**3. Discussion**

Per our findings, the BoNT-A urethral sphincter injection in non-SCI patients with voiding dysfunction produced a good response in 62.1% of patients after the urethral BoNT-A injection. In different types of voiding dysfunction, patients with DV had better treatment outcomes than those with DU and PRES. In multivariate analyses, DV, more voided volume, and recurrent UTI, were predictors of a good response to treatment, while the cervical cancer status post-radical surgery predicted a poor response. Although VUDS parameters did not differ significantly before and after treatment, they allow physicians to clearly observe the bladder outlet appearance during the voiding phase, which may provide insights into the pathophysiology of the voiding dysfunction [8]. We found that DV is a good predictor of the treatment response; so VUDS is considered to play an important role in making a precise diagnosis before treatment.

BoNT-A is believed to block the presynaptic release of acetylcholine in the neuromuscular junction in striated muscles, which achieves medical sphincterotomy effects. This could reduce the urethral sphincter resistance and improve voiding dysfunction [9]. The application of BoNT-A in urology was first used with urethral sphincter injections for the treatment of detrusor sphincter dyssynergia in patients with SCI and multiple sclerosis [10]. Double-blind placebo-controlled study then confirmed the validity and durability of the therapeutic efficacy of the BoNT-A urethral sphincter injection for patients with SCI and DSD [10]. Therefore, this treatment has been further used in treating non-SCI voiding dysfunction patients nowadays due to urethral sphincter hyperactivity, PRES, and DV or DU [5].

Voiding dysfunction is a frequently encountered clinical problem. In addition to anatomical obstruction-related voiding dysfunctions like benign prostatic hyperplasia and urethral stricture, functional problems like DU, DV, or PRES are more challenging for urologists. The current urodynamic study reported DU would possess in 12.4% of men [11] and 23.1% of women [12] with voiding dysfunction. Urethral sphincter hyperactivity was found in 17.0% of women, and PRES was noted in 39.5% of men and 17.6% of women with voiding dysfunction [11,13].

Treatment of DV is usually challenged because the actual pathophysiology has not been well explained currently and is thought to be a dysregulated urethral function with a spastic or non-relaxing external urethral sphincter during voiding [14]. DV results in difficult voiding and leads to a weak stream of urination and a large PVR. Therefore, attempts to reduce the hypertonicity or hyperactivity of the urethral sphincter via oral medication and resume smooth voiding are often futile. It is also postulated that voiding dysfunction due to psychological origins such as anxiety or depression might cause low detrusor contractility and urethral sphincter non-relaxation by inhibiting detrusor contraction [15]. Liao et al. previously reported an overall success rate of 86.7% for DV patients with sphincteric injections (50–100 units of Botox) [14]. Lee et al. also reported a 62.2% success rate in non-neurogenic DV [16]. In our study, approximately 76.6% of DV patients treated using BoNT-A urethral sphincter injections had a GRA of ≥2. VUDS also showed significantly decreased Pdet.Qmax and BOOI in female patients. On the other hand, our result showed no significant difference in male DV after treatment in VUDS data. It may be due to the case number being small (*n* = 10), and we also found that male DV baseline detrusor contractility is relatively not strong enough.

The etiology of DU is known to be neurogenic, myogenic, obstructive, or idiopathic. Sustained abdominal pressure is necessary to facilitate emptying the bladder [17]. Urethral BoNT-A sphincter injections help to decrease bladder outlet resistance and achieve successful outcomes. We need to be sure that the bladder neck should open during voiding. Otherwise, BoNT-A injections to the urethral sphincter may not be successful [14]. Therefore, if bladder neck dysfunction was confirmed by VUDS, patients with DU and voiding dysfunction should receive transurethral incision of the bladder neck (TUIBN) rather than urethral BoNT-A injection. In this study, we carefully excluded patients with bladder neck dysfunction and those previously treated for TUIBN (73.9%). In this study, approximately 50% of DU patients had GRA ≥ 2. This means the recovery of detrusor function combined with a hyperactive sphincter also suggested the potential neuromodulatory effect of the sphincteric BoNT-A injection. Sufficient abdominal pressure is necessary for triggering spontaneous voiding after urethral BoNT-A injections in patients with DU. In our previous study, female DU patients exhibited VE improvement after active treatment, and intact bladder sensations and smaller PVR had better treatment outcomes [18]. In this study, we also found that a large PVR is a negative predictor, and the receiver operating characteristic curve showed that PVR > 250 mL at baseline indicates a poor outcome. We supposed that a large PVR indicates a lower abdominal pressure or decreased bladder sensation. Another important factor for efficient urination is acceptable bladder sensation. The sensory afferents from the bladder urothelium and detrusor play important roles in the voiding reflex circuit. Decreased bladder sensation will render the initiation of voiding difficult [19]. Overall, DU patients treated using BoNT-A urethral sphincter injections in our study showed a 50% success rate. We also found that female patients with cervical cancer status post-radical hysterectomy had poor outcomes. We believe radical surgery causes nerve injury and, thus, irreversible DU; so, we could predict that patients with poor sensations and large PVRs were usually not satisfied with the treatment.

PRES, as a diagnosis, was determined based on the voiding phase in the VUDS, which shows non-relaxed surface EMG activity combined with a narrow membranous urethra [2]. The etiology of PRES was considered multifactorial, such as potential neuropathy, learned habituation, pelvic floor hypertonicity, and bladder hypersensitivity [20]. PRES is characterized by relatively small but stable bladders and low-pressure/low-flow during the voiding phase [21], which is different from the typical high-pressure/low-flow presentation in DV. Urethral BoNT-A injections may provide benefits by inhibiting acetylcholine release in the neuromuscular junction to reduce urethral resistance. Because the typical PRES is low-pressure during the voiding phase, we supposed that there was also inadequate detrusor contractility. Therefore, the success rate of urethral BoNT-A injections is not as high as that of DV. On the other hand, a previous study showed that detrusor contractility might be restored after BoNT-A injections in DU patients with PRES [22]. This result supports

the hypothesis that the low-pressure/low-flow dysfunction present in PRES might be the result of the detrusor suppression induced by non-relaxed urethral sphincter activity [7].

The primary limitation of this study is its retrospective design, different group sizes, and single-center scope of evaluation. Second, the 1–6-month follow-up VUDS was not consistent, which may have influenced the results of objective parameters. However, we believe the efficacy of BoNT-A durability continued for at least 6 months [23]. Moreover, patients without follow-up VUDS were not enrolled in this study, which might have caused selection bias. In this study, no obvious side was reported. However, we supposed mild side effects might exist in some patients, such as urinary incontinence. Finally, the patient groups were heterogeneous, with varying causes of voiding dysfunction. The identification of the underlying causes of failure may improve the success rate of the urethral sphincter BoNT-A injection.
