The Therapeutic Effects and Pathophysiology of Botulinum Toxin A on Voiding Dysfunction Due to Urethral Sphincter Dysfunction

Neurogenic and non-neurogenic urethral sphincter dysfunction are common causes of voiding dysfunction. Injections of botulinum toxin A (BoNT-A) into the urethral sphincter have been used to treat urethral sphincter dysfunction (USD) refractory to conventional treatment. Since its first use for patients with detrusor sphincter dyssynergia in 1988, BoNT-A has been applied to various causes of USD, including dysfunctional voiding, Fowler’s syndrome, and poor relaxation of the external urethral sphincter. BoNT-A is believed to decrease urethral resistance via paralysis of the striated sphincter muscle through inhibition of acetylcholine release in the neuromuscular junction. Recovery of detrusor function in patients with detrusor underactivity combined with a hyperactive sphincter also suggested the potential neuromodulation effect of sphincteric BoNT-A injection. A large proportion of patients with different causes of USD report significant improvement in voiding after sphincteric BoNT-A injections. However, patient satisfaction might not increase with an improvement in the symptoms because of concomitant side effects including exacerbated incontinence, urinary urgency, and over-expectation. Nonetheless, in terms of efficacy and safety, BoNT-A is still a reasonable option for refractory voiding function. To date, studies focusing on urethral sphincter BoNT-A injections have been limited to the heterogeneous etiologies of USD. Further well-designed studies are thus needed.


Introduction
Urethral sphincter dysfunction (USD) is one of the functional causes of voiding dysfunction (VD), which leads to slow or incomplete micturition in both males and females [1]. The condition can be either neurogenic or non-neurogenic. While detrusor sphincter dyssynergia (DSD) commonly stands for the well-known neurogenic cause [2], dysfunctional voiding (DV), Fowler's syndrome (FS), and poor relaxation of the external urethral sphincter (PRES) during voiding comprise the have been used to decrease outlet resistance, but the results of urodynamic studies have not been convincing [35]. Clean intermittent catheterization (CIC) is an effective alternative for patients to empty their bladder, but some patients are not able to tolerate it well due to upper limb impairment or psychological unwillingness [36]. Surgical sphincterotomy is another drastic option for those who fail in the treatments mentioned above, however, many of such patients experience a worsened quality of life due to persistent incontinence and a high long-term failure rate [37]. BoNT-A is thought to be an alternative, as it can block the release of acetylcholine from presynaptic vesicles at the neuromuscular junction, which causes temporary, reversible chemo-denervation of the external EUS [38].
Dykstra et al. [5] first described both transperineal and transurethral injections of BoNT-A to the EUS in 1988 as a new therapeutic approach for 11 male patients with SCI and DSD. In this preliminary study, weekly injections using different dosages of BoNT-A effectively decreased the MUP and post-void residual (PVR) urine volume. Electromyography (EMG) further confirmed sphincter denervation in all patients. The authors then conducted a small sample, randomized, double-blind placebo-control trial in five SCI male patients with DSD to compare low dose BoNT-A with normal saline by weekly injections for three weeks. They concluded that the therapeutic effects were found only in the BoNT-A injection group [39].
Instead of a once-weekly injection, a once-monthly dose has also been confirmed to effectively improve sphincter function as well as to decrease PVR [40]. In addition, a single transurethral dose of BoNT-A was further evaluated by Petit et al. [41] in 17 male patients with SCI and DSD. Their results showed similar clinical improvements in approximately 70% of the cases with an average therapeutic effect of two to six months. Phelan et al. [42] first confirmed the effectiveness of sphincteric BoNT-A injections in both male and female patients with various etiologies of DSD, including SCI, TM, and MS. De Sèze et al. [43] carried out a randomized, double-blind control trial to compare the efficacy of and tolerability to BoNT-A with a lidocaine single transperineal injection in patients with DSD. Higher patient satisfaction scores and significant decreases in PVR, MUP, and EMG activity were found only in the BoNT-A group. Most of the patients tolerated the treatment well without major complications.
Recently, more data describing the clinical experiences related to sphincteric BoNT-A injections in patients with DSD have been reported. Both transperineal injections [5,40,[43][44][45][46][47] and transurethral injections using cystoscopy [5,[40][41][42][48][49][50][51][52][53] showed promising outcomes in terms of reducing sphincteric activity. Although transurethral injections have been reported to be more effective in terms of decreasing the MUP than the use of a transperineal route [40], no further trials have ever directly compared these two injection methods. Most of the recent publications on this topic involve the use of a sphincteric injection of 100 units of Botox in DSD patients since a more prolonged therapeutic effect was found, as compared to 50 units, in a previous randomized control trial [54].
Urodynamic parameters were commonly used as measurements of objective outcomes after sphincteric BoNT-A injections in patients with DSD. As expected, a significant reduction in EMG activity [5,39,43] and a decrease in MUP [5,40,41,43,44] during the voiding phase have been found. Improvements in PVR [5,[40][41][42][43]49], detrusor contraction pressure [5,44,48,49], and maximal flow rate (Qmax) [42,48] were also reported as a result of decreased sphincteric resistance. However, unlike SCI patients, a randomized, double-blind, placebo-control trial in 86 patients with MS and DSD showed that a single injection of BoNT-A did not decrease the PVR [44]. The authors posited that the unchanged PVR could be attributed to the lower baseline detrusor contraction pressure. In addition to the outcome measurement, urodynamic parameters have also been used to predict therapeutic outcomes. Several pre-treatment urodynamic parameters, including higher baseline detrusor contraction pressure [41,53], lower baseline sphincteric tone [48,50,55], and a synergic bladder neck [53,55] have been confirmed as predictors of a favorable outcome.
Although objective urodynamic results have been improved after EUS BoNT-A injection, Kuo et al. [52] reported inconsistencies between urodynamic outcomes and patient satisfaction. Even though the PVR and detrusor contraction pressure were improved, patients were not satisfied with the outcomes mainly because of the increased incontinence grade [51]. Notably, for patients with SCI and DSD, a detrusor BoNT-A injection provided a much better quality of life than the case with an EUS injection [52] This result emphasized the importance of continence in a patient's quality of life. Other causes of patient dissatisfaction included persistent difficult urination [51], increases in urgency episodes [51], and the need for repeated injections [45]. For therapeutic outcomes, patient satisfaction was mainly due to improved voiding conditions and fewer autonomic dysreflexia (AD) episodes [52,53]. To summarize the subjective outcomes of sphincteric BoNT-A injections for patients with DSD, 61-88% experienced clinical improvement [5,[40][41][42]48,51] for a therapeutic duration lasting two to six months [40][41][42]44], more than 80% could regain spontaneous voiding and successfully remove the indwelling catheter or stop CIC [42,48,49], and episodes of autonomic dysreflexia were reduced in more than 50% of the cases [5,43,50,51]. Table 1 summarizes the clinical studies on efficacy and adverse events related to sphincteric BoNT-A injections in the treatment of patients with DSD.
In spite of the benefits of sphincteric BoNT-A injections in patients with DSD based on the literature, the use of different injection protocols among various etiologies makes intergroup comparisons difficult. Further randomized control trials with large case numbers focusing on a single etiology of DSD are necessary to evaluate the therapeutic impacts on both objective and subjective parameters, quality of life, duration of effect, and long-term durability after repetitive injections.  Sphincteric injections used with preparation other than the typical BoNT-A commercial form, including Botox or Disport, were denoted as "BoNT-A". a UDS improvements were found in the urethral Botox group. b Individual results in specific disease groups were not available.

Urethral Sphincter BoNT-A Injections in Children with Dysfunctional Voiding
DV is characterized by an intermittent or fluctuating flow rate, owing to intermittent contractions of the periurethral striated muscles or pelvic floor muscles during voiding in neurologically normal patients [56]. In 1973, Hinman and Baumann first described the symptom complex including enuresis, daytime wetting, UTI, and upper tract dilatation in 14 boys without neurologic defects and suggested that the condition is a functional discoordination between detrusor contraction and external sphincter relaxation [57]. This syndrome was then described by other authors as Hinman syndrome, occult neuropathic bladder, non-neurogenic neurogenic bladder, learned voiding dysfunction, and dysfunctional voiding [58][59][60]. In children, the typical symptoms of DV include urinary incontinence, recurrent UTI, voiding difficulty, urinary retention, and hydronephrosis [61]. To establish the diagnosis, uroflowmetry with an EMG is required to confirm that a sudden change in flow rate in the form of a staccato or intermittent pattern is related to sphincter contraction. Also, a "spinning-top" urethra can also be seen in a video-urodynamic study (vUDS) or voiding cystourethrography, indicating discoordination of the EUS and detrusor contraction during voiding [56].
The conventional treatments for children with DV include non-pharmacological urotherapy [62,63] and alpha-blockers [64]. Since DV and DSD share similar pathophysiology in terms of abnormal sphincteric activity during voiding, applying BoNT-A to the EUS seems to be a reasonable therapeutic option. A sphincteric BoNT-A injection was first introduced as a novel treatment for children with DV by Steinhardt et al. [65], who successfully improved incontinence and recurrent UTI in a 7-year-old girl, and also demonstrated a marked improvement in the degree of urethral dilatation.
Several case series with small samples also discussed the therapeutic outcome of BoNT-A in children with DV who failed traditional urotherapy and medical management [66][67][68][69][70]. According to these data, 80-85% of patients showed improvement in daytime incontinence or enuresis [68,69], total dryness was found in 45-80% of patients after sphincteric BoNT-A injections [67][68][69][70], and approximately 45-75% of patients were free from recurrent UTI even without prophylaxis antibiotics [68][69][70]. A small case series reported by Mokhless et al. [66] revealed that nine children who were catheterized preoperatively experienced recovery of spontaneous voiding after sphincteric BoNT-A injections. In the case of urodynamic parameters, PVR improvement was found in most of the studies, and a flow pattern changed to bell-shaped curve was also reported [69,70]. Unlike the usual dose of 50 to 100 units of Botox in pediatric sphincter injections, Franco et al. [67] used a higher dose ranging from 200 to 300 units in 16 children with DV. They reported long-lasting improvements in PVR at six months, and the majority of their patients did not require repeated injections. The authors hypothesized that BoNT-A could block sensory feedback of overactive guarding reflex, making it possible to retrain these children to void appropriately. No acute complications, including nausea, dysphagia, respiratory distress, or paralysis, were found in any of these studies. Clinical studies of sphincteric BoNT-A injections for children with DV are summarized in Table 2.
Although the effects and safety of BoNT-A use in children with DV seem to be convincing, we should remember that all these study designs were nonrandomized, without controlled variables, and comprised small samples. Further better-designed trials with longer follow up are necessary to arrive at an accurate conclusion.

Urethral Sphincter BoNT-A Injections in Adults with Dysfunctional Voiding
The precise prevalence of DV in the adult population is still unknown. In a urodynamic database review of 1015 adults who were evaluated for voiding symptoms, around 2% could be defined as having DV using strict vUDS criteria [60]. Adult DV may come from persistent disease since childhood or adult-onset symptoms due to non-neurological etiologies [72]. Although adults and children with DV share similar characteristics and are defined similarly [56,73], the clinical characteristics of these two groups are quite different. Unlike children, adult patients typically present with obstructive symptoms, followed by frequency, nocturia, and urgency. Recurrent UTI and urinary incontinence are less prominent in adults [60].
Data discussing the therapeutic effect of sphincteric BoNT-A injections in adults with DV are limited and are mostly provided by Kuo and his colleagues [48,55,71]. In a prospective nonrandomized study without controlled variables, the authors performed sphincteric injections using 50 to 100 units of Botox in 20 adults with DV and reported a subjectively excellent outcome in 30% of the patients, where the remaining 70% showed improvement [48]. Liao and Kuo also reported an overall success rate of 86.7% in adults with DV by sphincteric injections with 50 to 100 units of Botox in a five-year retrospective review. DU with low abdominal straining pressure, spastic EUS, and bladder neck obstruction were the most common causes of treatment failure [55].
A randomized, double-blind, placebo-controlled study was conducted in 31 adults with DV to compare the therapeutic effect of 100 units of Botox with normal saline [71]. Even though the detrusor voiding pressure and voided volume were significantly improved in the BoNT-A group, there were no significant between-group differences in the subjective success rate. The author hypothesized that the local injection itself might have some unknown therapeutic effects on the relaxation of the EUS [71]. This concept is similar to the dry needling effect on myofascial trigger point pain, which can relax the actin-myosin bonds and normalize muscle tone [74]. Additional well-designed studies to enroll more adult patients with DV are necessary to elucidate the therapeutic effect of sphincteric BoNT-A injections, normal saline injections, or even the dry needle effect. Clinical studies of sphincteric BoNT-A injections for adults with DV are summarized in Table 2.

Urethral Sphincter BoNT-A Injections in Patients with Fowler Syndrome
Fowler's syndrome (FS), a specific cause of unexplained urinary retention in young women, was first described by Fowler in 1986 [75]. The condition is characterized by EUS relaxation failure with unique components of complex repetitive discharges and decelerating bursts presented in concentric needle EMG [3]. The typical feature of FS in vUDS include a large bladder capacity, reduced bladder sensation in the storage phase, decreased or no detrusor contraction with a patent bladder neck, and narrowing in the midurethra with or without ballooning of the proximal urethra [76]. The decrease in sensation and motor function in the bladder were thought to be a result of abnormally strong urethra afferent activity, which inhibits the bladder afferent signals to reach the brain as a spinal 'pro-continence' mechanism [77]. These findings are different from the pattern of the typical pattern of high-pressure low-flow in DV also caused by involuntary EUS or pelvic floor muscle contraction during voiding [78]. Whether FS is a subgroup of DV or a totally different entity remains currently unanswered.
Few studies have evaluated the effect of BoNT-A on the management of FS [55,79,80]. The first study was performed by Fowler and colleagues in 1992, where six women with FS were enrolled [79]. Two hundred units of BoNT-A (Division of Biologics, Porton Down, Salisbury, UK) were given to one side of the EUS under EMG guiding via a hollow cannula electrode. No improvements in voiding function were noted in any of the patients. One patient even developed transient stress urinary incontinence. In 2007, Liao and Kuo also reported no restoration of efficient voiding in two patients suspected to have FS with high MUP lacking a typical abnormal needle EMG pattern after injections with 100 units of Botox in four to eight EUS sites [55]. However, decreases in MUP and abdominal voiding pressure by 20 to 25 cm H 2 O after injections were noted by vUDS during follow up. In contrast to the poor outcome in the aforementioned studies, a 10-patient open-level pilot study in 2016 did find promising outcomes in the management of FS using BoNT-A [80]. The injections were done with 1 mL 2% lidocaine on either side of the external urethral meatus, followed by 100 units of Botox equally divided on either side of the EUS under EMG guidance. Four of five women with complete urine retention could void spontaneously four weeks after injections. Seven of the 10 women stopped CIC ten weeks after injections. Significant improvement in the Qmax, PVR, International Prostate Symptom Score (IPSS), and urethral pressure profile were also noted at ten weeks. No serious adverse effects were reported. Clinical studies on sphincteric BoNT-A efficacy and adverse FS events are summarized in Table 3.
Due to the rarity of the disease, the difficulties associated with arriving at a definitive diagnosis that needs special equipment, the techniques required for performance, and interpretation of concentric needle EMG, these studies were all limited to a small number of patients without adequate control groups. Further large cohort studies are needed to validate these outcomes. The contradictory findings might be the result of different BoNT-A injection techniques or the different etiology behind this disease. Compared to sacral neuromodulation, BoNT-A urethra injections might serve as a less invasive, low resource, safer alternative to other methods used to treat this disease.  None of these studies were randomized or controlled. Sphincteric injections were given with preparation other than typical BoNT-A commercial form including Botox or Disport denoted as "BoNT-A". a The subjects enrolled were not typical FS patients. They had a very high baseline MUP but did not have typical patterns of FS presented in a concentric needle electromyographic study. b Data were analyzed using combined groups. Individual results for specific disease groups were not available.

Urethral Sphincter BoNT-A Injections in Patients with Poor Relaxation of The External Urethral Sphincter
PRES as a diagnosis was first described by Kuo in 2000 and was determined based on non-relaxed surface EMG activity combined with a narrow membranous urethra during the voiding phase in the vUDS [82]. It was believed to have a different pathophysiology beyond prostatic obstruction or bladder neck dysfunction in non-neurogenic male voiding dysfunction refractory to alpha-blocker or transurethral resection of prostate [83,84]. The concept was further applied to non-neurogenic females with the same EMG findings and narrowing of the distal urethra in vUDS [85]. The cause of PRES was posited to be multifactorial, including learned habituation, pelvic floor hypertonicity, increased bladder sensitivity, or occult neuropathy [86]. However, the exact etiology responsible for the poor relaxation of the EUS or pelvic floor remains to be elucidated [82].
The cardinal symptoms of PRES are hesitancy, small urine caliber, and terminal dribbling with an IPSS voiding-to-storage subscore ratio >1 [82,87]. PRES is characterized by relatively small but stable bladder [88] and low-pressure low-flow during voiding phase [89], which is different from the typical high-pressure low-flow presentation in DV or extremely large, compliant bladder in FS. The prevalence rates were 12-20% [87,89,90] and 17.6% [4] in male and female non-neurogenic VD refractory to medication patients, respectively. The incidence increased in young males [89] in patients with bladder pain syndrome [88] and in idiopathic DU patients [91]. Sphincteric BoNT-A injections might provide chemo-denervation of the EUS by inhibition of acetylcholine release in the neuromuscular junction to relieve the USD in PRES [23].
The improvement rate in clinical or urodynamic parameters after injection of 100 units of BoNT-A in patients with PRES was reported to be 79 to 96% [48,55]. However, with a stricter definition of excellent outcome, only 42% of such patients had restored spontaneous voiding or had a >25% improvement in urodynamic parameters [48]. Great patient satisfaction was also reported to be approximately 47-52% [81]. It was concluded that the major predictive factors for a successful outcome were opening of the bladder neck and a higher baseline Qmax, but not the type of USD [81]. An increased recovery rate of detrusor contractility was further reported in idiopathic DU combined with PRES [25]. This result supported the hypothesis suggesting that the low-pressure low-flow dysfunction presented in PRES might be the result of detrusor suppression induced by non-relaxed EUS activity. With the aid of EUS relaxation after BoNT-A injections, the suppressed detrusor function was resumed. Clinical studies on sphincteric BoNT-A efficacy and adverse events of PRES are summarized in Table 3. Notably, most of the therapeutic effects of EUS BoNT-A injections in PRES came from studies conducted by Kuo's research group. Further work from other clinical facilities and laboratories might lead to more prudent inferences.

Conclusions
In recent decades, BoNT-A has been used in the treatment of VD caused by various types of USD. It has been reported to be effective in the management of DSD, DV, PRES, and has shown promise in treating FS. However, patient satisfaction might not correlate well with objective improvement. BoNT-A injections may serve as a less invasive and safer option in treatment of USD refractory to conventional medications. The mechanism by which BoNT-A improves USD is thought to be a result of a decrease in urethral resistance via inhibition of acetylcholine released in the presynaptic neuron of the EUS, and through the recovery of detrusor muscle contractility via neuromodulation. Studies focused on BoNT-A injections at the EUS have often been limited to distinct etiologies of USD. Further well-designed clinical and basic studies are needed to confirm its effect.