The role of lactate composition of cerebrospinal fluid (CSF) with vasospasm severity and rabbit neurologic status in subarachnoid hemorrhage was determined. The neurologic status of 20 New Zealand rabbits were graded initially and then, anesthetized and basal angiograms were performed. Then 1.0 mL of CSF was withdrawn through cisterna magna and then 1 mL autologous arterial blood was injected in all rabbits over 1 minute. After 5 days, neurologic severity score (NSS) and vertebrobasilar angiograms of all rabbits were repeated. Rabbits without radiologic vasospasm or spasm under 50% (n=7) were termed as group 1. Rabbits whose cerebral vasospasm were 50% or over 50% (n=7) and NSS is lesser than 3 were termed as groups 2, and rabbits whose cerebral vasospasm were 50% or above 50% (n=7) and NSS is greater than 3 were termed groups 3. On day 7, the CSF lactate values of each group were significantly different (P<0.05) with each other. But when compared with only CSF baseline lactate values groups 2 and 3 were significantly different (P<0.05). However, the NSSs were similar in groups 1 and 2, but group 3 significantly differed from groups 1 and 2 (P<0.05). All groups significantly differed from baseline NSSs (P<0.05). The data showed clearly that the degree of vasospasm correlates not only with neurologic status but also with CSF lactate levels. We suggest that CSF lactate level may be useful as a surrogate marker of cerebral vasospasm degree after subarachnoid hemorrhage in clinics where invasive cerebral angiography could not be assessed for whatever reasons. © 2007 Lippincott Williams & Wilkins, Inc.