Transcranial Dopplers Revisited: Development of Novel Markers for Cerebral Vasospasm After Aneurysmal Subarachnoid...

Background Cerebral vasospasm has been monitored by conventional angiography or transcranial Doppler (TCD). While angiography is the most accurate and reliable method for detection, TCDs are a noninvasive alternative to monitor onset and resolution of vasospasm. We aim to determine whether alternative TCD parameters rather than Lindegaard ratio lead to an improved method to diagnose and potentially prevent cerebral vasospasm. Methods A total of 103 consecutive patients with subarachnoid hemorrhage (SAH) were retrospectively reviewed and TCD studies were performed during the first 14 days post-bleed or longer if indicated. Multivariate logistic regression models were developed using significant univariate characteristics. Receiver operating characteristic (ROC) curves evaluated the mean middle cerebral artery (MCA), peak systolic MCA (PSV MCA), and end diastolic MCA (EDV MCA) velocities as well as ratios when compared to the ipsilateral extracranial internal carotid artery (ICA). The area under the curve was calculated to compare accuracy for symptomatic vasospasm. Results Thirteen patients (12.6%) were observed to develop cerebral vasospasm. Aneurysm location (p = 0.51), Hunt and Hess grade (p = 0.44), Fischer grade (p = 0.87), comorbidities, age (p = 0.67), or gender (p = 0.41) did not appear to have any effect in predicting the presence of vasospasm. ROC curves demonstrated that MCA EDV appeared to be slightly better compared to MCA velocity in predicting symptomatic vasospasm. PSV MCA/extracranial ICA and the EDV MCA/extracranial ICA ratios appeared to be an improvement to the Lindegaard ratio in the prediction of symptomatic vasospasm. Conclusion The utility of peak systolic and end diastolic velocities, instead of the classically referenced mean velocities and Lindegaard ratio, may improve diagnostic sensitivity of cerebral vasospasm after subarachnoid hemorrhage.