Fairly large volumes of intracarotid mannitol (20% to 25%) are required to disrupt the blood brain barrier (BBB), that is, 200 to 300 mL/30 s in humans or 10 mL/40 s in rabbits. During transient cerebral hypoperfusion blood flow to the rabbit brain is decreased to 0.2 to 0.3 mL/30 s. We therefore hypothesized that if the disruption of the BBB by intracarotid mannitol was primarily due to its osmotic effects, injection of 0.2 to 0.3 mL of mannitol during transient cerebral hypoperfusion will be sufficient to disrupt the BBB, thereby dramatically (by 20-folds) decrease the dose requirements compared with injections during normal blood flow. After preliminary studies, 4 doses of intracarotid mannitol were first tested: (1) 2 mL with cerebral hypoperfusion, (2) 4 mL with cerebral hypoperfusion, (3) 4 mL without cerebral hypoperfusion, and (4) 8 mL without cerebral hypoperfusion. Next, we compared the extent to which methods of drug delivery (infusion vs. bolus injection) affected BBB disruption in 12 rabbits. Finally, we assessed the duration of BBB disruption with intracarotid mannitol in another 12 rabbits. We observed that BBB disruption during injection of 4 mL of mannitol with cerebral hypoperfusion was comparable to 8 mL mannitol without cerebral hypoperfusion. Bolus injections of 4 mL mannitol were more effective than steady-state infusions. The BBB disruption with intracarotid mannitol lasted for 60 minutes postinjection. We conclude that cerebral hypoperfusion decreases the dose of intracarotid mannitol by a modest 2-fold. Our results suggest that mechanical factors may play a significant role in the osmotic disruption of the BBB by intracarotid mannitol.