**4. Materials and Methods**

This study retrospectively analyzed consecutive patients with voiding dysfunction who were refractory to medical treatment and received 100 U of BoNT-A (onabotulinumtoxinA, Allergan, Irvine, CA, USA) via urethral sphincter injection from 2011 to 2019. All patients underwent VUDS and cystoscopy to identify the underlying pathophysiology of lower urinary tract dysfunction before the BoNT-A injections. Patients with anatomical BOO, such as urethral stricture, bladder neck obstruction, or benign prostatic obstruction were excluded from the study. Only patients with DU who required spontaneous voiding by abdominal straining, had voiding dysfunction due to non-relaxing urethral sphincter, had spinal cord lesion(s) and DSD, and had DV were included in the final analysis. This study was approved by the ethics committee of the institution (IRB 105-151-B). Date of approval: 29 December 2016 to 14 December 2017. Informed consent was waived by the committee due to the retrospective nature of the study. All patients were informed of the potential adverse events after BoNT-A injection before treatment.

The treatment was performed in the operating room under light intravenous general anesthesia. A total of 100 U BoNT-A was given via transurethral sphincter injections [8]. One vial of 100 U onabotulinumtoxinA was reconstituted with normal saline to 4 mL, making the concentration equivalent to 25 U per mL. One mL of BoNT-A solution was injected into the urethral sphincter at the 3, 6, 9, and 12-o'clock positions transurethrally in men, and transcutaneous injection to the urethral sphincter along the urethral lumen at 1, 4, 7, and 10 o'clock positions of the sides of the urethral meatus in women.

A Foley catheter was indwelled overnight after BoNT-A injections and the voiding condition was requested to be reported at the out-patient clinic. The effect of BoNT-A on the urethral sphincter function appeared about 2–3 days after the treatment. The maximum therapeutic effect would reach about 2 weeks after BoNT-A injection [8]. Patients with DU and large PVR were instructed to void by Crede maneuver or abdominal straining, and CIC was recommended instead of an indwelling Foley catheter. The Qmax and PVR were measured and CIC was continued until PVR was reduced to less than 50% of the voided volume. Antibiotics were routinely given for 3 days to prevent urinary tract Infections and all medications for reduction of urethral resistance were discontinued after BoNT-A injections.

VUDS was performed in all patients at baseline. The VUDS parameters included the opening or closed bladder neck during voiding cystourethrography, the FSF, CBC, Qmax, Pdet, PVR, intra-abdominal pressure to void (in patients with DU), and VE were recorded. The terminology used in this study was based on the recommendations of the International Continence Society [33]. Patients with voiding dysfunction caused by SCI, CVA, MS, or peripheral neuropathy were categorized as having NVD, otherwise, patients were considered as NNVD. For analysis of the treatment outcome, the bladder dysfunctions were categorized into hypersensitive bladder, DO, DHIC, and DU groups. The urethral sphincter dysfunctions were caegorized as DV, DSD, and PRES, according to the electromyographic characteristics and images during the voiding phase.

The treatment outcome was assessed by self reported satisfaction and the change of the VE at 1 month after the BoNT-A injection. Patients were requested to report their global response assessment (GRA) after BoNT-A injection as: Excellent (+3), markedly improved (+2), mildly improved (+1), no change (0), or worsened (−1). Treatment success was defined as a report with GRA ≥ 1 and the VE ≥ 50% after BoNT-A injection. If patients had VE of less than 50% and GRA equaled to 0 or −1, they were considered to have had failed treatment. Adverse events after BoNT-A injections were also recorded and appropriate treatments were given if needed.

Continuous variables were expressed as means ± standard deviations (SD). Categorical data were presented as numbers and percentages (%). We used chi-square test for categorical variables, and the Wilcoxon rank sum test for continuous variables to determine the *p*-values between successful and failed subgroups for statistical comparisons. All statistical assessments were two-sided and considered significant at a *p* < 0.05. All calculations were performed using SPSS for Windows, version 16.0 (SPSS, Chicago, IL, USA).
