictor of symptomatic cerebral vasospasm. To predict symptomatic cerebral vasospasm, strict observations are required, because CSWS and symptomatic cerebral vasospasm which follows, develop rapidly.