BACKGROUND AND PURPOSE: Microemboli have been implicated in the etiology of neuropsychological deficits after cardiopulmonary bypass. This study examined the incidence of high-intensity transcranial signals (microemboli) and their relation to changes in neuropsychological performance after surgery. METHODS: Transcranial Doppler ultrasonography was used to measure middle cerebral artery blood flow velocity and detect microemboli. The number of high-intensity transcranial signals was determined and related to a neurological examination and absolute changes in neuropsychological performance as well as the number of patients considered to exhibit a neuropsychological deficit. Data were available on 100 consenting patients undergoing routine cardiopulmonary bypass. Fifty of the patients were randomly assigned to a procedure that included a 40-microns arterial line filter, and 50 had the procedure without any arterial line filter. RESULTS: Significantly more patients were found to have neuropsychological deficits in the group without the arterial line filter at both 8 days (P < .05) and 8 weeks (P < .03) after surgery. In addition, more "soft" neurological signs were found in the nonfiltered group 24 hours after surgery (P < .05). More high-intensity transcranial signals were found in the nonfiltered group, and the number of high-intensity transcranial signals was found to be related to the likelihood of a patient having a neuropsychological deficit at 8 weeks. CONCLUSIONS: These data suggest that neuropsychological deficits after routine cardiopulmonary bypass are related to the number of microemboli delivered during surgery. Furthermore, the numbers of microemboli may be reduced by including a 40-microns filter on the arterial line.