PURPOSE: To evaluate the accuracy of transcranial Doppler (TCD) sonography using different criteria for predicting cerebral infarction due to symptomatic vasospasm. METHODS: We retrospectively evaluated the clinical and radiologic data of consecutive patients admitted with acute aneurysmal subarachnoid hemorrhage (SAH) in the anterior cerebral circulation between January 2001 and June 2002. TCD sonographic examinations were performed on alternate days up to 20 days after admission. Cerebral infarction was defined on CT as a new hypodensity in the vascular distribution with corresponding clinical symptoms. Vasospasm was diagnosed as mild or severe when TCD sonography revealed a mean blood flow velocity (MBFV) greater than 120 and 180 cm/s in the middle or anterior cerebral artery and in the intracranial part of the internal carotid artery, respectively. RESULTS: A total of 93 patients with aneurysmal SAH in the anterior cerebral circulation were included. Vasospasm was demonstrated by TCD sonography in 60 patients (64.5%) and was shown via multivariable logistic regression analysis to be predictive of cerebral infarction (OR 3.11, 95% CI 1.46-6.59), with an 82.6% and 69.6% sensitivity, a 41.4% and 77.1% specificity, a 31.7% and 50.0% positive predictive value, and an 87.9% and 88.5% negative predictive value when the MBFV was greater than 120 and 180 cm/s, respectively. CONCLUSIONS: Vasospasm on TCD was found to be predictive of symptomatic cerebral infarction on CT, but its positive predictive value remained low despite the adoption of restrictive TCD criteria for vasospasm.
PURPOSE: To evaluate the accuracy of transcranial Doppler (TCD) sonography using different criteria for predicting cerebral infarction due to symptomatic vasospasm. METHODS: We retrospectively evaluated the clinical and radiologic data of consecutive patients admitted with acute aneurysmal subarachnoid hemorrhage (SAH) in the anterior cerebral circulation between January 2001 and June 2002. TCD sonographic examinations were performed on alternate days up to 20 days after admission. Cerebral infarction was defined on CT as a new hypodensity in the vascular distribution with corresponding clinical symptoms. Vasospasm was diagnosed as mild or severe when TCD sonography revealed a mean blood flow velocity (MBFV) greater than 120 and 180 cm/s in the middle or anterior cerebral artery and in the intracranial part of the internal carotid artery, respectively. RESULTS: A total of 93 patients with aneurysmalSAH in the anterior cerebral circulation were included. Vasospasm was demonstrated by TCD sonography in 60 patients (64.5%) and was shown via multivariable logistic regression analysis to be predictive of cerebral infarction (OR 3.11, 95% CI 1.46-6.59), with an 82.6% and 69.6% sensitivity, a 41.4% and 77.1% specificity, a 31.7% and 50.0% positive predictive value, and an 87.9% and 88.5% negative predictive value when the MBFV was greater than 120 and 180 cm/s, respectively. CONCLUSIONS:Vasospasm on TCD was found to be predictive of symptomatic cerebral infarction on CT, but its positive predictive value remained low despite the adoption of restrictive TCD criteria for vasospasm.
Authors: Anne-Katrin Hickmann; Sönke Langner; Michael Kirsch; Jörg Baldauf; Cornelia Müller; Alexander Khaw; Henry W S Schroeder Journal: Neurosurg Rev Date: 2012-10-27 Impact factor: 3.042