During the diagnostic evaluation of patients with vasculogenic impotence duplex scanning has been proposed as a reliable noninvasive method to evaluate the cavernous arteries. However, the sensitivity and specificity of this test have never been elucidated. To provide insight into the clinical value of this test 25 men who presented with impotence and failed to respond to 60 mg. intracavernous papaverine were evaluated with duplex scanning and the results of this modality were compared to penile angiography. The cavernous arteries were considered normal by duplex scanning if the artery demonstrated either a 60% or greater increase in diameter and/or a peak flow velocity of greater than 25 cm. per second 5 minutes after papaverine injection. Penile angiography was considered normal if both cavernous arteries were visualized and appeared to be normal radiographically (after intracorporeal papaverine). In the 25 patients studied there was no significant difference in peak flow velocity between patients with normal or abnormal cavernous arteries by angiography. In addition, arterial dilatation (more than 60%) after papaverine injection did not correlate with the results of the angiogram. These data demonstrate that duplex scanning of the cavernous arteries does not correlate well with what is observed by penile angiography in a selected population of men with vasculogenic erectile dysfunction. Since penile angiography itself has limitations in the evaluation of the penile arteries, the reliability of duplex scanning to identify cavernous artery dysfunction needs additional confirmation.