Costs of vasospasm in patients with aneurysmal subarachnoid hemorrhage.

Abstract

To assess the impact of vasospasm on costs, length of stay, and mortality among inpatients with aneurysmal subarachnoid hemorrhage. We combined hospital accounting and physician billing data for a consecutive cohort of 198 patients who underwent surgical clipping or endovascular coiling for subarachnoid hemorrhage repair. We considered patients with transcranial Doppler (TCD) velocity of 120 cm/s or greater in the middle cerebral artery to have TCD-defined vasospasm and patients with delayed ischemic neurological deficit to have symptomatic vasospasm. We compared outcomes of patients with TCD-defined vasospasm (n = 116) and those without (n = 73) and patients with symptomatic vasospasm (n = 62) and those without (n = 127), adjusting for demographic and clinical characteristics. In adjusted analyses, the incremental cost attributable to TCD-defined vasospasm was 1.20 times higher (95% confidence interval, 1.06-1.36; P = .004) than for patients without TCD-defined vasospasm. Length of stay was an estimated 1.22 times longer for patients with TCD-defined vasospasm (95% CI, 1.07-1.39; P < .01). For symptomatic vasospasm, adjusted costs were 1.27 times higher (95% CI, 1.12-1.43; P < .001) and length of stay was an estimated 1.24 times longer (95% CI, 1.09-1.40; P < .01) for patients with vasospasm than for those without. There was no significant relationship between either type of vasospasm and in-hospital mortality. Patients with subarachnoid hemorrhage and TCD-defined or symptomatic vasospasm incur higher inpatient costs and longer hospital stays than those without vasospasm.