Relative changes in transcranial Doppler velocities are inferior to absolute thresholds in prediction of symptomatic vasospasm after subarachnoid hemorrhage.

The absolute transcranial Doppler (TCD) velocity threshold has been validated as a screening tool for vasospasm after subarchnoid hemorrhage (SAH). We assessed whether relative changes in velocity were superior to absolute TCD thresholds in the detection of symptomatic vasospasm. We reviewed consecutive patients with aneurysmal SAH who underwent serial TCD monitoring and survived at least 7 days. We recorded initial flow velocity (IFV) and maximal flow velocity (MFV) of the middle cerebral artery (MCA) serially up to 14 days from admission. We calculated relative flow velocity changes (MFV/IFV) and maximum change in mean flow velocity (FVmean) over any consecutive 2 days in addition to standard absolute measures of Lindegaard ratio (LR) and FVmean. We calculated receiver operating characteristic curve and area under curve (AUC) values, sensitivity, specificity, and positive predictive and negative predictive values for these parameters, optimal cutpoints, and various combinations. Forty-eight of 211 patients (23%) developed symptomatic MCA vasospasm. AUC values for various TCD parameters were 0.80 for MCA MFV >175 cm/s, 0.71 for LR >6, 0.64 for MFV/IFV >2, and 0.64 for >70% change in MFV over 2 days. The best characteristics were observed for the combination of MFV >175 cm/s and/or maximal LR >6 (AUC 0.81). Our data suggest that absolute thresholds of TCD FVmean provide the most accurate prediction of symptomatic MCA vasospasm after SAH. Other thresholds, including relative change from baseline and day-to-day changes, are inferior to established absolute thresholds.