Transcranial Doppler validation pilot study.

: The use of transcranial Doppler sonography (TCD) for the evaluation of patients with ischemic cerebrovascular disease remains controversial. This study was organized to gather preliminary data regarding the sensitivity and specificity of TCD when compared to cerebral angiography in detecting stenosing lesions and collateral flow patterns of the anterior cerebral circulation. Forty-two patients from six medical centers were prospectively enrolled. Each received cerebral angiography and TCD testing within 24 hours of each other. Based on TCD criteria established a priori, the results were first analyzed by a blinded investigator and then by computer. Computerized analyses were then repeated with modified criteria. Collateral flow through the anterior communicating and ophthalmic arteries was detected with sensitivities of 62% and 100%, and specificities of 98% and 92%, respectively. Internal carotid artery proximal and distal severe ( greater than 70%) stenoses were detected with sensitivities of 79% and 100% and specificities of 88% and 97%. Middle and anterior cerebral artery stenoses and middle cerebral artery occlusion were detected with specificities exceeding 98%; however, the data were insufficient to determine sensitivity. Computerized analyses did not permit improvement of sensitivity and specificity of the baseline criteria. The selected TCD criteria are highly specific in detecting intracranial stenoses and collateral flow patterns of the anterior circulation. The criteria have limited but acceptable sensitivity and specificity in detecting internal carotid artery origin severe stenoses, and are highly sensitive in detecting ophthalmic artery retrograde flow. A study with a larger sample is necessary to provide definitive guidelines for diagnosis.