arotid endarterectomy (CEA) reduces the risk for stroke in selected patients with symptomatic and asymptomatic carotid atherosclerosis. Because of its observed benefit and the prevalence of carotid bifurcation atherosclerosis, CEA is the most commonly performed vascular reconstructive procedure. In general, the benefit of CEA is related to the degree of stenosis, although in two national trials of patients with asymptomatic disease, observations of the stenosis-specific benefit did not persist. Carotid arteriography is commonly considered the most reliable method for evaluating carotid atherosclerosis, and the North American Symptomatic Carotid Endarterectomy Trial (NASCET) investigators standardized the method of quantifying the degree of carotid stenosis. Carotid duplex ultrasound scanning compares favorably with arteriography in quantifying carotid atherosclerosis. In an effort to reduce the risk for ischemic complications from carotid arteriography, many physicians are using carotid duplex scanning as the definitive diagnostic procedure before CEA. Diagnostic criteria for carotid duplex scanning have been established and accepted as reliable from laboratories accredited by the Intersocietal Commission for the Accreditation of Vascular Laboratories with ongoing quality-assurance programs. However, despite efforts to maintain accurate diagnostic criteria that reflect arteriographically demonstrable disease, personal observations were made at a worrisome frequency that carotid duplex scanning resulted in overestimation of disease severity in women, yet false-positive studies were infrequently observed in men. This raised the concern that women may have higher velocities in their carotid arteries than men do for similar amounts of disease, and led to this retrospective analysis of carotid duplex ultrasound scanning versus arteriography stratified by gender. The purpose of this study was to examine whether there were velocity differences based on gender in patients with carotid artery disease and whether different velocity criteria should be used in women than men, especially at clinically relevant thresholds of disease.