We compared measurements of venous blood flow velocity in the superior sagittal sinus (SSS) obtained by magnetic resonance imaging (MRI) and quantitative measurements of cerebral blood flow (CBF) obtained by xenon-enhanced CT, in neurosurgical patients and normal volunteers, to assess the clinical usefulness of MRI for determination of CBF. Flow measurements were obtained in 15 neurosurgical patients and 3 normal volunteers. SSS velocimetry was performed using the direct bolus imaging technique with a 1.5-tesla MRI system near the point of lambda on a sagittal MR image. Quantitative CBF was measured by xenon-enhanced CT and correlated with SSS flow velocity. We also examined the effect of surgical evacuation of a hematoma or cranioplasty in 3 patients, mannitol in 5 patients and acetazolamide in 4 patients by performing SSS velocimetry. The peak SSS flow velocity during the entire cardiac cycle was 27.2 +/- 6.3 cm sec-1 (mean +/- SD) in normal volunteers and 23.5 +/- 8.9 cm sec-1 in patients. The SSS flow velocity increased after surgery. Mean SSS peak flow velocity increased by 24% and 48% at 10 min after administration of mannitol and acetazolamide, respectively. SSS peak flow velocity showed significant linear correlations with both cerebral cortical (r = 0.74) and hemispheric flows (r = 0.73). Our results suggest that SSS flow velocity reflects CBF and that MRI assessment of SSS flow provides a simple method for assessing and monitoring global changes in cerebral hemodynamics.