**4. Conclusions**

Urethral sphincter BoNT-A injection is effective in treating refractory functional nonneurogenic voiding dysfunction in both genders. The overall successful rate was 64%. Baseline VE and history of recurrent UTI positively correlate with a successful outcome. DU is a predictive factor for a poor treatment outcome.

## **5. Materials and Methods**

The study was initiated following approval by the Institutional Review Board of the author' hospital (IRB 105-151-B). From January 2010 to November 2019, patients who received urethral sphincter BoNT-A injection due to refractory functional, non-neurogenic voiding dysfunction were retrospectively reviewed. All patients available for baseline and follow-up VUDS data were included. Patients with anatomical urethral conditions including uncorrected BPO and high-grade pelvic organ prolapse, history of lower urinary tract reconstruction, urethral stenosis and urethral tumor were excluded. Patients with uncorrected bladder neck dysfunction, neurogenic abnormality related detrusor sphincter dyssynergia, cauda equina syndrome or peripheral neuropathy were also excluded [21]. The voiding dysfunction among patients with cerebral vascular accident, Parkinson's disease or dementia with subtle neurological clinical manifestation were not considered neurogenic since the lower urinary tract symptoms manifest in these diseases were predominantly detrusor overactivity with or without incontinence [22]. The bladder neck dysfunction was corrected first if the VUDS revealed a narrow bladder neck during the voiding phase in patients with voiding dysfunction. Because the patients still have difficulty in urination after TUI-BN or TUR-P, they were recommended to receive urethral sphincter BoNT-A injection for the urethral sphincter dysfunction.

VUDS performed in accordance with the International Continence Society (ICS) recommendation [23] were utilized for baseline urinary function assessment of every patient with refractory voiding dysfunction before the urethral sphincter BoNT-A injection. The cause of voiding dysfunction was determined by VUDS and electromyography (EMG) as DU or urethral sphincter dysfunction. DU was defined as having a bladder contractility index 100 in men, and maximal detrusor pressure (Pdet) < 10 cm H2O with maximum flow rate (Qmax) < 10 mL/s and post-void residual (PVR) > 150 mL in women. The external urethral sphincter dysfunction was subclassified into dysfunctional voiding (DV) or poor relaxation of the external sphincter (PRES) according to the features of VUDS and EMG.

DV was diagnosed as the stasis of contrast at the level of urethral sphincter presenting with the typical feature of a "spinning top" urethra during the voiding phase of VUDS with increased external urethral sphincter EMG activity at the same time [24]. PRES was defined as the narrowing of the distal urethra without the presentation of a "spinning top" urethra during the voiding phase of VUDS without the concomitant relaxation of the external urethral sphincter EMG activity [25].

Other parameters of VUDS included first sensation of filling (FSF), full sensation (FS), urge sensation (US), compliance in the storage phase and Pdet, Qmax, BOO index (BOOI), PVR, cystometric bladder capacity (CBC) and VE in the voiding phase. The bladder contractility index was calculated as Pdet + 5× Qmax and BOOI was calculated as Pdet—2 × Qmax [26]. CBC was calculated by voided volume plus PVR in the VUDS. VE was defined as the voided volume divided by the CBC in the VUDS. Major comorbidities including diabetes mellitus, hypertension, chronic kidney disease, chronic obstructive lung disease, coronary artery disease, and neurogenic disease beyond the sacral spinal cord–brainstem pontine micturition center pathways, as well as history of recurrent urinary tract infection, transurethral resection of the prostate (TUR-P) or transurethral incision or resection of the bladder neck (TUI-BN) were collected from medical records.

All patients received 100 units onabotulinumtoxinA (BOTOX, Allergan, Irvine, CA, USA) external urethral sphincter injections in the operation room under light intravenous general anesthesia [27]. The location of the external urethral sphincter was identified by direct visualization under cystoscopy in both men and women. In male patients, urethral sphincter injections were performed transurethrally using a 23-gauge needle (22 Fr, Richard Wolf, Knittlingen, Germany) with 4–8 injections circumferentially distributed in the external urethral sphincter at a depth of 0.5 cm along the longitudinal direction of the urethral lumen. Female patients, on the other hand, received urethral sphincter injections periurethrally using 27-gauge 1 mL syringe needles with 4–8 injections circumferentially into the external urethral sphincter at a depth of 1.5 cm along the longitudinal direction of the urethral lumen. A detailed description of the urethral sphincter injection technique was reported in our previous review [28]. Treatment outcomes of urethral sphincter BoNT-A injections were assessed at around 3 months after the procedure since the average therapeutic duration was reported at around 6 months [9]. Subjective outcomes were measured by global response assessment (GRA) as excellent (+3), markedly improved (+2), mildly improved (+1) or no change (0), according to the patients' perception of the voiding condition after the BoNT-A injections. Patients with an excellent outcome can ge<sup>t</sup> rid of the catheter, and patients with marked improvements still need CIC occasionally. A successful outcome was defined as GRA equal to or greater than 2. Objective outcomes were also assessed by VUDS follow-up after the injections.

All analyses were performed through SAS Statistics for Windows, Version 9.4, Cary, NC, USA: SAS Inc. Two-sided *p*-values less than 0.05 were considered significant. The continuous variables of baseline demographics were expressed as the mean ± standard deviation whereas the categorical ones were expressed as number (percentage). Differences between gender of the above variables were examined with independent t-test in continuous variables and chi-square test in the categorial ones. We applied Fisher's exact test in circumstances when more than 20% of the expected frequencies were less than five. Changes in post-treatment variables in each gender were examined with the paired samples t-test and McNemar test for continuous and categorical variables, respectively. The distribution of the GRA grades after injections between DU and urethral sphincter dysfunction was examined with the chi-square test. Univariate logistic regression analysis was performed to find out the predictive factors for a successful treatment outcome. Variables demonstrating significant differences in the univariable analysis, including age and gender, were further evaluated in the multivariable model.

**Author Contributions:** Conceptualization, H.-C.K.; Data curation, Y.-C.O., Y.-L.K. and K.-J.T.; Formal analysis, Y.-L.K.; Investigation, Y.-C.O.; Methodology, Y.-L.K.; Project administration, H.-C.K.; Resources, H.-C.K.; Supervision, H.-C.K. and K.-J.T.; Visualization, Y.-C.O. and Y.-L.K.; Writing— original draft, Y.-L.K.; Writing—review and editing, H.-C.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** The work was supported by grants from the National Cheng Kung University Hospital NCKUH-11103038 and Buddhist Tzu Chi Medical Foundation TCMF-MP-110-03-01 and TCMF-SP-108-01.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of the Buddhist Tzu Chi General Hospital (IRB 105-151-B).

**Informed Consent Statement:** Patient consent was waived due to the retrospective study design.

**Data Availability Statement:** Data are available on request to the corresponding author.

**Acknowledgments:** The work was greatly enhanced by the assistance of Chia-Min Liu.

**Conflicts of Interest:** The authors declare no conflict of interest.
