A STATISTICAL-ANALYSIS OF FACTORS RELATED TO SYMPTOMATIC CEREBRAL VASOSPASM

A retrospective review of patients presenting to our institution with aneurysmal subarachnoid hemorrhage between 1980-1990 was accomplished. Eleven variables were examined as to their relationship to clinical v asospasm: age, sex, clinical grade, amount of subarachnoid blood on CT , aneurysm location, incidence of vasospasm, incidence of complication s, use of calcium channel blockers, time to surgery, length of stay, a nd outcome. Data were analyzed with univariate and multivariate logist ical regression methodology. By univariate analysis, age under 20, amo unt of subarachnoid hemorrhage, and clinical grade were associated wit h a higher risk of vasospasm. Using multivariate logistic regression, these factors, along with age under 35, were correlated as being predi ctive of clinical vasospasm. When all patients are grouped into either good or bad outcome, and a similar analysis is performed, only in the poor outcome group is the amount of subarachnoid hemorrhage and clini cal grade correlated with vasospasm. This suggests that there is a gro up of patients with a predisposition to vasospasm that is independent of subarachnoid hemorrhage and clinical grade, and that these patients may have a more favorable outcome.