Doppler sonographic parameters for detection of carotid stenosis: Is therean optimum method for their selection?

OBJECTIVE. A wide range of Doppler threshold values for carotid stenosis is found in the literature. We undertook this study to compare methods of der ivation and to determine if an optimum strategy of threshold selection exis ts for a high-risk population. MATERIALS AND METHODS. From the sonograms of all patent internal carotid ar teries, peak systolic velocity in the internal carotid artery (ICA(PSV)) an d the ratio of peak systolic velocity in the internal carotid artery to tha t of the common carotid artery (ICA(PSV)/CCA(PSV)) were compared with the p ercentage of angiographically determined stenosis, Receiver operating chara cteristic curves were generated for levels of stenosis greater than or equa l to 60% and greater than or equal to 70%. Doppler thresholds were chosen o n the basis of maximum accuracy and on the basis of greater than or equal t o 90% sensitivity and specificity. Patients were then segregated into sympt omatic and asymptomatic cohorts, and the above process was repeated. An eff ectiveness analysis was also conducted using various Doppler thresholds. Th resholds derived using these three methods were compared and optimal values chosen. RESULTS. Of 333 carotid arteries that fit inclusion criteria, 132 were foun d in asymptomatic patients and 201 in symptomatic patients. Maximum accurac y, greater than or equal to 90% sensitivity and specificity, and effectiven ess analysis each produced different ranges of thresholds. We chose final t hresholds that maintained patient outcome profiles. For asymptomatic patien ts at the greater than or equal to 60% stenosis level, thresholds were ICA( PSV) = 200 cm/sec and ICA(PSV)/CCA(PSV) = 3.0. For symptomatic patients wit h stenosis greater than or equal to 70%, thresholds were ICA(PSV) = 175 cm/ sec and ICA(PSV)/CCA(PSV) = 2.5. CONCLUSION. Considerable latitude exists in the choice of carotid Doppler t hresholds. We propose a rational strategy for threshold selection based on a combination of three commonly used methods. Our observations indicate tha t it appears advisable to consider symptomatic and asymptomatic patients se parately and to apply appropriately derived thresholds.