Treating voiding dysfunction without anatomical obstructions is challenging. Urethral onabotulinum toxin A (BoNT-A) is used in treating voiding dysfunction; however, the success rate varies widely, and patients may not be satisfied with the treatment outcome. This study compared the efficacy of the urethral BoNT-A injection between patients with different non-spinal cord injury (SCI) voiding dysfunctions. This study retrospectively analyzed patients with refractory voiding dysfunction, including detrusor underactivity (DU), dysfunctional voiding (DV), and poor relaxation of the external sphincter (PRES) who received the urethral sphincter 100 U BoNT-A injection. The treatment outcomes were assessed via a global response assessment (GRA) one month after treatment. Baseline and follow-up videourodynamic study (VUDS) parameters were also compared. Totally, 161 patients (60 with DU, 77 with DV, and 24 with PRES) with a mean age of 58.8 ± 20.2 were enrolled, of which 62.1% had a good response (GRA ≥ 2) after urethral BoNT-A injection. DV patients had a higher success rate (76.6%) than DU (50%) and PRES (45.8%) patients (p = 0.002). A diagnosis of DV, higher voided volume and recurrent urinary tract infection were predictors of a good treatment response, while the cervical cancer status post-radical surgery predicted a poor response. Receiver operating characteristic (ROC) curve analyses identified PVR > 250 mL as a negative predictor (p = 0.008) in DU patients. The urethral BoNT-A injection provides a satisfactory success rate for non-SCI voiding dysfunction. Patients with DV benefit most from both subjective and objective parameters. Approximately 50% of patients with DU and PRES also had a fair response. PVR > 250 mL was a negative predictor in DU patients.