History. The clinician should carefully elicit the patient's bladder symptoms to document duration of symptoms and baseline symptom levels, to ensure that symptoms are not the consequence of some other condition and to determine whether the patient constitutes a complex OAB presentation that may require referral. Questions should assess bladder storage symptoms associated with OAB (e.g., urgency, urgency incontinence, frequency, nocturia), other bladder storage problems (e.g., stress incontinence episodes) and bladder emptying (e.g., hesitancy, straining to void, prior history of urinary retention, force of stream, intermittency of stream). The symptom of urgency as defined by IUGA/ICS is the "complaint of sudden compelling desire to pass urine which is difficult to defer."^4,27 The interpretations of "sudden" and "compelling" are highly subjective and difficult to quantitate. Nevertheless, the clinician can simply ask if the patient has a problem getting to the bathroom in time, assuming the patient has normal mobility. 

Bladder function is related to amount and type of fluid intake. Excessive fluid intake can produce voiding patterns that mimic OAB symptoms. For this reason, an inquiry into fluid intake habits should be performed, including asking patients how much fluid and of what type (e.g., with or without caffeine) they drink each day, how many times they void each day and how many times they void at night. Patients who do not appear able to provide accurate intake and voiding information should fill out a fluid diary. Urinary frequency varies across individuals. In community-dwelling healthy adults, normal frequency consists of voiding every three to four hours with a median of approximately six voids a day.^{30, 31}  Current medication use also should be reviewed to ensure that voiding symptoms are not a consequence of a prescribed medication, particularly diuretics.

The degree of bother from bladder symptoms also should be assessed. If a patient is not significantly bothered by his/her bladder symptoms, then there would be less compelling reason to treat the symptoms. Degree of bother can affect different domains of daily activities related to work and leisure. Patients may avoid certain activities (e.g., travel, situations that do not allow easy access to a toilet) because of their bladder symptoms. 

Co-morbid conditions should be completely elicited as these conditions may directly impact bladder function. Patients with co-morbid conditions and OAB symptoms would be considered complicated OAB patients. These co-morbid conditions include neurologic diseases (i.e., stroke, multiple sclerosis, spinal cord injury), mobility deficits, medically complicated/uncontrolled diabetes, fecal motility disorders (fecal incontinence/constipation), chronic pelvic pain, history of recurrent urinary tract infections (UTIs), gross hematuria, prior pelvic/vaginal surgeries (incontinence/prolapse surgeries), pelvic cancer (bladder, colon, cervix, uterus, prostate) and pelvic radiation. The female patient with significant prolapse (i.e., prolapse beyond the introitus) also may be considered a complicated OAB patient. Patients with urgency incontinence, particularly younger patients, or a patient with extremely severe symptoms could represent a complicated OAB patient with an occult neurologic condition. A patient who has failed multiple anti-muscarinics to control OAB symptoms could also be considered a complicated OAB patient. If the history elicits any of these co-morbid conditions and/or special situations, then the clinician should consider referring these patients to a specialist for further evaluation and treatment. 

Physical Examination. A careful, directed physical exam should be performed. An abdominal exam should be performed to assess for scars, masses, hernias and areas of tenderness as well as for suprapubic distension that may indicate urinary retention. Examination of lower extremities for edema should be done to give the clinician an assessment of the potential for fluid shifts during periods of postural changes. A rectal/genitourinary exam to rule out pelvic floor disorders (e.g., pelvic floor muscle spasticity, pain, pelvic organ prolapse) in females and prostatic pathology in males should be performed. In menopausal females, atrophic vaginitis should be assessed as a possible contributing factor to incontinence symptoms. The examiner should assess for perineal skin for rash or breakdown. The examiner also should assess perineal sensation, rectal sphincter tone and ability to contract the anal sphincter in order to evaluate pelvic floor tone and potential ability to perform pelvic floor exercises (e.g., the ability to contract the levator ani muscles) as well as to rule out impaction and constipation. 

Cognitive impairment is related to symptom severity and has therapeutic implications regarding goals and options. The Mini-Mental State Examination (MMSE)³² is a standardized, quick and useful assessment of cognitive function. An MMSE should be conducted on all patients who may be at risk for cognitive impairment to determine whether symptoms are aggravated by cognitive problems, to ensure that they will be able to follow directions for behavioral therapy and/or to determine the degree of risk for cognitive decline with anti-muscarinic therapy. In the Panel's experience, the ability of the patient to dress independently is informative of sufficient motor skills related to toileting habits.  

Urinalysis.  A urinalysis to rule out UTI and hematuria should be performed. A urine culture is not necessary unless indication of infection (i.e., nitrites/leukocyte esterase on dipstick, pyuria/bacteriuria on microscopic exam) is found and may be done at the discretion of the clinician. If evidence of infection is detected, then a culture should be performed, the infection treated appropriately and the patient should be queried regarding symptoms once the infection has cleared. If evidence of hematuria not associated with infection is found, then the patient should be referred for urologic evaluation.  

Urine culture. Urinalysis is unreliable for identification of bacterial counts below 100,000CFU/mL. In some patients with irritative voiding symptoms but without overt signs of infection, a urine culture may be appropriate to completely exclude the presence of clinically significant bacteriuria. 

Post-void residual (PVR). Measurement of the PVR is not necessary for patients who are receiving first-line behavioral interventions or for uncomplicated patients (i.e., patients without a history of or risk factors for urinary retention) receiving anti-muscarinic medications. Because anti-muscarinic medications can induce urinary retention,³³ particularly in complicated patients with retention risk factors, PVR should be assessed in patients with obstructive symptoms, history of incontinence or prostatic surgery, neurologic diagnoses and in other patients at clinician discretion when PVR assessment is deemed necessary to optimize care and minimize potential risks. It should be noted, however, that the occurrence of symptomatic urinary retention or asymptomatic elevations of PVR after the addition of anti-muscarinic agents occurs in a small proportion of patients; previously undiagnosed poor detrusor function may be unmasked in those individuals.

PVR should be measured with an ultrasound bladder scanner immediately after the patient voids.  If an ultrasound scanner is not available, then urethral catheterization may be used to assess PVR. For any patient on anti-muscarinic therapy, the clinician should be prepared to monitor PVR during the course of treatment should obstructive voiding symptoms appear. As there is considerable overlap between storage and emptying voiding symptoms, baseline PVRs should be performed for men with symptoms prior to initiation of anti-muscarinic therapy. Anti-muscarinics should be used with caution in patients with PVR >250-300 mL.³⁴ Most randomized trials that evaluated anti-muscarinics for OAB treatment used a PVR of 150-200 mL as an exclusion criterion; the overwhelming majority of participants in these trials were women.  

Bladder diaries. Diaries that document intake and voiding behavior may be useful in some patients, particularly the patient who cannot describe or who is not familiar with intake and voiding patterns. Diaries also are useful to document baseline symptom levels so that treatment efficacy may be assessed. 

In particular, self-monitoring with a bladder diary for three to seven days is a useful first step in initiating behavioral treatments for OAB. At a minimum, the patient documents the time of each void and incontinence episode and the circumstances or reasons for the incontinence episode. Rating the degree of urgency associated with each void and incontinence episode also can be useful. Adding measures of voided volumes can provide a practical estimate of the patient's functional bladder capacity in daily life and estimate the amount of overall fluid intake. Recording voided volumes also can be useful to differentiate between polyuria (characterized by normal or large volume voids) from OAB (characterized by frequent small voids). It is more burdensome, however, and is usually completed for only 24 to 48 hours. 

The bladder diary is a useful tool for both the clinician and the patient. In the evaluation phase, it provides information that can help the clinician plan appropriate components of intervention, particularly behavioral intervention. Recording the times that the patient voids provides a foundation for determining voiding intervals in bladder training programs. During the course of treatment, it can be used to monitor symptoms to track the efficacy of various treatment components and guide the intervention. For the patient, the self-monitoring effect of completing the diary enhances awareness of voiding habits and helps them recognize activities that can trigger incontinence. Twenty-four hour pad weights also can provide useful information regarding the severity of incontinence symptoms.

Symptom questionnaires. Validated symptom questionnaires have been utilized in OAB clinical trials to quantitate bladder symptom and bother changes with OAB therapies. Among these questionnaires are the Urogenital Distress Inventory (UDI), the UDI-6 Short Form, the Incontinence Impact Questionnaire (II-Q) and the Overactive Bladder Questionnaire (OAB-q).^35-37 The rationale for utilization of these validated questionnaires is to quantitate and follow the patients' responses to OAB treatment as well as to obtain baseline and post-treatment levels of bother.

Urodynamics, cystoscopy and diagnostic renal and bladder ultrasound are not recommended in the initial diagnostic workup of the uncomplicated OAB patient. For complicated patients or refractory patients who have failed multiple OAB treatments, the choice of additional diagnostic tests depends on patient history and presentation and clinician judgment. In some cases, additional information may make clear that the patient has neurogenic OAB rather than non-neurogenic OAB and requires a different treatment plan. Patients with hematuria should be referred for a urologic work up. In the low-risk uncomplicated patient without microscopic hematuria, urine cytology is infrequently associated with atypia requiring further investigation, engendering costs and possibly resulting in morbidity. Urine cytology is not recommended in the routine evaluation of patients with uncomplicated OAB without hematuria who respond to therapy.

Initiating treatment for OAB generally presumes that the patient can perceive an improvement in his or her QOL. In patients who cannot perceive symptom improvements, treatment may not be appropriate, may be potentially unsafe or may be futile (e.g., in the very elderly or demented patient) except in patients for whom OAB symptoms present a significant health risk (e.g., risk for skin breakdown). It is important for clinicians who treat this problem to recognize this issue and to set feasible therapeutic goals with the patient and/or caregiver. The presence of an overactive bladder frequently accompanies other disorders such as deficiencies in cognition (i.e., dementia) and/or poor mobility, which can complicate treatment. To treat incontinence, optimally the patient must have a desire to be continent or have a desire for symptom improvement. In patients with cognitive deficits, this desire may not be present and family and/or caregivers may have difficulty understanding that simply giving a medication will not correct the problem. The other common situation associated with OAB is severely reduced mobility. Causes can range from dementia, severe arthritis, severe obesity, hemiparesis/plegia, and lower extremity amputations. In these situations, despite receiving an urge to urinate, the patient physically cannot get from their current position without assistance to a toileting facility. This cannot be corrected pharmacologically and should be recognized by the treating physician.

In certain patients for whom hygiene and skin breakdown are major concerns, treatment may be considered regardless of the patient's perceptions when it is in the patient's best interests. In these patients, behavioral strategies that include prompted voiding and fluid management may be helpful. Pharmacologic treatments and invasive treatments, however, are generally not appropriate for these individuals. The Panel also recognizes that untreated incontinence can result in falls when a patient with compromised mobility attempts to move quickly to a toileting facility. The treating physician, the patient and the caregiver must weigh these risks when making the decision whether and how to treat OAB.

Prior to initiating treatment, the clinician should provide patient education regarding normal and abnormal bladder function, including voiding frequency and toileting behavior. Explaining what is normal can help the patient understand how their condition diverges from normal and gives them a comparator (or goal) for judging their own progress in treatment. Education also empowers the patient to engage and participate in their treatment, which is essential when using interventions that rely on behavior change. Patients must understand that voiding is a behavior that can be managed and that successful OAB treatment requires a willing participant who is informed and engaged in the treatment process. 

Patients should be informed that OAB is a symptom complex with a variable and chronic course that needs to be managed over time, that it primarily affects QOL, that there is no single ideal treatment and that available treatments vary in the effort required from the patient as well as in invasiveness, risk of adverse events and reversibility. An effective treatment plan depends on patients having realistic goals for treatment and a clear understanding of the risks and burdens of particular treatments. In this context, it is important to understand the patient's expectations of treatment, not only in terms of its outcome, but regarding what is required of them as well. Expectations are important because they affect motivation and adherence, and they can influence the patient's interpretation of treatment effects and satisfaction with outcomes.

Most OAB treatments can improve patient symptoms but not eliminate them. The available OAB treatments, with the exception of behavioral therapies, present risks for adverse events, some of which are serious. When initiating behavioral interventions, it is crucial that the patient understands that treatment progress and outcomes will depend on their active participation and persistence over time. It is also useful for them to understand several other aspects of behavioral change: progress is usually gradual, change can be irregular with good days and bad days and long-term change in symptoms depends on their long-term change in behavior.

Patients may decide that the symptomatic improvement achieved with a particular therapy (e.g., from 5 incontinence episodes per day to 3 incontinence episodes per day) is not worthwhile given the adverse events associated with that treatment (e.g., dry mouth and constipation associated with anti-muscarinic therapy) and choose to discontinue therapy despite symptomatic improvement. Choosing to forego treatment is a valid decision. It is the opinion of the Panel that patients seeking treatment initially and at any point in the treatment algorithm should be told that they may opt for no treatment with minimal adverse effects on their health and no impact on the success of later management should they choose to pursue treatment in the future.

Behavioral treatments are designated as first-line treatments because they are as effective in reducing symptom levels as are anti-muscarinic medications, and they consist of many components that can be tailored to address the individual patient's needs and capacities. In addition, they are relatively non-invasive and, in contrast to medications, are associated with virtually no adverse events. They do require the active participation of the patient and/or of the patient's caregiver, however, as well as time and effort from the clinician. Behavioral therapies should be offered to all OAB patients, including OAB patients who require a caregiver; caregivers can be instructed in behavioral techniques in order to optimize patient symptom control (i.e., prompted voiding, timed voiding). 

Most of the literature on overactive bladder focuses on the treatment of urinary incontinence, and most studies have been performed with women. Although most patients do not experience complete symptom relief with behavioral intervention, the literature indicates that most patients experience significant reductions in symptoms and improvements in QOL. The literature provides clear support for the effectiveness of bladder training (incremental voiding schedules done with distraction and self-assertions)^41,42 and behavioral training (PFMT with urge suppression techniques).⁵³ Typical mean improvements range from 50% to 80% reduction in the frequency of incontinence. Reductions in voiding frequency have also been documented in men⁴³ and women.^{41, 44, 45}

There is also a good body of literature addressing the effects of weight loss on incontinence specifically. The most definitive trial reported that a six-month behavioural weight loss intervention resulted in an 8.0% weight loss in obese women, reduced overall incontinence episodes per week by 47% (compared to 28% in the control group) and reduced UUI episodes by 42% (compared to 26% in controls).⁴⁰

One study evaluated fluid management and reported that a 25% reduction in fluid intake reduced frequency and urgency.⁴⁶ The Panel notes that when attempting intake reduction, baseline intake levels must be considered to determine whether reduction is appropriate. There is also evidence from a study of bladder training that reducing caffeine intake results in greater reductions in voiding frequency.⁴⁷

Based on a limited literature, no single component of behavioral therapy appears to be essential to efficacy, and no single type of behavioral therapy appears to be superior in efficacy. In comparing behavioral training that was administered with biofeedback, with verbal feedback or self-administered using a pamphlet, all three approaches had similar effects to reduce incontinence and increase bladder capacity.⁴⁸ Patients in the two feedback conditions, however, reported greater treatment satisfaction and better perceptions of symptom control, suggesting that feedback may be important in subjective outcomes. However, in a comparison of PFMT with and without biofeedback, incontinence, pelvic muscle strength and QOL improved more in the group that received feedback.⁴⁹ In a study that compared PFMT to bladder training or PFMT in combination with bladder training, patients in the combined group initially had greater incontinence reductions and QOL improvements; however, at 3 month follow-up all three groups had similar improvement levels.⁵⁰

The literature review of comparative effectiveness randomized trials indicated that various types of behavioral treatment were generally either equivalent to^{44, 51, 52} or superior to^{42, 45, 53} medications in terms of reducing incontinence episodes, improving voiding parameters such as frequency^{43, 54, 55} and nocturia⁵⁶ and improving QOL. Most studies evaluated oxybutynin (both the IR and the ER formulations).^{43, 44, 51-53} One study evaluated tolterodine.⁵⁴ One study evaluated flavoxate hydrochloride and imipramine.⁴² One study evaluated trospium.⁴⁵

The Panel interpreted these data to indicate that behavioral therapies can result in symptomatic improvements similar to anti-muscarinics without exposing patients to adverse events. Evidence strength is Grade B because although the majority of studies were randomized trials and findings were generally consistent across studies (both randomized and observational), most of the randomized trials were of moderate quality, follow-up durations were short in most studies (12 weeks) and sample sizes were small. 

Behavioral and drug therapies are often used in combination in clinical practice to optimize patient symptom control and QOL. A limited literature indicates that initiating behavioral and drug therapy simultaneously may improve outcomes, including frequency, voided volume, incontinence and symptom distress.^{45, 54, 57-59} In patients who are not adequately improved on behavioral or drug therapy alone, there also is evidence that continuing the initial therapy and adding the alternate therapy using a stepped approach can produce additional benefit.⁶⁰ In the Panel's judgment there are no known contraindications to combining pharmacologic management and behavioral therapies. Evidence strength is Grade C because of the limited evidence base consisting of relatively few trials, small sample sizes, and limited follow-up durations.

Anti-muscarinic medications. The choice of oral anti-muscarinics as second-line therapy reflects the fact that these medications reduce symptoms but also can commonly have non-life-threatening side effects such as dry mouth, constipation, dry or itchy eyes, blurred vision, dyspepsia, UTI, urinary retention and impaired cognitive function. Rarely, life-threatening side effects such as arrhythmias have been reported. An extensive review of the randomized trials that evaluated pharmacologic therapies for OAB (including trials with placebo control groups as well as trials with active treatment comparison groups) revealed no compelling evidence for differential efficacy across medications.^{44, 51-54, 57, 61-125} This finding is consistent with the conclusions of several published systematic reviews.^126-129 

These data were not suitable for meta-analysis due to of lack of variance information (e.g., standard deviations, variances, standard error of the mean) for outcomes in many studies. Qualitative analysis revealed, however, that for 24-hour frequency, urgency incontinence and incontinence, baseline symptom level was closely related to degree of symptom reduction across medications. Specifically, patients with more severe symptoms, on average, experienced greater symptom reductions. For urgency incontinence and total incontinence episodes, only patients with relatively low baseline symptom levels were likely to experience complete symptom relief. 

This relationship was evident both within and across medications regardless of study inclusion criteria or dosing regimens (see Figure 1 for UUI data). 

Figure 1. Baseline UUI (episodes/day) and UUI Reduction (episodes/day) for randomized trials by drug.

For urgency and nocturia, however, there was no apparent relationship between baseline symptom levels and symptom reduction (See Figure 2 for urgency data). 

Figure 2. Baseline urgency (episodes/day) and urgency reduction (episodes/day) for randomized trials by drug.

Due to the similar efficacy observed for all oral anti-muscarinic medications, the choice of medication for a particular patient depends on the patient's history of anti-muscarinic use, information regarding adverse events experienced in the past, the impact on the patient of adverse events, patient preferences, comorbidities, use of other medications and the availability of and resources to acquire specific medications. In addition, although there was no evidence of differential efficacy across medications, both qualitative analysis and meta-analysis of all randomized trial arms revealed different adverse event profiles for dry mouth and constipation.* This information may be relevant if a patient is particularly sensitive to one of these adverse events. 

With regard to dry mouth, meta-analysis revealed that on average 6.90% of placebo patients experienced dry mouth (40 placebo arms; 95% CI: 5.6% to 8.5%). Rates of dry mouth in active drug treatment arms for the newer medications (i.e., darifenacin - 9 arms, fesoterodine - 11 arms, solifenacin - 15 arms) and for trospium (8 arms) ranged from 20.0% to 40.0%. Within each medication, there was no clear relationship between rate and dose. Across medications, rates were statistically indistinguishable with overlapping confidence intervals and derived from relatively few trial arms for each medication; the Panel interpreted these findings as preliminary and descriptive rather than definitive until more data are available. 

The majority of the available studies evaluated oxybutynin (25 trial arms) and tolterodine (40 trial arms). The rate of dry mouth for oxybutynin at 61.4% was statistically significantly higher (95% CI: 52.5% to 69.5%) than the 23.7% rate for tolterodine (95% CI: 20.7% to 26.9%) (p<0.001). Although there was no clear relationship with dose, there was heterogeneity within each medication based on whether the immediate release (IR) or the extended release (ER) formulation was administered. 

With regard to constipation, on average 3.6% of placebo patients experienced this adverse event (36 placebo arms; 95% CI: 2.7% to 4.8%). Constipation rates in active drug treatment arms for fesoterodine (11 arms), solifenacin (15 arms), and trospium (5 arms) ranged from 7.0% to 9.0%. These rates were statistically indistinguishable with similar 95% confidence intervals spanning 5.0% to 12.0%. The constipation rate for darifenacin (9 arms), however, was significantly higher at 17.0% (95% CI: 13.0% to 21.0%). Within each medication, there was no clear relationship between rate and dose. Since these data were derived from relatively few trial arms, the Panel again interpreted them as descriptive rather than definitive until more data are available. 

The majority of the available studies evaluated oxybutynin (21 trial arms) and tolterodine (34 trial arms). The rate of constipation for oxybutynin was 12.1% (95% CI: 7.9% to 18.0%). The rate for tolterodine was statistically significantly lower (p<0.001) at 4.9% (95% CI: 4.1% to 5.7%). There were no differences based on dose or between the IR and ER formulations for either medication. 

The Panel interpreted the oxybutynin and tolterodine data to indicate that the probability that a patient will experience dry mouth and/or constipation appears to be higher overall with the administration of oxybutynin compared to tolterodine. Evidence strength was Grade B because most trials were of moderate quality and follow-up durations were relatively short (i.e., 12 weeks).

β 3-adrenoceptor agonists: Mirabegron. The 2014 update literature review retrieved a newly-published set of studies that evaluated the benefits and risks/burdens of mirabegron, a β₃-adrenoceptor agonist, for overactive bladder. Seven studies evaluated mirabegron in comparison to a placebo group and/or an active control group^{130,131,132, 133, 134,135, 136} in a total of 9,310 patients; 5,884 of these patients were in the mirabegron groups. For the five studies that used an active control group, the active control was tolterodine ER 4 mg. Five studies were Phase III trials evaluating safety and efficacy. One study was a Phase II proof-of-concept study,¹³⁶ and one study was a Phase II dose-ranging trial.¹³⁵ Five studies followed patients for 12 weeks. The proof-of-concept study followed patients for four weeks.¹³⁶ One of the Phase III trials followed patients for one year.¹³³ Inclusion criteria were similar across studies, generally requiring patients to have OAB symptoms for ≥ three months, ≥ eight voids/day, and at least three urgency episodes over a three-day period. Patient ages were similar across studies, ranging from mean 55.4 years to mean 61 years. All studies used a two-week single-blind placebo run-in period. Baseline symptom levels were remarkably similar across studies, with baseline voids/24 hours ranging from mean 10.9 to mean 12.3, baseline incontinence episodes ranging from mean 2.4 to 3.6 but with all but one study arm in the 2.4 to 3.0 range, and baseline UUI episodes ranging from 1.7 to 3.5 but with all but one study arm in the 1.7 to 2.7 range. Only two studies measured urgency episodes, and patients ranged from mean 4.1 episodes per day to mean 6.6 episodes per day.^134,136 The same two studies also reported nocturia outcomes; at baseline patients had mean 1.7 to 2.1 episodes per night. 

All studies focused on voids per day and incontinence episodes. Some studies also reported urge urinary incontinence, urgency episodes, and nocturia. Most studies included some measure of QOL (e.g., Treatment Satisfaction-Visual Analogue Scale, Patient Perception of Bladder Condition, OAB-q). In general, most mirabegron doses produced statistically significant symptom reductions for voids per day and incontinence episodes per day compared to placebo. Improvements in UUI, urgency episodes, and QOL measures also occurred but were not as consistently statistically significant. Among studies with an active control group administered tolterodine ER 4 mg/daily, mirabegron generally performed similarly to tolterodine. Higher doses of mirabegron did not produce greater effects.

Pooled analyses and meta-analyses. Nitti Khullar et al.¹³⁷ (2013) reported findings from a pre-specified pooled efficacy and safety analysis using data from the USA/Canada Phase III trial,¹³¹ the Europe/Australia Phase III trial,¹³⁰ and the Europe/USA/Canada Phase III trial.¹³²  Mirabegron at doses of 50 and 100 mg once daily significantly reduced incontinence episodes per day (50 mg: -1.49 [95% CI -1.63 to -1.36]; 100 mg: -1.50 [95% CI -1.67 to -1.34]) and number of voids per day (50 mg: -1.75 [95% CI -1.89 to -1.61]; 100 mg: -1.74 [95% CI -1.91 to -1.56] compared to placebo incontinence episodes: -1.10 [95% CI -1.23 to -0.97]; voids per day: -1.20 [95% CI -1.34 to -1.06]) at 12 weeks of treatment. Significant improvements in mean voided volume/micturition, mean level of urgency, mean number of urgency episodes (grade 3 or 4)/day, mean number UUI episodes per day, and mean number nocturia episodes per day also were noted for both doses compared to placebo. The proportion of patients reporting zero incontinence episodes was significantly higher in the mirabegron groups (50 mg: 44.1%; 100 mg: 46.4%) compared to placebo (37.8%). Efficacy in patients who had used anti-muscarinics compared to treatment-naïve patients was similar across doses. The 25 mg dose significantly reduced frequency and incontinence episodes but generally not other endpoints. The 50 and 100 mg doses also significantly improved ratings on the Treatment Satisfaction-VAS compared to placebo. There was no dose-response gradient for the 50 mg dose compared to the 100 mg dose with both doses producing similar effects. Cui (2014)¹³⁸ reported findings from a meta-analysis as standardized mean differences (SMDs) between mirabegron groups collapsed across dose and placebo groups.  The SMDs were (statistically significant at p <0.05 unless otherwise noted): number of incontinence episodes per day: -0.44 (95% CI -0.86 to -0.12); voids per day: -0.62 (95% CI -0.89 to -0.25); urgency episodes per day: -0.60 (95% CI -0.85 to -0.36); nocturia episodes per day: -0.12 (95% CI -0.23 to -0.01); TS-VAS: 0.75 (95% CI 0.45 to 1.05; non-significant); OAB-q: -5.34 (95% CI -7.11 to -3.57); PPBC: -0.20 (95% CI -0.33 to -0.07). 

Additional useful information is provided by a report from the National Institute for Health and Care Excellence (NICE 2013) that evaluated the mirabegron evidence in preparation for making recommendations to United Kingdom practitioners. ¹³⁹ NICE performed a mixed treatment comparison analysis - a Bayesian statistical procedure that combines data across trials and compares treatments that were never tested head-to-head. NICE concluded that all medications, including mirabegron, have similar efficacy to reduce frequency. With regard to incontinence episodes, mirabegron and other medications also were similar with the exception of solifenacin (5 and 10 mg), which was statistically significantly more effective. However, NICE notes that the size of differences across medications was small. The Panel interpreted these findings to indicate that, in general, mirabegron has similar efficacy to the anti-muscarinics. 

The pooled safety analysis of the three Phase III trials¹³⁷ reported that overall rates of treatment-emergent adverse events (TEAEs) were similar across groups (Placebo: 47.7%; 25 mg mirabegron: 48.6%; 50 mg mirabegron: 47.1%; 100 mg mirabegron: 43.3%; tolterodine ER 4 mg: 46.7%) without evidence of a dose-response relationship across the mirabegron groups. The incidence of serious adverse events was similar across groups: placebo -2.1%; mirabegron 25 mg - 1.6%; mirabegron 50 mg - 2.1%; mirabegron 100 mg - 2.8%; and tolterodine ER 4 mg - 2.2%. The proportion of patients withdrawing from the studies for medication adverse events also was similar across groups: placebo - 3.3%; mirabegron 25 mg - 3.9%; mirabegron 50 mg - 3.9%; mirabegron 100 mg - 3.7%; and tolterodine ER 4 mg - 4.4%.

Detailed adverse events are reported in Nitti, Khullar¹³⁷ (2013) (see table below). Generally, rates of adverse events were similar across groups with the exception of dry mouth, which was higher for the tolterodine ER 4 mg group. Changes in blood pressure and pulse rate were minor and comparable across groups, although pulse rate increases were dose-dependent across the mirabegron groups (from 0.6 to 2.3 beats per minute with increasing dose). In comparison, the tolterodine groups exhibited increases of 1.0 to 2.1 beats per minute. Rates of tachycardia were less than 5% in each group and comparable to placebo rates.

The meta-analysis reported by Cui¹³⁸ (2014) indicated that rates of hypertension, cardiac arrhythmias, and urinary retention in mirabegron-treated patients were indistinguishable from rates in placebo groups. Chapple¹³³ (2013) reported adverse events during a 12-month trial of mirabegron (50 and 100 mg) compared to tolterodine ER 4 mg. Rates of adverse events were generally similar across groups with the exception of dry mouth, which was higher in the tolterodine group. Blood pressure changes were minimal (< 1 mm Hg) as were pulse rate changes (< 2 beats per minute). Below are selected adverse events from Chapple¹³³ (2013).

Additional useful information is provided by Malik (2012), ¹⁴⁰ which reports findings from a randomized, double-blind, placebo- and active-controlled study of mirabegron effects on cardiac repolarization in healthy volunteers. The active control was moxifloxacin. Continuous ECGs were recorded during crossover treatment with once-daily 50, 100, or 200 mg mirabegron or a single 400 mg dose of moxifloxacin (each participant exposed to only one daily dose). Mirabegron did not prolong the QT interval at the 50 mg and 100 mg doses but did result in prolongation at the 200 mg dose in females only (>10 msec from 30 min post-dose to 6 hours post-dose).

The MTC presented by NICE (NICE 2013) further indicates that the probability of dry mouth and constipation were statistically indistinguishable for mirabegron compared to placebo. ¹³⁹ All of the anti-muscarinics had a significantly higher probability of dry mouth compared to mirabegron 50 mg. Mirabegron was similar to most anti-muscarinics in terms of risk of constipation except for solifenacin (5 mg and 10 mg), fesoterodine (8 mg) and trospium (60 mg), all of which had a higher risk of constipation. 

Overall, the Panel interpreted the mirabegron data to indicate that mirabegron appears to be similar in efficacy to the anti-muscarinics and has lower rates of dry mouth than any of these medications. Mirabegron may produce lower rates of constipation than some of the anti-muscarinics. This lower incidence of bothersome adverse events may inform the selection of medications for patients who already present with dry mouth (e.g., secondary to Sjogren's syndrome) and/or constipation or for patients who experience efficacy from the anti-muscarinics but cannot tolerate these adverse events.

The body of evidence strength for the benefits and risks/burdens of mirabegron is Grade B. The available RCTs appear to have been well-conducted and evaluated large numbers of patients. However, follow-up in most studies is limited to 12 weeks, the magnitude of symptom reductions is relatively modest, and the patients in most trials cluster on the lower end of OAB symptomatology. These limitations result in uncertainty regarding long-term use of mirabegron in broader patient populations, particularly those with more severe OAB symptoms. In addition, there are no data on the use of mirabegron in patients with significant comorbidities, such as cognitive impairments, glaucoma, or uncontrolled hypertension. Clinicians should be cautious in management of these potentially vulnerable patient groups and engage in regular monitoring for adverse events. Further, there is no information regarding whether patients are more likely to adhere to mirabegron than to the anti-muscarinics. 

A meta-analysis of adverse events indicated that the ER formulations of oxybutynin and tolterodine resulted in statistically significantly fewer patient reports of dry mouth than the IR formulations of both medications. Specifically, the rate for the oxybutynin ER formulation was 40.0% (95% CI: 28.0% to 53.0%) and was statistically significantly lower than the oxybutynin IR rate of 69.0% (95% CI: 60.6% to 76.5%). The dry mouth rate for ER tolterodine was 18.0% (95% CI: 14.8% to 21.4%) and was statistically significantly lower (p<0.001) than the IR rate of 28.8% (95% CI: 25.1% to 32.8%). Within each medication, there was no relationship with dose. There were insufficient trospium trial arms to meta-analyze the IR versus ER formulations; however, a similar pattern was evident. The IR trospium trials reported dry mouth rates that ranged from 19.8% to 41.4% of patients;^{45, 82, 102, 108, 123, 141} in contrast, the ER trospium trials reported dry mouth rates of 8.7 to 12.9%.^{77, 111}

Because OAB is a chronic condition and treatment with anti-muscarinics generally would be required long-term, optimizing medication tolerability is critical to obtaining patient compliance. Adverse drug events, particularly dry mouth, are the major reasons that patients fail to comply with anti-muscarinic therapy; thus choosing the formulation with the lowest likelihood of adverse events may improve compliance.^{142, 143} In addition, compliance with a once-daily treatment has been shown to be greater than with medications that are taken more than once a day.¹³³ The decision to prescribe an IR versus an ER formulation, however, should be made in the context of the patient's prior experience with anti-muscarinics and the availability of medications, including insurer constraints, in order to minimize patient burden. Insurer constraints may be such that a patient may need to be prescribed an IR formulation and either have inadequate symptom control or have intolerable side effects prior to obtaining approval to be prescribed an ER formulation. The Panel notes that if a patient has good symptom control and tolerable side effects on an IR formulation, then there is no need to change to an ER formulation. 

Transdermal preparations of oxybutynin may be offered instead of oral anti-muscarinics to patients who are at risk of or who have experienced dry mouth with oral agents. Six randomized trials evaluated transdermal oxybutynin preparations. Four trials that included placebo control groups evaluated the TDS patch.^{76, 87, 145, 147} One trial with a placebo control group evaluated oxybutynin chloride topical gel.¹⁴⁶ An additional trial compared the oxybutynin TDS patch to oral oxybutynin.⁷⁴ Dmochowski (2002) evaluated three oxybutynin doses administered via TDS patch (1.3 mg, 2.6 mg and 3.9 mg) and reported reductions in incontinence episodes per day (reductions of 2.8 episodes with placebo, 2.7 episodes with 1.3 mg, 2.6 episodes at 2.6 mg and 3.3 episodes at 3.9 mg) and reductions in 24 hour frequency (reductions of 1.7 episodes with placebo, 1.8 with 1.3 mg, 1.8 with 2.6 mg and 2.3 with 3.9 mg); only the reductions with 3.9 mg were significantly different from placebo.¹⁴⁵ A 12-week open-label extension of this study reported larger incontinence episode reductions, ranging from 3.4 to 3.9 episodes. Dmochowski (2003) evaluated only the 3.9 mg dose compared to r mg tolterodine ER and placebo in known responders to anti-muscarinics and reported similar findings for frequency for both the TDS and oral medication groups.⁷⁶ In addition, in this trial, urgency incontinence episodes were reduced by 2.8 episodes a day using 3.9 mg compared to 3.2 episodes a day with tolterodine and 2.1 episodes a day with placebo. Cartwright (2010) also evaluated 3.9 mg compared to placebo and reported significant reductions in urgency episodes/day (reduction of 1.23 episodes in the 3.9 mg group compared to a reduction of 0.21 episodes in the placebo group) and non-significant reductions in frequency and UI episodes/day.¹⁴⁷ The primary outcome in this trial was patient-selected goals for treatment. A greater proportion of patients in the 3.9 mg group reported goal achievement (41.9%) than in the placebo group (32.2%), but the difference was not significant. The authors note that the relatively low proportions of patients who reported achieving treatment goals may indicate why many patients discontinue anti-muscarinic treatment. Homma and Koyama (2006) compared 2.6 mg, 3.9 mg and 5.2 mg TDS oxybutynin to placebo and reported reductions in incontinence episodes of 1.5 episodes with 2.6 mg, 2.0 episodes with 3.9 mg, 1.6 episodes with 5.2 mg and 1.4 episodes with placebo.⁸⁷ Baseline incontinence levels were lower in this trial (average of 3 episodes/day) than in the Dmochowski (2002) trial (average of 4 episodes/day), which might account for the smaller magnitude of change. In patients known to be responsive to oral oxybutynin, Davila (2001) used 1.3 mg to 3.9 mg TDS oxybutynin or 5 mg to 22.5 mg oral oxybutynin, depending on the patient's prior tolerance for oxybutynin.⁷⁴ Reduction of approximately 4.8 incontinence episodes/day occurred in both groups. Staskin (2009) evaluated 1 mg oxybutynin chloride topical gel and reported significant decreases in urgency incontinence (3 fewer episodes/day) and frequency (2.7 fewer episodes/day) compared to the placebo group (2.5 fewer UI episodes/day and 2 fewer frequency episodes/day).¹⁴⁶ Newman (2010) reported on the same patients and noted that treatment with gel improved health-related QOL measures more than did treatment with placebo.¹⁴⁸

Five trials reported adverse events. Dmochowski (2002) reported dry mouth rates of 8.3% with placebo, 4.6% with 1.3 mg TDS oxybutynin, 6.8% with 2.6 mg and 9.6% with 3.9 mg and constipation rates of 3.0% with placebo, 5.4% with 1.3 mg, 2.3% with 2.6 mg and 0.8% with 3.9 mg.¹⁴⁵ Dmochowski (2003) reported dry mouth rates of 4.1% with 3.9 mg TDS oxybutynin, 7.3% with tolterodine and 1.7% with placebo; constipation rates were 3.3% with 3.9 mg and 5.7% with tolterodine (constipation rate not reported for placebo group).⁷⁶ Davila (2001) reported dry mouth rates of 38% in the TDS group compared to 94% in the oral oxybutynin group (constipation rates not reported).⁷⁴ Cartwright (2010) reported that 38.2% of patients in the active treatment group experienced erythema or pruritus (compared to 27.1% in the placebo group) and 14.9% experienced at least one systemic adverse event, the most common of which was dry mouth (compared to 12.5% in the placebo group).¹⁴⁷ Staskin (2009) reported dry mouth rates of 6.9% in the oxybutynin gel group compared to 2.8% in the placebo group and rates of other adverse events at 1% or less in both groups.¹⁴⁶

The Panel interpreted these data to indicate that transdermal oxybutynin (patch and gel) is effective in reducing incontinence episodes, in particular, with dry mouth rates that appear to be less than the meta-analyzed rates of 40.0% for oral oxybutynin ER and 68.0% for oral oxybutynin IR. Because the number of studies evaluating TDS oxybutynin was relatively few with different patient inclusion criteria (i.e., known responders to anti-muscarinic medications in some trials), the body of evidence strength was designated as Grade C.

In the Panel's experience, patients who experience inadequate symptom control and/or unacceptable adverse drug events with one anti-muscarinic medication may experience better symptom control and/or a more acceptable adverse drug event profile with another anti-muscarinic. In addition, in some patients, dose modification (i.e., reducing dose or reducing dose and combining medication with behavioral techniques) may achieve a better balance between efficacy and adverse drug events. A small literature composed of observational studies supports this experience, particularly when switching from an older medication to a newer medication. Patients who had prior unsatisfactory symptom control and/or unacceptable adverse events with tolterodine^149-151 or oxybutynin^{151, 152} reported better efficacy and/or more acceptable adverse event profiles with fesoterodine,¹⁴⁹ solifenacin^{150, 152} or darifenacin.¹⁵¹ Based on the Panel's clinical experience and this limited literature, the Panel advises that clinicians should not abandon anti-muscarinic therapy if trial of one medication appears to fail or produces an unacceptable adverse event profile. Further, clinicians may also switch patients to a β₃-adrenoceptor agonist (e.g., mirabegron) given an efficacy profile that appears similar to the anti-muscarinics and a relatively lower adverse event profile.

The 2019 update literature review uncovered a number of studies looking at combination therapy for the treatment of OAB. Typical pharmacotherapy options for the management of OAB include both anti-muscarinics and β₃-adrenoceptor agonists. While these therapies have different mechanisms of action, co-administration appears to have no noticeable effects on pharmacokinetics.¹⁵³ Further, studies have demonstrated improved efficacy with combination therapy without any significant effect on the safety profile when compared to monotherapy. While the strongest evidence for combination therapy comes from the SYNERGY I/II^154,155 and BESIDE¹⁵⁶ trials, which utilized solifenacin (5 mg) and mirabegron (25 or 50 mg), additional combinations utilizing various other drug classes and dosing schedules have been tested and are discussed below. At this time, the Panel feels that such data are less robust; as such, more data will be required before such additional combinations can be recommended.

In the 18-week SYNERGY trial, Herschorn et al. (2017)¹⁵⁴ randomized 3,398 wet OAB patients (77% women) to solifenacin (5 mg) plus mirabegron (25 mg), solifenacin (5 mg) plus mirabegron (50 mg), placebo, mirabegron (25 mg) monotherapy, mirabegron (50 mg) monotherapy, or solifenacin (5 mg) monotherapy in a 2:2:1:1:1:1 ratio. While the solifenacin/mirabegron 50 mg group was superior to solifenacin monotherapy for mean adjusted difference in UI episodes per 24 hours (-0.20, 95% CI -0.44 to 0.04, P= 0.033), statistical superiority was not seen versus mirabegron 50 mg monotherapy (-0.23, 95% CI -0.47 to 0.01, P= 0.052). In secondary analyses, all active treatment groups showed greater improvements in UI episodes per 24 hours versus placebo (nominal P values all <0.05). Effect sizes for the combined therapy groups (solifenacin/mirabegron 25 mg: -0.70; solifenacin/mirabegron 50 mg: -0.65) were higher than those obtained with monotherapy (range -0.37 for mirabegron 25 mg to -0.45 for solifenacin 5 mg). All active treatment groups had greater improvements in the mean numbers of micturitions per 24 hours versus placebo, with effect sizes for the combined therapy groups (solifenacin/mirabegron 25 mg: -0.85; solifenacin/mirabegron 50 mg: -0.95) higher than with mirabegron monotherapy (25 mg: -0.36; 50 mg: -0.39) and solifenacin 5 mg (-0.56). Researchers noted that the combined solifenacin/mirabegron 50 mg group was statistically significantly superior to both monotherapies at end of treatment for UUI episodes, urgency episodes and nocturia, with effect sizes that appeared to be additive. Adverse events including dry mouth, constipation, and dyspepsia, were slightly increased in the combination therapy groups compared to monotherapies. Additionally, events indicative of urinary retention were reported more frequently in the combination groups compared to monotherapy and placebo.

In the follow up SYNERGY II trial, Gratzke et al. (2018)¹⁵⁵ evaluated the safety and efficacy of combination therapy over 12 months. The trial randomized 1,829 wet OAB patients (80% women) to combination 5 mg solifenacin plus 50 mg mirabegron, solifenacin monotherapy, mirabegron monotherapy, or placebo. The primary objective was safety as measured by TEAEs; additionally, efficacy was measured as the change from baseline to the end of treatment in the mean number of incontinence episodes and micturitions per 24 hours. Adverse events were reported more frequently in the combination group (49%) than in the mirabegron group (41%) or solifenacin group (44%), with dry mouth being the most commonly reported TEAE. Combination therapy was statistically superior to mirabegron and solifenacin monotherapy for decreasing number of incontinence episodes (mirabegron, p<0.001; solifenacin, p=0.002) and micturitions (mirabegron, p<0.001; solifenacin, p=0.004). 

Drake et al. (2016)¹⁵⁶ in the BESIDE trial evaluated the efficacy, safety, and tolerability of combination therapy (solifenacin 5 mg plus mirabegron 50 mg) versus monotherapy (solifenacin 5 or 10 mg) in a 1:1:1 randomized trial for OAB patients remaining incontinent after 4 weeks of solifenacin 5 mg. A total of 2,174 patients were randomized (83% women). At end of treatment, combination therapy was found to be superior to solifenacin 5 mg, with significant improvements in daily incontinence (p = 0.001), daily micturitions (p < 0.001), and incontinence noted in a 3-day diary (p = 0.014). Combination was noninferior to solifenacin 10 mg for key secondary end points, including a change from baseline to end of treatment in the mean number of micturitions per 24 hours and number of incontinence episodes noted in a 3-day diary at end of treatment. Combination was superior to solifenacin 10 mg for improving daily micturitions. 

One RCT by Kosilov et al. (2018)¹⁵⁷ and one prospective cohort study by Kosilov et al. (2013)¹⁵⁸ evaluated combination solifenacin and trospium in women with OAB. The Panel notes that the doses of solifenacin and trospium utilized in the Kosilov et al. studies are higher than the standard recommended doses. The RCT randomized 312 women (60-83 years of age) to high dose solifenacin (20 mg/day) plus high dose trospium (60 mg/day), or a lower dose of solifenacin (10 mg/day) plus a lower dose trospium (30 mg/day), or placebo.¹⁵⁷ Following treatment, QOL parameters in terms of the functional state of the lower urinary tract were improved compared with placebo, but with no differences between the higher dose and lower dose combination groups (p<0.05). The prospective cohort study was designed in two phases and included 229 women (65-77 years of age).¹⁵⁸ In the first phase all women received 60 mg trospium and 40 mg solifenacin daily, while in the second phase, defined as the maintenance phase, women were allocated to additional trospium plus solifenacin treatment, electrical stimulation of detrusor muscle, laser puncture using a helium-neon laser applied to acupuncture points, or placebo for one month. The study reported that maintenance with the combination therapy resulted in effective maintenance of clinical and urodynamic results for at least seven months, while electrical stimulation was less effective, and laser puncture was inefficient. 

Another RCT by Ellington et al. (2016) evaluated combination tolterodine and intravaginal estradiol cream in 58 menopausal women.¹⁵⁹  Women were randomized to either oral tolterodine or estradiol cream for 12 weeks and then offered addition of the alternative therapy with follow-up at week 24 and week 52.  Both monotherapies resulted in decreased symptom bother scores at 12 weeks when compared with baseline values. Following addition of the alternate therapy at the 12 week time-point, within group OAB symptom severity and HRQL scores remained improved significantly at 24 and 52 weeks from baseline, though it should be noted that only addition of tolterodine at 12 weeks to the estradiol group resulted in further significant improvement in the OAB-q symptom bother score and the HRQL scores above the single therapy effect noted at 12 weeks. 

Finally, an RCT by Rovner et al. (2018)¹⁶⁰ compared a 3 month combination of oral desmopressin 25 µg plus 4 mg tolterodine with 4 mg tolterodine monotherapy in 106 women.  When evaluating the overall population, combination therapy resulted in a non-significant reduction in nocturnal voids over monotherapy (p=0.112).  In a subgroup of patients with nocturnal polyuria, combination therapy resulted in improved nocturnal void volume (p=0.034) and time to first nocturnal void (p=0.045).

Clinicians should not use anti-muscarinics in patients with narrow angle glaucoma unless the treating ophthalmologist approves and should use anti-muscarinics with extreme caution in patients with impaired gastric emptying or a history of urinary retention, carefully weighing the benefits versus the significant risks. If the patient is at risk for or has a history of gastric emptying problems, then the patient should be seen by or receive clearance from a gastroenterologist. If the patient has a history of or is at risk of urinary retention, then urology consultation should be strongly considered. It is useful to obtain a PVR in any patient the clinician suspects has a higher than normal risk of urinary retention. Anti-muscarinics are also contraindicated in patients using solid oral forms of potassium chloride, as the reduced gastric emptying potentially caused by the anti-muscarinics may increase the potassium absorption of these agents. If these patients can be switched to alternative forms of potassium chloride, then anti-muscarinic therapy may be possible with caution. In weighing the risks of anti-muscarinic therapy in high-risk patients, it is important to remember that OAB may compromise QOL, but it is not a life-threatening condition. Clinicians, therefore, should exercise extreme caution in using treatments that may present life-threatening risks.

One of the main limitations of anti-muscarinic therapy is that the majority of patients discontinue after a few weeks or months.^{142, 144} Although there may be several factors involved in this decision, side effects are commonly cited as the reason for discontinuation.¹⁴³ One way clinicians can help patients benefit from anti-muscarinic therapy is to proactively monitor for and manage common side-effects. Even before initiating anti-muscarinic therapy, patients should be educated about the possible effects of medication on bowel function and the roles of adequate dietary fiber and fluid, psyllium-based fiber supplements, regular exercise and normal bowel habits. Preparing for dry mouth might include advice on oral lubricants, avoiding mouthwashes with alcohol, taking small sips of water, sucking on sugar-free hard candies and chewing sugar-free gum. When dosing options are available, dose reduction can provide relief from side-effects while retaining some therapeutic effects. In older patients who may metabolize drugs differently, it is often advisable to start with a minimal dose and then increase it if it is tolerated well. With multiple drugs available for OAB, trying alternate anti-muscarinics may identify a medication that the patient can more easily tolerate.

The concurrent use of other medications with anti-cholinergic activity may potentiate the side effects of the anti-muscarinic class of OAB medications. These medications include tricyclic antidepressants, those used in the treatment of Parkinsonism and other extra-pyramidal diseases and of Alzheimer's disease, and include benzotropine, biperiden HCl, galantamine, rivastigmine and trihexyphenidyl HCl. Certain anti-nausea medications and those with atropine-like properties, such as trimethaphan, methscopolamine bromide and ipratropium, may also potentiate these side effects. Providers also should exercise caution in patients who are prescribed acetycholinesterase inhibitors such as donepezil. This list is not intended to be exhaustive; prescribers should be aware of precautions and contraindications for these medications. 

In addition, most clinical studies of OAB medications have been conducted on relatively narrow patient populations and provide only short-term data (i.e., 12 weeks) on adverse drug events. In the absence of long-term data on patients neither eligible for nor included in clinical trials, the prevalence and severity of adverse drug events is largely unknown. 

In frail patients, defined as patients with mobility deficits (i.e., require support to walk, have slow gait speed, have difficulty rising from sitting to standing without assistance), weight loss and weakness without medical cause and who may have cognitive deficits¹⁶¹ the use of OAB medications may have a lower therapeutic index and a higher adverse drug event profile. OAB medication studies generally are not conducted in the frail elderly, resulting in a lack of data in this group. In the Panel's experience, however, adverse drug events in addition to the typically reported events of dry mouth and constipation may occur, including impaired thermoregulation that can cause dangerous core temperature elevation. Clinicians should begin with the lowest possible dose and increase doses slowly while carefully assessing for the balance between symptom control and adverse events. The use of transdermal anti-muscarinics should be monitored to ensure that the skin where the medication is applied remains intact. 

Cognitive deficits, particularly memory difficulties, have been reported in response to anti-muscarinics,^{132-152,157,161-164} and clinical experience suggests that elderly patients may be particularly prone to these adverse effects. There is some suggestion that the newer agents (e.g., darifenacin) are less likely to produce cognitive deficits in elderly patients than are the older agents, but the literature is limited and the two-week drug administration period in these studies is not long enough to yield definitive conclusions.^{165, 166} Kay (2006) notes, however, that patients may not recognize that memory deterioration has occurred, making it essential for the clinician, family members and caregivers to monitor for these effects.¹⁶⁶ In addition, polypharmacy is common in community dwelling patients who are frail,¹⁶⁷ placing them at higher risk for adverse drug events, including impaired cognition. In dementia patients, anti-muscarinics should be used with extreme caution or may be contraindicated entirely depending on the level of cognitive impairment. The Panel further notes that presently there are no data on the use of β₃-adrenoceptor agonists (e.g., mirabegron) in the frail patient, the patient with significant comorbidities, or the patient on multiple medications. The clinician should consider these possibilities in prescribing anti-muscarinics or β₃-adrenoceptor agonists to frail patients and reassess the balance between benefits and risks/burdens with the patient, caregiver and/or family on a regular basis and/or when functioning appears to change. In patients who cannot tolerate anti-muscarinics or for whom pharmacologic management is not appropriate, behavioral strategies that include prompted voiding and fluid management may be helpful. 

The Panel defines the refractory patient as the patient who has failed a trial of symptom-appropriate behavioral therapy of sufficient length to evaluate potential efficacy and who has failed a trial of at least one anti-muscarinic medication administered for 4 to 8 weeks. Failure of an anti-muscarinic medication may include lack of efficacy and/or inability to tolerate adverse drug effects. The Panel notes that this definition is a minimum definition; individual clinicians and patients may decide that it is in the best interests of the patient to persevere with behavioral and/or pharmacologic therapy for longer periods, to combine behavioral and pharmacologic therapies to achieve better efficacy, or to try alternate medications before judging that a patient is refractory.

Behavioral therapies present no risks to patients and medications present risks that cease when the medication is stopped. The remaining treatment levels present increasing risks to patients that must be balanced with potential efficacy. Before a patient is exposed to these therapies, a comprehensive evaluation should be conducted to ensure that the patient's symptoms are attributable to OAB and not to some other disease process that requires other kinds of treatment and the patient's desire for further treatment should be ascertained. 

The Panel designated intradetrusor onabotulinumtoxinA treatment as a Standard in the thoroughly-educated and carefully-counseled patient with moderate to severe OAB symptoms because a body of moderate-quality evidence indicated sustained improvements in voiding and QOL outcomes and rates of adverse events that could compromise QOL or lead to serious illness were less likely to occur with use of the FDA-approved dose of 100U.  The available literature on intravesical use of onoabotlinumtoxinA is reviewed below; the Panel focused its deliberations on injections into the detrusor specifically because most studies assessed this injection location. There is insufficient evidence at present to comment on the relative efficacy of injections into other intravesical locations. 

The original literature searches retrieved four randomized trials with placebo control groups (reported in five papers), two randomized trials without placebo control groups and 15 observational studies without control groups that evaluated the effects of onabotulinumtoxinA in patients with non-neurogenic OAB who had inadequate symptom control with anti-muscarinics or intolerable side effects.^168-189 All studies reviewed evaluated onabotulinumtoxinA (BoTox R, Allergan, Inc., Irvine, CA) except for one¹⁷⁵ which evaluated abobotulinumtoxinA (Dysport TM, Medicis Pharmaceutical Co., Scottsdale, AZ). Studies varied in onabotulinumtoxinA dose and in injection location. Doses of onabotulinumtoxinA are not equivalent to doses of abobotulinumtoxinA.

The four RCTs with placebo control groups evaluated injections into the detrusor of 200 U onabotulinumtoxinA,^168,169 200-300 U onabotulinumtoxinA,¹⁷⁰ or 50-300 U onabotulinumtoxinA.^{169, 184} The randomized trials without placebo control groups injected 100 U onabotulinumtoxinA into the suburothelial space, the detrusor or the bladder base¹⁷³ or injected 100 U onabotulinumtoxinA into the bladder body, bladder body and trigone or bladder base and trigone.¹⁷⁴ Significant reductions in incontinence episodes^{168, 170, 174} and in urgency^{171, 173, 174} were reported in active treatment groups (but not in placebo controls where included). Frequency data were less clear with reductions occurring in most active treatment groups (except for patients injected in the suburothelial space in Kuo 2007 who experienced increased frequency) but with a large range of reductions (e.g., from a non-significant 1.3 episodes per day in Flynn 2009, to 6.1 episodes per day in Sahai 2007, to 14.2 episodes per day in Kuo 2007 in the detrusor injected group).^{170, 171, 173} To some extent this range may be related to the inclusion of patients with different baseline frequency levels (e.g., patients in Flynn 2009 had baseline 24-hour frequency of 10.5 episodes per day compared to patients in Kuo 2007 who had 24-hour frequencies ranging from 17.8 to 29.8 episodes per day),^{170, 173} but the largest reductions were reported in Kuo (2007),¹⁷³ which is the trial with the lowest onabotulinumtoxinA dose (100 U). Flynn (2009) also reported reductions in nocturia (0.5 fewer episodes/night) and in pad use (2.2 fewer pads/day) at six weeks post-injection.¹⁷⁰ In addition, Sahai (2007) reported improvements in a variety of urodynamic parameters and improved scores on the IIQ-7 and the UDI-6 in onabotulinumtoxinA patients but not in placebo patients.¹⁷¹ Two additional papers reporting on the same group of patients noted improved scores in onabotulinumtoxinA patients on the King's Health Questionnaire during the randomized trial as well as during an open-label extension study¹⁹⁰ and that QOL improvements and improved urodynamic parameters were restored in patients who required repeat injections.¹⁹¹ In Kuo (2007), three months after injection of 100 U onabotulinumtoxinA, the proportion of patients reporting a status of excellent or moderately improved was 93% of the detrusor group, 80% of the suburothelial space group and 67% of the bladder base group.¹⁷³ These proportions dropped to 67% (detrusor), 47% (suburothelial space) and 13% (bladder base) at 6 months and to 20% (detrusor), 20% (suburothelial space) and 6.7% (bladder base) at 9 months. Dmochowski (2010) compared responses across a wide range of onabotulinumtoxinA doses (50 to 300 U) and reported that doses of 100 U or greater were sufficient to reduce urgency incontinence episodes and improve QOL measures but without clear dose-response effects such that doses above 150 U did not contribute additional clinically-relevant symptom improvement.¹⁶⁹ Additional information on the same patients was provided by Rovner (2011).¹⁸⁴ This paper reported that doses of ≥ 100 U all resulted in significant improvement of OAB symptoms (i.e., reductions in UI episodes and frequency) without clear dose-response effects. Findings also were broken out between patients with and without DO; similar improvements were reported in both groups. 

The observational studies injected doses of onabotulinumtoxinA ranging from 100 - 300 U. Most studies injected into the detrusor except for two studies, which injected into the detrusor and sphincter^176,186 and two studies,^{179, 180} which injected into the submucosa of the bladder wall. Jeffery (2007) injected 500 U of abobotulinumtoxinA into the detrusor.¹⁷⁵ As a group, the observational studies reported reductions in frequency, nocturia, pad use and incontinence; improvement in urodynamics parameters and improvement in QOL measures. Follow-up durations ranged from 1.5 weeks to 145 weeks. In the longer studies, improvements diminished over time and repeat injections were required to restore improvements. Gamé (2010) reported that in patients who had up to five repeat injections to maintain improvement, QOL improved after each injection.¹⁹²

The 2014 update literature review retrieved 27 new studies, including five randomized trials with placebo control groups,^{131,193–196} two randomized trials with active control groups (Kuo 2011 - compared 100U in three different injection sites; AlTaweel 2011 - compared 100U to 200U in the detrusor),¹⁹⁷ and 20 observational studies.^198-217 Patient selection criteria were similar to those in the prior evidence (e.g., patients with moderate to severe baseline levels of UUI, incontinence, frequency, and urgency) and, in aggregate, the new randomized trials reported on responses of 887 patients treated with BTX-A - more than double the number of patients in the active treatment groups from the prior randomized trials. The lack of long-term follow-up remains, with the largest trials reporting outcomes at 12 weeks. A few trials reported additional outcomes at longer intervals (e.g., AlTaweel 2011 - nine months; Denys 2012 - 6 months);^194,197 the general pattern is of diminishing effectiveness. With the exception of Kuo (2011) who compared injection sites, most studies injected into the detrusor.²⁰⁸ In contrast to prior evidence, the most commonly used dose was 100U rather than 200U. 

In general, most trials reported statistically significant improvements in measured voiding outcomes (UUI, incontinence, frequency, urgency, nocturia, pad use) and in QOL outcomes compared to placebo groups. Fowler (2012)²¹⁸ reported QOL data from the Dmochowski (2010)¹⁶⁹ dose-finding trial and noted that the I-QOL and KHQ exhibited significant improvement compared to placebo for all groups administered 100U or greater. Studies that measured urodynamics parameters also reported improvements (e.g., in maximum bladder capacity). Denys (2012)¹⁹⁴ compared placebo to 50U, 100U, and 150U; at three months post-procedure, >50% improvement in urgency and UUI was reported by 30% of placebo patients, 37% of the 50U patients, 68% of 100U patients, and 58% of 150U patients (sample sizes were <30 in each group; only the 100U group was statistically significantly different from placebo). The 100U and 150U groups exhibited significantly reduced frequency compared to placebo and this reduction persisted for 30 days in the 100U group and was still significant at 60 months for the 150U group. The number of patients who achieved complete continence at three months was significantly greater in the 100U group (55%) and the 150U group (50%) compared to the placebo group (11%). At five months post-treatment, these differences were maintained. Kuo (2011) compared injection of 100U into the bladder body, bladder body plus the trigone, and the bladder base plus the trigone and found no differences across injection sites in success rates (70 to 74% of patients in each group) or in reductions in urgency or UUI episodes.²⁰⁸

These outcomes occurred, however, in the context of high rates of adverse events in the active treatment groups in some studies. Rates of UTIs ranged from 3.6% to 54.5% with four of the RCTS reporting rates of >40.0% and Dmochowski (2010) reporting that rates generally increased with dose with rates ranging from 33.9% to 48.1% across active treatment groups.¹⁶⁹ The definition of elevated PVR varied across studies from 100 mL to 400 mL with most studies defining an elevated PRV as 100 - 150 mL. It should be noted, however, that that the highest rates of urinary retention were not necessarily reported in studies that used the lowest PVR thresholds. Rates of urinary retention were reported in 10 studies and ranged from 0% to 43% with rates of 43.0% and 30.0% reported in one RCT (elevated PVR defined as 200 cc)¹⁶⁸ and one observational study (elevated PVR defined as 250 cc),¹⁷² respectively. Rates of PVR increase were reported in 14  studies and ranged from 0% to 75% with half of these studies reporting rates of 43.0% or higher (Kuo 2011);^{168,172,174,175,179,182,208} The proportion of patients who needed to perform self-catheterization was reported in 20 studies and ranged from 0% to 43% with six studies reporting rates higher than 20.0%.^{168, 169, 171, 175-178} Increased PVRs and the need for self-catheterization persisted for six to nine months in some patients.^{172, 175, 185} It should be noted, however, that Kessler (2009) examined QOL outcomes in women who had to perform self-catheterization post-onabotulinumtoxinA treatment compared to those who did not and found no differences in UDI-6 and IIQ-7 scores.¹⁷⁷ Bauer (2011) focused more broadly on side effects and interviewed patients (n = 56) who had been administered onabotulinumtoxinA (100, 150 or 200 U) or abobotulinumtoxinA (500 U) regarding the occurrence of gross hematuria, dry mouth, dysphagia, speech problems, impaired vision and weakness of the eyelids, arms, legs, torso and/or whole body.²¹⁹ Approximately 54% of patients reported at least one side effect, including urinary retention (8.9%), gross hematuria (17.9%), UTI (7.1%), dry mouth (19.6%), dysphagia (5.4%), impaired vision (5.4%), eyelid weakness (8.9%), arm weakness (8.9%), leg weakness (7.1%) and torso weakness (5.4%). The authors note that symptoms other than urinary retention and UTI were transient and resolved without the need for further treatment. These data indicate, however, that patients may experience neurological adverse events in addition to the more commonly reported events of urinary retention and UTIs.

Additional useful information regarding efficacy and adverse events is provided by the large group of observational studies. 

Diabetic patients. Wang (2013) reported on 48 type II diabetes mellitus (DM) patients compared to 48 non-diabetic age-matched control patients. These patients were older, on average, than in the randomized trials (DM patients mean age 73 years; non-DM patients mean age 72 y).²¹⁷ All patients had UDS-verified DO. Patients received 100U BTX-A in the suburothelial space and were followed at three and six months. At three months, frequency, UUI episodes, and PPBC scores had improved similarly in both groups; urgency episodes were reduced in both groups but not significantly. Success rates (defined as PPBC score improvement of two or more points) were statistically similar at 56% for DM patients and 61% for non-DM patients and remained at 6 months. DO disappeared in 56.3% of DM patients and 47.8% of non-DM patients. After the six months point, success rates began to decrease. However, the DM patients had significantly higher rates of large PVR volumes (60.4%) and general weakness (10.4%) compared to the non-DM patients (large PVR - 33.3%; general weakness - 0%). 

BTX-A in the frail elderly patient. Liao & Kuo (2013) reported on 166 patients, including 61 frail elderly (mean age 75.8 years), 63 elderly without frailty (mean age 75.7 years), and 42 younger patients (mean age 44.6 years).²⁰⁶ Frailty was defined as age >65 years and three or more of the following: unintentional weight loss, self-reported exhaustion, weakness, slow walking speed, low physical activity. Frail patients averaged 1.6 comorbidities, and 5.7 medications per day. All patients had idiopathic DO refractory to medications and persistent UUI/urgency. Patients received 100U of BTX-A injected into the suburothelial space. At three months, significant improvement had occurred in UUI and PPBC scores for all three groups and CBC and PVR also had increased significantly for all three groups. Success rates (defined as improvement of two or more points on the PPBC) were similar at three months (frail elderly - 83.4%; elderly - 91.2%; younger - 88.9%) and at six months across groups (frail elderly - 44.9%; elderly - 52.1%; younger - 49.4%). By 12 months post-injection, however, success rates among frail elderly were significantly lower at 6.82% then rates for elderly (22.3%) and younger patients (23.1%) and cumulative success rates (Kaplan-Meier analysis) also were significantly lower at follow-up to 15 months post-injection. Rates of acute urinary retention (AUR), straining to void, and hematuria were similar across groups. Frail elderly, however, were more likely to have a PVR >150 mL (60.7%) compared to elderly (39.7%) and younger patients (35.7%). Although the difference in AUR rates was not significant, the recovery period to spontaneous voiding was longer in frail elderly (median 3.5 months) compared to one and 0.5 months for elderly and younger patients respectively. Frail elderly also were more likely to report generalized weakness (6.6%) compared to elderly (0%) and younger patients (0%). Interestingly, the highest rates of UTI occurred among younger patients (28.6%) with lower rates among frail elderly (13.1%) and elderly (9.5%). 

Long-term follow-up: Kuo (2011), Chen (2012), and Ke (2012) reported on the same group of 174 patients with idiopathic DO and UUI who received 100U BTX-A using various injection sites (e.g., detrusor or suburothelial injection in bladder body and/or bladder base including the trigone). Improvements of at least 2 points on the PPBC were designated as successes.^208,210,211 At three months, success rates were similar for males (79%) and females (80%), for patients aged 75 or older (80%) and younger patients (79%), and similar across different injection sites. Patients were monitored for adverse events for up to 24 months. Adverse events included: AUR in 6.9%; PVR >150 mL in 46.6%; straining to void in 42%; gross hematuria in 9.8%; UTI in 15.5%; and general weakness in 3.4%.

Okamura (2013) reported on 17 patients (mean age 67 years) who received one injection of BTX-A 100U submucosally and were followed every month for one year.²¹² Urgency episodes, UUI, and daytime frequency were significantly decreased up to 11 months post-injection. Measures of QOL were significantly improved for eight to ten months post-injection. Scores on a GRA measure demonstrated improvement above baseline for eight months. Median time to failure was 11 months but women had longer intervals to failure (11.7 months) compared to men (7.2 months). Two patients had a PVR >200 mL that resolved without the need for clean intermittent self-catheterization (CISC).

Granese (2012) reported on 68 female patients (mean age 56 years) who underwent an injection of BTX-A 100U into the detrusor and were followed for 12 months.²¹⁴ At baseline, patients had moderately severe symptoms (9.5 urgency episodes/day; 15.1 voids/day; 2.5 nocturia episodes/day; 5.7 incontinence episodes/day). Symptoms improved markedly up to six months and then began to decay although at 12 months symptom levels had not yet returned to baseline. The same pattern was evident with some urodynamic parameters (e.g., bladder compliance, detrusor contraction). QOL measures also exhibited long-term improvement. A subset of 20 patients elected to have a second injection at 12 months and improvements in most measured parameters again occurred (follow-up to 3 months). PVR > 100 mL with lower urinary tract symptoms was judged as indication for CISC. At one month post-injection, 35% of patients after the first injection and 40% of patients after the second injection required CISC. At two months post-injection, 22% of patients after the first injection and 25% of patients after the second injection required CISC.  At three months post-injection, 3% of patients after the first injection and 1% of patients after the second injection required CISC. UTI rates were 6% (after first injection) and 5% (after second injection) but resolved after month one. 

Repeated injections: injection intervals, discontinuation rates, and discontinuation reasons. Mohee (2012) reported on 137 patients (mean age 57.3 years; all patients had DO with a minority having neurogenic DO) with at least 36 months follow-up.²¹⁵ At the beginning of the study, patients with idiopathic DO were administered 200U of BTX-A (in the detrusor); later the dose was reduced to 100U as published literature emerged indicating lower doses had equivalent efficacy. Mean re-treatment interval was 8 months. Of 104 patients with idiopathic DO, 66% (n=69) stopped treatment with BTX-A. The only difference between patients who continued and those who discontinued was the presence of incontinence at baseline (patients more likely to discontinue) and age (patients aged <50 years more likely to discontinue). Reasons for discontinuation were evaluated for the group as a whole (104 with idiopathic DO; 33 with neurogenic DO); these were: primary failure -- patients who had a single injection without benefit -16.8%; secondary failure -- patients who had benefit from the initial injection but not after two or more - 11.7%; tolerability -- patients who stopped BTX-A for reasons other than efficacy (e.g., having to perform CISC, urinary retention, recurrent UTIs) - 55.9% with an additional 16 patients who had primary and secondary failure also citing tolerability issues as a reason for discontinuation. Tolerability issues consisted of worsening symptoms (4.8%), need for CISC (49.2%), recurrent UTIs (36.5%), and unable to perform CISC (9.5%).

Veeratterapillay (2014) reported on 125 patients (median age 53 years, all with DO including a minority of patients with neurogenic DO) who had two or more BTX-A 200U injections (2 injections - 125 patients, 3 injections - 60 patients, 4 injections - 28 patients, 5 injections - 14 patients, 6 injections - 3 patients, 7 injections - 3 patients, 8 injections - 2 patients).²¹⁶ The overall median interval between injections was 14.4 months. Mean intervals between injections were: 17.6 months between injections 1 and 2; 15.7 months between injection 2 and 3; 15.4 months between injections 3 and 4; and, 11.6 months between the 4^th and all subsequent injections. Of the 125 patients, 17% did not respond to repeated injections and opted for other treatments (e.g., long-term catheterization, SNS, surgery). Approximately 26% of patients developed PVR ≥ 150mL with 91% of these episodes occurring with the first two injections and all occurring within the first three injections. Approximately 18% of patients developed recurrent UTI.

Dowson (2012) reported on 100 patients, 63 of whom had two or more injections (53 had two, 20 had three, 13 had four, 10 had five, 5 had six, 3 had seven, 1 had eight, 1 had nine and 1 had ten injections).²⁰⁰ Significant reductions in frequency, urgency, and UUI were observed following each injection. The most common reasons for discontinuing injections were poor efficacy (13%) and the need for CISC (11%). CISC was needed by 35% of patients after the first injection; 21% developed UTIs. AUR occurred in one patient; two patients required indwelling catheters. One patient developed urosepsis after his third injection. Additional adverse events were bladder pain persisting >1 week (2 patients), flu-like symptoms for 1 week (1 patient), transient arm paresthesia (1 patient), and leg weakness (1 patient).

Abeywickrama (2013) reported on 33 female patients (mean age 59.3 years) with idiopathic DO who had up to three injections of Dysport.¹⁹⁸ Patients received 500U which was increased to 750U if symptoms returned within six months; location of injections was not specified. After each injection, significant improvements occurred in frequency, nocturia, mean voided volume, and maximum voided volume. Scores on the ICIQ also improved after each injection. The mean interval between the first and second injections was 15.2 months and between the second and third injections was 19.2 months (statistically significantly longer). About 10% of patients required CISC (3 of 33); two patients developed UTIs after each injection.

More information regarding adverse events: Jackson (2012) reported on 94 patients who had 200U of BTX-A injected into the detrusor. ²⁰⁴ Patients with idiopathic DO had similar success rates (81%) as patients without idiopathic DO (89%) (success defined as patient reporting symptom improvement). Post-injection, 29% of patients required CISC. Kanagarajah (2012) reported on 32 patients without idiopathic DO, 19 of whom had OAB-dry and 13 of whom had OAB-wet.²⁰⁵ Patients had 100U or 150U of BTX-A injected into the detrusor. QOL measures improved significantly by 3 months post-injection (UDI-6, VAS scale). Frequency improved significantly among OAB-dry patients; UUI improved significantly among OAB-wet patients, dropping by more than 50%. Three patients who received 150U and one patient who received 100U developed large PVRs and required CISC. Five patients developed UTIs.

Injection location: Manecksha (2012) reported on 22 patients randomized to trigone-including  versus trigone-sparing injections of 500U Dysport into the bladder wall.²⁰⁷ Patients were followed for up to 26 weeks. Both groups exhibited improvement on the OABSS at 12 and 26 weeks; the urgency subscale also showed improvement at 6, 12, and 26 weeks. The size of the improvement, however, was greater in the trigone-including injected group. No patients developed vesicoureteral reflux and magnitude of PVR increase and rates of CISC (2 patients in each group) were similar. 

Dysport. Irwin (2013) reported on 73 patients administered 250U of Dysport into the detrusor and suburothelium.²⁰² Significant improvements were documented in frequency, nocturia, urgency, UI, and on QOL measures. CISC was required after 16.8% of procedures (some patients had more than one injection), was discontinued after three weeks in four cases, after three months in five cases, but the remaining seven patients "continue to self-catheterise 5-24 months following BTX treatment (median 11 months)."

BTX-A  versus Dysport: Ravindra (2013) reported on 207 patients (detrusor injections including the trigone), 101 treated with BTX-A (200U) and 106 treated with Dysport (500 U initially later dropped to 300U).²¹³ Both forms of BTX significantly reduced daily frequency, nocturia, and incontinence episodes. Patient groups reported similar global improvement rates (81% of BTX-A group; 90% of Dysport group). Therapeutic effects had similar durations (mean 10.7 months in BTX-A group; mean 10.9 months in Dysport group). However, adverse event rates differed significantly between the two preparations. Among BTX-A patients, 23% required CISC compared to 42% of patients administered Dysport. Among patients who received 300U Dysport, the CISC rate was 46%; the rate in the 500U group was 37%.

Patient satisfaction: Brubaker (2012) reported on the same patients evaluated in Dmochowski (2010; the dose-finding trial in the current guideline).¹⁹⁹ Responses on the PSTQ and assorted global assessment measures indicated that greater proportions of patients in the BTX-A groups attained their primary OAB treatment goal (34.5% to 65.3%) compared to those in the placebo group (23.7%). El-Azab (2013) reported on 31 patients who had either BTX-A injections (100 or 200U into the detrusor) or augmentation ileocystoplasty (AC) for refractory idiopathic OAB. Treatment satisfaction was measured with the OAB-SAT-q at three and six months post-procedure. The AC patients had significantly higher OAB-SAT-q scores than did the BTX-A patients. BTX-A patients cited the need for repeat treatments to maintain symptom control as a primary reason for dissatisfaction. Makovey (2011) reported on 85 patients who had 150 or 200U BTX-A injected into the detrusor.²⁰⁹ All patients had failed anticholinergic medications either for lack of efficacy (57 patients) or because of intolerable side effects (28 patients). Success was defined based on patient-reported symptomatic improvement and was more likely to occur in patients who could not tolerate medication side effects (86%) compared to patients who did not experience medication efficacy (68%).

Presence of antibodies. Hegele (2011) reported on the presence of antibodies post BTX-A injection in 31 patients.²⁰³ Eleven patients were treated once, 16 were treated twice, and four were treated three times. Blood was collected before and three months post injection. In five patients (16%) BTX-A antibodies were detectable. One of these patients had a strongly positive titer and experienced complete failure of the treatment. The other four had slightly positive titers; one patient had a poor response to the second injection. Authors speculate that the presence of antibodies is involved in poor treatment responses.

Systematic review. Cui (2013) reports a systematic review and meta-analysis of 12 randomized BTX trials for idiopathic OAB.²²⁰ Their meta-analysis indicates that in trials that measured incontinence (Flynn 2009, Sahai 2009, Tincello 2012), the mean difference in incontinence episodes between BTX-treated patients (any dose) and placebo patients is -3.85 (95% CI -4.79 to -2.90).^170,171,196 The mean difference in frequency in trials that reported that outcome (Flynn 2009, Sahai 2009) between BTX-treated patients and placebo-treated patients is -5.13 (95% CI -7.86 to -2.39).^170,171 The relative risk ratio for CISC in BTX-treated patients compared to placebo-treated patients (based on data from Brubaker 2008, Denys 2012, Flynn 2009, Fowler 2012, Sahai 2009, Tincello 2012) was 5.25 (95% CI 2.47 to 11.16).^{168,170,171,194,196,218} The relative risk ratio for UTI (using data from Brubaker 2008; Denys 2012; Flynn 2009; Fowler 2012; Tincello 2012) in BTX-treated patients compared to placebo-treated patients was 2.36 (95% CI 1.58 to 3.53). ^{168,170,194,196,218}

The Panel interpreted these data to indicate that onabotulinumtoxinA injections can improve moderate to severe OAB symptoms in the context of adverse events that could require secondary intervention (e.g., an untreated UTI, undiagnosed urinary retention). The Panel notes that at the FDA-approved dose of 100U some adverse events appear to occur less frequently. For example, rates of urinary retention in study arms that administered 100U were <20% compared to up to 43% in study arms administering higher doses. Similarly, the percent of patients requiring CISC who were administered 100U was generally less than 10% compared to up to 43% in studies using higher doses. The Panel judged that the benefits of BTX-A at the 100U dose in the carefully counseled patient outweigh its risks/burdens and designated it a Standard. The Panel notes that patients considering onabotulinumtoxinA treatment must be counseled regarding the possible need to perform self-catheterization for long periods (or to have a caregiver perform catheterization) and should be willing to accept this possibility. OnabotulinumtoxinA treatment also may require access to a clinician who can measure PVR on a periodic basis if necessary. Further, effects diminish over time for most patients; therefore, patients also should be informed that repeat injections are likely to be necessary to maintain symptom reduction. The Panel also believes that this procedure should be performed by experienced personnel familiar with intravesical injection techniques.

Evidence strength is Grade B given a body of evidence constituted by randomized trials. Limitations of the available evidence include short follow-up durations in the best-designed studies (ranging from 4 to 12 weeks for the RCTs), the variability in doses and injection sites across studies, and adverse event reporting that was variable. 

In the original literature review, eight studies reported in ten publications assessed the efficacy of PTNS to treat OAB symptoms.^{103, 109, 221-228} The majority of studies were single-group observational designs that evaluated patients with refractory OAB symptoms. Patients in these studies are characterized by having moderately severe baseline levels of incontinence (ranging from 2.2 to 9.8 episodes per day) with most studies assessing patients with more than 3 episodes a day. Patients had moderately severe frequency symptoms at baseline, ranging from 11.8 to 16.5 episodes per day. Most trials report improvements in all measured symptoms, including incontinence (typical reductions of 1 to 3 episodes/day), frequency (reductions of 2 to 5 episodes/day), nocturia (reductions of 1 to 2 episodes/night) and QOL. The most common protocol was the application of 30 min of stimulation once a week for 12 weeks (the trial duration; for continued benefit, weekly stimulation would have to continue). Peters (2009) compared PTNS to tolterodine ER 2-4 mg daily and reported similar improvements in both groups in voiding parameters but a greater proportion of patients in the PTNS group indicating subjective improvement.¹⁰³ Sancaktar (2010) compared tolterodine ER 4 mg daily with and without PTNS and noted that the combined treatment group improved more than did the tolterodine alone group.¹⁰⁹ Two reports followed patients for long periods of time (44 weeks in Klingler 2000; 52 weeks in MacDiarmid 2010 - a long-term report on patients initially evaluated in Peters 2009) and indicate that improvements were maintained as long as the treatment was maintained.^{103, 221-223} Additional studies did not report raw voiding data but reported improvements in symptoms and QOL with treatment^{229, 230} that ceased when treatment ceased.²²⁹ The validity of PTNS treatment responses is supported by Peters (2010), which compared a PTNS group to a sham-PTNS group and found that only the active treatment group exhibited improvements in frequency, nocturia and urgency incontinence. Adverse events were relatively uncommon and mild.

The 2014 update literature search retrieved eleven new publications that reported outcomes from nine studies, including one RCT (Finazzi-Agro 2010),²³¹ one randomized design (Souto 2014),²³² and seven observational designs. Patient inclusion criteria remain varied with some studies requiring frequency and nocturia and others requiring UUI. Although most studies reported outcomes at 12 weeks, several reported longer-term findings. Peters (2013a, 2013b) reported findings in a group of responders from the SUmiT trial who continued with PTNS therapy for up to 36 months.^233,234 Yoong (2013) reported one-year findings for a group of PTNS responders.²³⁵ Several other papers reported partial findings beyond the formal study end date (see text below). Sample sizes remained relatively small (range 14 to 60 patients) with most studies having fewer than 25 patients in each treatment arm.

The best quality evidence comes from Finazzi-Agro (2010), a double-blind RCT (placebo arm received stimulation for 30 sec in the gastrocnemius and the device was then turned off; all patients were told they may not experience sensation during treatment).²³¹ PTNS patients had twelve 30-minute sessions three times a week; placebo patients had the same number and duration of sessions but with only 30 sec of active current in the alternate muscle location. Patients in the PTNS group, but not in the placebo group, experienced statistically significant improvements in incontinence, frequency, voided volumes, and I-QOL scores. In the PTNS group, 71% were characterized as responders (defined as experiencing at least a 50% reduction in UUI episodes) compared to 0% in the placebo group. In Souto (2014), patients were randomized to three groups: PTNS, oxybutynin ER 10 mg/daily, and PTNS + oxybutynin ER 10 mg/daily.²³² PTNS patients had treatments twice a week, for 30 min, for 12 weeks. At 12 weeks, all three groups showed similar improvements in frequency, incontinence, nocturia, ICIQ-SF scores, ICIQ-OAB scores, and symptom bother scores. The authors followed patients after treatment cessation for another 12 weeks. At week 24, the oxybutynin group had significantly worse scores compared to week 12 on the QOL measures - but not the two groups that had PTNS. Frequency, incontinence, and nocturia data at 24 weeks were only reported as proportions of patients exhibiting these symptoms; it appears that the oxybutynin only group had decaying responses compared to the PTNS groups. Three additional observational studies (Ugurlucan 2013; Zhao 2011; Onal 2012) generally reported improvements in voiding parameters and QOL outcomes after five to 12 weeks of treatment. ^236-238

Treatment effects duration. Several studies focused on documenting the duration of treatment effects once treatment ceased (e.g., Marchal 2011; Sherif & Abdelwahab 2013; Arrabal-Polo 2012).^239-241 Using protocols that ranged from 12 weekly sessions to a protocol that used tapering session frequency over five to six months, these studies generally reported that treatment benefits were retained for four to six months once treatment ceased.

Longer-term outcomes in patients who undergo maintenance treatments. Peters (2013a, 2013b) reported long-term outcomes for PTNS responders identified during the SUmiT trial who were willing to continue with therapy.^233,234 Patients in the SUmiT trial (Peters 2010) had 12 weekly PTNS treatments.²⁴² Responders received PTNS therapy in sessions that tapered during a transition phase: two treatments at 14-day intervals, two treatments at 21-day intervals; one treatment at a 28-day interval. After tapering, patients had a customized treatment plan in which PTNS sessions were prescribed based on patient report of increasing OAB symptoms. Throughout the 36 month follow-up period, patients reported significant improvements in frequency, nocturia, urgency episodes, and UUI. Scores on the OAB-q and HRQOL as well as symptom severity scores also remained significantly improved. Among study completers, patients received a median of 1.0 treatments per month (IQR 0.9 to 1.2) with 41% receiving <1 treatment/month, 55% receiving 1.0 to <2.0 treatments/month, and 4% receiving 2.0 to <3.0 treatments/month.

Yoong (2013)²³⁵ also reported long-term follow-up data on a group of PTNS responders originally discussed in Yoong (2010).²⁴³ Of the original 30 patients who reported positive responses defined as OAB symptoms no longer dominant, 50% reduction in frequency, and 25% reduction in IIQ-7 scores, 23 continued to receive maintenance treatments (30 min sessions). The sessions were scheduled by the patients when they felt they needed a treatment. At two years, frequency, UUI, nocturia, pad use, and IIQ-7 scores were statistically indistinguishable from those recorded after initial responses to treatment, indicating treatment effects had been maintained. Patients received a median 8.42 treatments per year and median time between treatments was 64.3 days.

Systematic reviews and meta-analyses. Since the completion of the original guideline, five systematic reviews assessing PTNS have been published.^244-248 In general, these reviews conclude that there is convincing evidence for the efficacy of PTNS in comparison to placebo or sham conditions, that from 37% to 100% of patients are reported to meet criteria for success (but success criteria varied across studies), and that adverse events are minimal. Burton (2012) also provided a meta-analysis.²⁴⁵ When the randomized trials with placebo or sham control groups are considered, the relative risk ratio (RR) for successful treatment is 7.02 (95% CI 1.69 to 29.17); that is, PTNS patients were seven times more likely to report success compared to placebo patients. In an analysis that included prospective non-randomized trials, the pooled subjective success rate in PTNS patients was 61.4% (95% CI 57.5% to 71.8%; differing definitions of success) and the pooled objective success rate (based on voiding parameters but also with different definitions of success) was 60.6% (95% CI 49.2 to 74.7%). In trials that compared PTNS to anti-muscarinics, there were no differences in efficacy. The authors conclude that PTNS significantly improves OAB symptoms, that the effects are similar in magnitude to anti-muscarinics, but that PTNS has a better adverse event profile. The primary weakness identified by this group of papers is the lack of long-term follow-up in a randomized design.

The Panel interpreted these data to indicate that PTNS can benefit a carefully selected group of patients characterized by moderately severe baseline incontinence and frequency and willingness to comply with the PTNS protocol. Patients must also have the resources to make frequent office visits both during the initial treatment phase and to obtain maintenance treatments in order to achieve and maintain treatment effects. Reported adverse events were minor; the most frequently reported events were painful sensation during stimulation that did not interfere with treatment and minor bleeding at the insertion site. In the Panel's view, benefits outweigh risks/burdens for the use of PTNS in the thoughtfully-selected and counseled patient who is highly-motivated to make the required office visits.

As a group, the PTNS studies constitute Grade C evidence because of the predominant observational designs, varying patient inclusion criteria, small sample sizes, and short follow-up durations for most studies.

In the original literature review thirteen studies, predominantly single-group observational designs, evaluated sacral neuromodulation in patients with severe refractory OAB symptoms, many of whom had failed multiple other therapies.^249-261 In general, patients were characterized by extremely severe levels of baseline incontinence, ranging from 5.0 to 11.6 episodes per day, by severe frequency (most studies reporting baseline levels of more than 13 episodes per day) and by pad use of more than 4 per day at baseline in most studies. This group of studies is characterized by much longer follow-up durations than in other OAB studies, with follow-up ranging from 24 weeks to 260 weeks and most studies following patients for more than a year. In general, studies reported that all measured parameters, including QOL and subjective improvement, show improvement with treatment and that improvement dissipates if treatment ceases. Siegel (2000), Janknegt (2001) and van Kerrebroeck (2007) evaluated the same groups of urgency incontinence patients compared to urgency-frequency patients and reported that at 5 years post-surgery greater than a 50% improvement was reported by 68% of the UI group and 56% of the urgency-frequency group.^{252, 257, 260} Groen (2011) reported that treatment success (defined as ≥ 50% decrease in the number of daily incontinence episodes or pads used) was 87.0% of patients at one month post-surgery with a decline to 62.0% at five years.²⁵⁰ An additional study reported on urodynamics outcomes for patients evaluated by Schmidt (1999) and noted that patients with UI, with and without DO, had similar improvements in urodynamic parameters.^{256, 262} Leong (2011) assessed long-term satisfaction with SNS and reported that 90% of 207 patients surveyed reported being satisfied with the treatment (median post-implant interval of 77 months).²⁵⁴ In an effort to reduce adverse events and possibly limit nervous system adaptation and diminished efficacy that may occur with continuous stimulation, Oerlemans (2011) tested an on-demand protocol in which patients turned the apparatus off for several hours a day.²⁵⁵ Approximately 63% of patients were able to maintain symptom improvement by using the on-demand procedure during the two-week test. 

The 2014 updated literature review retrieved an additional 16 relevant treatment studies, including one prospective randomized multi-center trial,²⁶³ one crossover study,²⁶⁴ and 14 observational studies. Seigel (2014) reported findings at six months of follow-up for the InSite trial, an ongoing FDA-mandated post device approval study that included a subsample of patients randomized to SNS or to standard medical therapy (SMT; anti-muscarinic medications).²⁶³ The study used a less invasive procedure than in older studies and also used the newer tined lead. A total of 147 patients were randomized (SNS - 70; SMT - 77) and 130 patients completed six months of treatment (SNS - 59; SMT - 71).

Siegel (2014) differs from the studies discussed above in that patients had less severe symptom levels at baseline (SNS: mean 11.2 voids/day; mean 2.4 incontinence episodes/day, mean 1.1 pads/day; SMT: mean 11.9 voids/day, mean 2.7 incontinence episodes/day; mean 1.5 pads/day).²⁶³ In addition, the primary outcome was OAB therapeutic success defined as ≥50% improvement in average incontinence episodes/day or voids/day or a return to normal voiding frequency of <8 voids/day rather than change in voids or incontinence episodes. At 6 months, the OAB success rate was 61% in the SNS group compared to 42% in the SMT group (p=0.02). In addition, <8 voids/day was achieved by 61% of SNS patients compared to 37% of SMT patients (p=0.04). The SNS group also improved more on the OAB-QOL than did the SMT group (p<0.001), SNS female patients reported a greater improvement in sexual function than did SMT female patients (p<0.05), and the SNS group exhibited greater improvements in Beck Depression Inventory scores than did the SMT group (p=0.01). However, limited information is reported regarding the intervention content of the SMT arm; the report notes only that 96.1% of SMT participants used OAB medications between randomization and the six month follow-up, that 70% used medication on at least 80% of the days during the follow-up period, and that some participants (proportion not specified) used more than one medication. Lack of information regarding medication use and that fact that patients had mild symptoms compared to most other SNS studies limits the interpretability of SMT outcomes compared to SNS outcomes.

The crossover study evaluated whether different stimulator settings altered outcomes.²⁶⁴ Patients in this study had had an SNS implant with a tined lead for at least three months prior to study beginning and were refractory to conventional treatments including medications at the time of SNS implant. Settings were 5.2 Hz, 14 Hz, or 25 Hz and were maintained for one week. Numbers of incontinence episodes and pad changes were significantly affected by rate such that the 14 Hz and 25 Hz settings reduced these outcomes compared to the 5.2 Hz setting.

The observational studies evaluated a wide range of patient subgroups, clinical questions, and outcomes. More than half of the studies followed patients for one year or longer.

Testing phase outcomes. Davis (2013) reported success rates during the testing phase in patients who presented for SNS because of lack of medication efficacy or intolerable medication side effects.²⁶⁵ Success rates at the testing phase were similar (70% and 71% respectively). Yazdany (2011) also reported on testing phase success in a group of patients with severe incontinence (mean 10.4 episodes/day); the authors note that patients with >10 incontinence episodes per day were more likely to have a successful stage I trial compared to those with less than 5 episodes/day.²⁶⁶ Levin (2012) reported on the impact of obesity on stage I success rates.²⁴⁷ Of 149 patients, 80 (53.7%) were obese (BMI mean 37.3) and 69 (46.3%) were non-obese (BMI mean 25.6). The overall stage I success rate was 81% and the success rates for non-obese patients (83%) was statistically indistinguishable from the success rates for obese patients (78%). Gleason (2013) reported on stage I SNS effects on periurethral sensation and urethral sphincter activity.²⁶⁷ Baseline urethral sensation did not differ between patients who were classified as stage I responders versus non-responders, however, responders had larger amplitude, longer duration, and more turns and phases at baseline, indicating more successful urethral re-innervation, than did non-responders. 

Treatment phase outcomes. Lee (2013) compared motor and sensory responses to SNS and noted that patients with compound muscle action potential (cMAP) were more likely to report the sensation of stimulation than patients without cMAP.²⁶⁸ Nineteen of 31 patients did not have cMAP. Of these, 16 were successfully reprogrammed to achieve cMAP, resulting in improvements in nocturia, incontinence, and UI that did not reach statistical significance. Cardarelli (2012) reported significant decreases in UUI episodes, frequency, nocturia, and pad use as well as significant increases in voided volumes at 48 weeks post-implant.²⁶⁹ Moon (2013) evaluated patients with severe levels of UUI, frequency, and nocturia and reported at 12 months post-implant significant decreases in urgency episodes, UUI, frequency, nocturia, and in the severity of urgency episodes.²⁷⁰ Various urodynamics parameters also improved as did scores on the OAB-q. Improvements were similar for patients with OAB-wet compared to OAB-dry. Yih (2013) reported on changes in sexual functioning in women after SNS implant.²⁷¹ At 12 months post-implant, FSFI scores improved significantly from mean 13.5 at baseline to mean 15.9 and ICSI-PI composite score improvements were similar among women with baseline FSFI scores of < 26 compared to ≥ 26. 

Older patients. Angioli (2013) reported in patients >age 65 years (mean patient age 76 years) that at 12 months post-implant, 27.8% of patients reported improvement and 55.5% of patients reported complete success with cessation of UUI episodes.²⁷² Overall, UUI episodes decreased from mean 6.3/day to mean 0.5/day. Incontinence episodes, frequency, nocturia, and number of pads used daily also significantly decreased. All subscales of the OAB-q exhibited significant improvement. Peters Killinger (2013) compared patients aged 40 to 64 years with patients older than age 64.²⁷³ Most of the patients in these two groups had OAB (a subset had IC or urinary retention). For both groups, incontinence episodes, frequency, urgency, nocturia, OAB-q scores, and ICSI-PI scores improved significantly at 26 weeks.

Patients with concomitant bowel dysfunction. Faucheron (2012) evaluated patients with both urinary and fecal incontinence.²⁷⁴ At more than 5 years post-implant, UUI episodes (and fecal incontinence episodes) were significantly reduced. Approximately 74% of patients were satisfied with their improvements in both forms of incontinence and all QOL measures also exhibited significant improvement (e.g., the Fecal Incontinence QOL measure and the Ditrovie score). Gill (2012) reported on patients with UUI of which 83% had concomitant bowel dysfunction (either constipation or fecal incontinence).²⁷⁵ At 4 months post-implant, Likert scale ratings (patient perception of disease) were significantly improved from baseline and indicated that symptom severity and disease state had become "normal." UDI-6 and CRADI-8 scores also reflected significant symptom improvement. When patients were examined separately based on type of bowel dysfunction, the improvement in scores appeared to occur for the constipation group rather than the incontinence group.

Patients who discontinued BTX-A. Smits (2013) reported on a group of patients who had discontinued BTX-A therapy because of lack of efficacy (85%) or the desire for a treatment that was more permanent (15%).²⁷⁶ The primary outcome was improvement in leakage, defined as greater than 50% improvement in leakage episodes and severity of leakage measured on a 1 to 4 scale. Secondary outcomes were improvement in frequency and urgency defined as greater than 50% improvement (measurement not specified but presumably also on a 1 to 4 scale). Mean interval between last BTX-A treatment and SNS test phase was 23 months. At one year post-implant, 11 of 14 patients (79%) reported satisfaction with treatment.

In contrast to PTNS studies, SNS studies reported frequent adverse events, including pain at the stimulator site (3.3 to 19.8% of patients), pain at the lead site (4.5 to 19.1% of patients), lead migration (1.1 to 8.6% of patients), infection/irritation (2.2 to 14.30% of patients), electric shock (5.5 to 10.2% of patients) and need for surgical revision (6.25 to 39.5% of patients). In most studies, the need for surgical revision occurred in greater than 30% of patients. There is some evidence that newer, less invasive surgical procedures and tined devices may be associated with fewer adverse events.²⁵⁹ Leong (2011) reported that although 90% of patients reported satisfaction with SNS, 56% reported adverse events, particularly pain at the stimulator site and when the stimulator was turned on and daily life limitations, such as difficulty passing through airport metal detectors and inability to undergo magnetic resonance imaging (MRI).²⁵⁴

Additional useful information is provided by Lai and Grewal (2013) who investigated the bacterial colonization rate of the connector and lead during staged testing.²⁷⁷ During the stage II procedure, aerobic and anaerobic cultures were obtained by swabbing the connector pocket, the connector, and the permanent lead itself. Of 38 patients, 9 (24%) had a positive culture at the connector or lead site. Of the 9 patients with a positive culture, 3 (33.3%) subsequently developed device infection that required explant compared to 3% of patients who did not have colonization but who subsequently developed an infection. Longer percutaneous testing was associated with a greater colonization rate. Of 10 patients who underwent >14 days of staged testing, 50% developed connector and/or lead colonization. In contrast, only 4 (14%) of the 28 patients who had ≤ 14 days of testing developed colonizations.

Cameron (2013) reported on reprogramming and battery explant rates from 1997 to 2007 in a 5% sample of Medicare beneficiaries at mean follow-up 60.5 months.²⁷⁸ Among OAB-wet patients and OAB-dry patients, during the first year there were mean 2.14 and 2.26 reprogramming events respectively. Reprogramming events declined over time; at the 5-year point, rates were 0.24 and 0.95 respectively. Explant rates for OAB-wet were 10.9% (32/294) and for OAB-dry 8.7% (10/115). In comparison, IC patients had an explant rate of 57.9%.

The Panel interpreted these data to indicate that in carefully selected patients, SNS is an appropriate therapy that can have durable treatment effects but in the context of frequent and moderately severe adverse events, including the need for additional surgeries. The Panel notes that patients should be counseled that the device requires periodic replacement in a planned surgical procedure and that the length of time between replacements depends on device settings. Patients also must be willing to comply with the treatment protocol because treatment effects typically are only maintained as long as the therapy is maintained and have the cognitive capacity to use the remote control to optimize device function. In addition, patients must accept that the use of diagnostic MRIs is contraindicated in individuals with the device implanted. Given the negative effects on QOL associated with severe incontinence and frequency, the Panel judged that benefits of SNS in the appropriate patient outweighed the risks/burdens and notes that patients should be carefully counseled regarding the risks/burdens. Evidence strength is Grade C because of the predominance of observational designs, the small sample sizes, the limited number of unique patient groups (i.e., there are multiple reports on the same patient groups followed over time) and limited information regarding the protocols used by patients to maintain symptom control.

The Panel members note that they regularly encounter patients who present for second-line (pharmacologic management) or third-line treatments (BTX-A, PTNS, or SNS) who have never undergone a comprehensive evaluation (i.e., completion of a voiding diary to ensure the OAB diagnosis is correct) or who have never had a trial of behavioral therapy or who have had an inadequate trial of behavioral therapy. Similarly, it is not uncommon for patients to present for non-medical treatments who have not had an adequate trial of medications. On the other hand, the Panel also encounters patients who are being treated with multiple simultaneous therapies without clear evidence of the efficacy of the individual therapies or their combination. The Panel encourages practitioners and patients to persist with new treatments (4 to 8 weeks for medications and 8 to 12 weeks for behavioral therapies) for a sufficient duration to achieve clarity regarding efficacy and adverse events for a particular therapy before abandoning the therapy prematurely or before adding a second therapy. If a comprehensive evaluation has demonstrated that the patient has signs and symptoms consistent with the OAB diagnosis and a particular therapy is not efficacious after a reasonable trial, then an alternative therapy should be tried. Combination therapeutic approaches should be assembled methodically, beginning with the establishment of confidence in the partial efficacy of one therapy, continuing with an adequate trial of any additional therapies one at a time until the patient experiences adequate symptom control in the context of tolerable adverse events. If a patient does not achieve adequate symptom control with this approach, then referral to a specialist should be considered.

In general, surgery is not recommended for OAB patients except in extremely rare cases. The vast majority of case series that document the effects of augmentation cystoplasty and diversion focus on neurogenic patients. Little is known regarding the impact of these procedures on non-neurogenic OAB patients and, particularly, on their QOL. There are substantial risks to these procedures, however, including the likely need for long-term intermittent self-catheterization and the risk of malignancy.²⁷⁹ In the Panel's judgment, therefore, a surgical approach to OAB treatment is appropriate only in the extremely rare patient.

In situations where the medical management of burdensome OAB, as outlined above, is not feasible, effective nor recommended, as in the patient with severe cognitive deficits or mobility issues, then other management options may need to be considered. Management with diapering and absorbent garments is always preferred to indwelling catheterization because of the high risk of indwelling catheter-associated UTIs, urethral erosion/destruction and urolithiasis. Intermittent catheterization may be an option when concomitant incomplete bladder emptying is present leading to overflow incontinence; however, this approach generally requires either patient willingness and ability or significant caregiver support. As a last resort, an indwelling catheter may be considered when urinary incontinence has resulted in the development and progression of decubiti, during the management of those decubiti, or rarely, where urinary incontinence is the predominant disability affecting activities of daily living and therefore may result in institutionalization. 

The purpose of follow-up is to assess compliance with treatment protocols, query patients regarding symptom improvements and any adverse events and present information about possible alternative treatments to patients who have insufficient symptom improvement and/or intolerable adverse events. There are many ways to measure symptom changes, including voiding diaries with or without frequency-volume charts and patient-rated global response scales for urgency, urgency incontinence, incontinence, frequency and nocturia. In addition, validated OAB-specific instruments may be used to assess the impact of OAB symptoms on QOL.  Ideally, clinicians should obtain baseline measures using the same instruments in order to chart progress. Patients should be encouraged to persist with a particular treatment for four to eight weeks; this time period will identify the majority of responders.⁶¹

Clinicians and any clinical personnel engaged in follow-up should be aware that the various treatment options for OAB have different requirements for efficacy and different adverse event probabilities and severities. For example, the efficacy of some treatments (e.g., behavioral therapies, neuromodulation) depends greatly on treatment compliance, and the efficacy must be balanced against possible adverse events. For other treatments, such as the use of anti-muscarinics, adverse events are common but vary in severity across patients. Patients should be informed about and subsequently queried regarding dry mouth and its severity (i.e., sufficient to impair alimentation), constipation, fecal retention and any possible central nervous system (CNS) effects. Queries of the patient and caregiver regarding CNS effects are particularly important in elderly or frail patients; clinical experience suggests that CNS effects can be severe enough to cause loss of independent living skills in some patients. Non-responders to anti-muscarinics should be tried on at least one other anti-muscarinic or mirabegron and/or dose modification attempted to determine if a better balance between efficacy and adverse events occurs. If adverse events are severe enough to compromise patient QOL, then strategies to manage specific adverse events, such as ameliorating constipation with appropriate bowel management, should be implemented before abandoning anti-muscarinic treatment.

For peripheral tibial nerve stimulation and sacral neuromodulation, pre- and post-therapy measures are essential to assess efficacy. The before-and-after evaluation should include baseline assessment with a voiding diary and assessment of urgency as well as a global response assessment. Adverse events such as pain and collateral stimulation should be assessed, and sacral neuromodulation wound complications should be evaluated.  

Patients treated with intradetrusor onabotulinumtoxinA should be followed for the possibility of increased PVRs and the need for self-catheterization. Patients who have undergone surgical treatments (e.g., augmentation cystoplasty with or without sling, supravesical diversion) or permanent or semi-permanent catheter placements also should be followed regularly for symptom level, QOL and any complications. Patients who are using incontinence pads, regardless of whether or how they are being treated, should be followed for appropriate skin care and skin integrity.