ObjectivesThe aim of this study was to analyze the diagnostic accuracy of different criteria of pharmaco‐penile duplex sonography in venous erectile dysfunction (ED).MethodsThe following parameters were measured after an intracavernous injection test in patients with ED from May 2016 to February 2017 at our hospital: diameter, peak systolic velocity, end‐diastolic velocity, and resistance index of the cavernous artery; diameter and peak velocity (if leak occurred) of the deep dorsal vein. Three ultrasonographic diagnostic criteria of venous ED were applied. Criterion A: continuous blood flow signals in the deep dorsal vein, peak velocity greater than 3 cm/s, peak systolic velocity greater than 30 cm/s, end‐diastolic velocity greater than 5 cm/s; Criterion B: resistance index less than 0.89 and other parameters corresponding with Criterion A; Criterion C: resistance index less than 0.80 and other parameters corresponding with Criterion A. The diagnostic results of each criterion were compared with the cavernosographic results.ResultsThirty‐six patients were diagnosed as venous ED by cavernosography in 54 ED cases. The diagnostic specificity, sensitivity, and accuracy of Criterion A were 70.6%, 91.7%, and 84.9%, respectively. Those of Criterion B were 82.4%, 69.4%, and 73.6%, while the results for Criterion C were 94.1%, 33.3%, and 52.8%, respectively. Criterion A had the highest diagnostic accuracy, the largest area under the receiver operating characteristic curve (area = 0.811), and the highest consistency (kappa = 0.642) with the cavernosographic results in the 3 criteria. The difference was statistically significant (P < .05).ConclusionsAmong the 3 commonly used ultrasonographic criteria, Criterion A is most appropriate in the diagnosis of venous ED.