Symptomatic vasospasm diagnosis after subarachnoid hemorrhage: Evaluation of transcranial Doppler ultrasound and cerebral angiography as related to compromised vascular distribution

Objective: To evaluate the reliability of transcranial Doppler ultrasound in detecting symptomatic vasospasm in patients after aneurysmal subarachnoid hemorrhage and monitoring response after hypertensive and endovascular treatments.Design: Retrospective chart review. Setting: Neurosciences critical care unit in a tertiary-care university hospital. Patients: All patients admitted to a neurosciences critical care unit with the diagnosis of subarachnoid hemorrhage between January 1990 and June 1997.Interventions: None Measurements and Main Results: We reviewed transcranial Doppler ultrasound data of 199 patients; 55 had symptomatic vasospasm. Clinical symptoms and corresponding vascular distributions were identified, as was angiographic vasospasm (n = 35). The sensitivity and specificity of transcranial Doppler ultrasound for anterior circulation vessels were calculated by using a mean cerebral blood flow velocity criterion of >120 cm/sec. Clinical diagnosis of symptomatic vasospasm was used as the standard to determine sensitivity and specificity of transcranial Doppler ultrasound and cerebral angiography. The sensitivity of transcranial Doppler ultrasound for anterior circulation in patients with symptomatic vasospasm was 73% with a specificity of 80%. The sensitivity of cerebral angiography was 80%. For individual vessels, the sensitivity and specificity of transcranial Doppler ultrasound were middle cerebral artery, 64% and 78%; anterior cerebral artery, 45% and 84%; and internal carotid artery, 80% and 77%, respectively. The mean times for symptomatic and transcranial Doppler ultrasound signs of vasospasm presentation were 6.4 ± 2 and 6.1 ± 3 days, respectively. In patients without symptomatic vasospasm, the mean time for mean cerebral blood flow velocities >120 cm/sec was 7.0 ± 3 days (p < .05). Symptomatic vasospasm also was associated with thickness of clot on head computed tomography scan and rapidly increasing mean cerebral blood flow velocities. Transcranial Doppler ultrasound signs of vasospasm improved after endovascular treatment in 30 patients. Conclusions: The reliability of transcranial Doppler ultrasound was better at detecting high mean cerebral blood flow velocities in patients with symptomatic vasospasm related to middle cerebral and internal carotid artery distributions than for anterior cerebral artery distribution. Transcranial Doppler ultrasound was as sensitive as cerebral angiography at detecting symptomatic vasospasm. High mean cerebral blood flow velocities can be apparent before the presence of symptomatic vasospasm. Daily transcranial Doppler ultrasound monitoring could provide early identification of patients with aneurysmal subarachnoid hemorrhage who are at high risk for symptomatic vasospasm and may be helpful at following success of endovascular treatment.