Erectile dysfunction has taken on increasing importance in urologic practice. Still open is the question of which tests are mandatory for adequate clinical assessment of erectile dysfunction. One accepted standard modality is the intracavernous pharmacotest with vasoactive agents. In contrast, color duplex sonography is not considered mandatory although it reveals detailed information about penile vessels and functional implications. The question remains whether the information gained by color duplex sonography is relevant for therapy, making it an indispensable standard procedure. Patients with erectile dysfunction were evaluated, without preselection, by extensive history, clinical evaluation, laboratory tests, tumescence and rigidity measurements, intracavernous administration of vasoactive drugs and color duplex sonography. Seventy-nine patients were available for final analysis. Color duplex sonography revealed 39 normal tests, 16 arterial impairments. 19 venous leakages and 5 arteriovenous fistulae. Intracavernous vasoactive agents (pharmacotest) revealed 44 reactions sufficient for intercourse and 35 insufficient responses. In 89% of patients, the diagnoses on color duplex sonography and intracavernous pharmaco-testing were similar. The accuracy of the two methods in diagnosing erectile dysfunction was not statistically different (McNemar's test). In contrast to intracavernous pharmaco-testing, color duplex sonography permitted further etiologic subdivision into arterial disease, venous leakage, arteriovenous fistula and normal result. This was made possible by measuring significantly (P < 0.01) different arterial peak flow velocities, end-diastolic velocities and calculated resistance index. These data did not imply direct clinical consequences. Color duplex sonography and intracavernous pharmacotesting reveal comparable results concerning the diagnosis of an erectile dysfunction. In contrast to pharmacotesting, color duplex sonography reveals details of the nature of the erectile dysfunction. Because this information has no profound implications for the choice of therapeutic procedure, color duplex sonography can not be recommended as a standard procedure in the evaluation of erectile dysfunction.