Cavernosal venous occlusive dysfunction (CVOD) accounts for the most common etiology of erectile dysfunction (ED) among non-responders to phosphodiesterase type 5 inhibitors (PDE5is). How to identify CVOD caused by hypersympathetic activity of cavernous smooth muscle (HS-CVOD) and effectively treat HS-CVOD non-responded to PDE5is have not been fully studied. A total of 166 patients with ED were assessed and grouped by nocturnal penile tumescence rigidity test (NPTR) and color duplex Doppler ultrasound (CDDU). The sympathetic activity of cavernous smooth muscle (amplitude and duration) were determined by corpus cavernosum electromyography (CC-EMG). Patients diagnosed with CVOD by CDDU and with psychological ED by NPTR were considered as HS-CVOD. HS-CVOD patients non-responded to PDE5is were treated by PDE5is combined trazodone for 4 weeks. There were 27 cases of HS-CVOD were considered. The CC-EMG amplitude of HS-CVOD group (302.65 ± 195.01 uv) were significantly greater than that of psychological ED combined with non-vascular ED group (183.71 ± 112.42 uv), organic ED combined CVOD group (162.67 ± 91.17 uv), organic ED combined CVOD group (162.67 ± 91.17 uv), and organic ED combined arterial or mixed ED group (145.67 ± 75.29 uv), respectively. 8 in 11 patients with HS-CVOD non-responded to PDE5is were successfully treated by PDE5is combined trazodone. These findings underscore that HS-CVOD can be preliminary considered by NPTR and CDDU, and assessed by CC-EMG with enhanced sympathetic activity of cavernous smooth muscle. The therapy of PDE5is combined trazodone could be an effective treatment strategy for HS-CVOD non-responded to PDE5is.