# Third-Line Therapies for Refractory OAB

## Overview

Third-line therapies are considered when behavioral interventions and pharmacological treatments, including antimuscarinics, beta-3 agonists, or their combinations, have failed or are not well tolerated by the patient. The American Urological Association (AUA) and the Society of Urodynamics, Female Pelvic Medicine & Urogenital Reconstruction (SUFU) endorse three primary third-line options: intravesical injection of onabotulinumtoxinA (Botox), sacral neuromodulation (SNM), and percutaneous tibial nerve stimulation (PTNS). There is no definitive evidence favoring one modality over the others, so the choice depends largely on patient preference, existing comorbidities, and the patient’s willingness to perform self-catheterization if necessary.

## OnabotulinumtoxinA (Botox) Injection

### Mechanism of Action

OnabotulinumtoxinA works by inhibiting the release of acetylcholine at the neuromuscular junction, which produces a motor effect by reducing detrusor muscle contractions. Additionally, it suppresses the release of sensory neurotransmitters such as ATP, substance P, and calcitonin gene-related peptide (CGRP) from afferent nerves, thereby diminishing bladder sensory signaling. This dual action reduces both involuntary detrusor overactivity and the sensation of urgency. The effects are temporary because the neuromuscular junction regenerates over a period of 6 to 12 months.

### Technique

The procedure is performed in the office setting using cystoscopy under local anesthesia. Local anesthetic is administered by instilling lidocaine into the bladder 15 to 20 minutes before the injection. For idiopathic overactive bladder (OAB), a dose of 100 units is used, whereas 200 units are recommended for neurogenic detrusor overactivity. The toxin is reconstituted in 10 mL of preservative-free saline, yielding a concentration of 10 units per milliliter. Injections of 0.5 mL are made into the detrusor muscle at 20 different sites distributed across the posterior bladder wall, dome, and lateral walls. The trigone is typically avoided in idiopathic OAB to reduce the risk of vesicoureteral reflux, although some centers now include trigonal injections to improve efficacy. The injections are superficial within the detrusor and do not penetrate full thickness.

### Efficacy

OnabotulinumtoxinA significantly reduces episodes of urgency urinary incontinence by 50 to 70%, with complete continence achieved in 20 to 40% of patients. It also improves symptoms of urgency, frequency, and nocturia. The onset of effect occurs within 1 to 2 weeks, with peak efficacy at 4 to 6 weeks post-injection. The duration of benefit averages 6 to 9 months, necessitating repeat injections for sustained symptom control. Repeated treatments maintain efficacy without evidence of tachyphylaxis.

### Complications

Urinary retention or elevated post-void residual (PVR) occurs in 5 to 8% of patients, requiring temporary clean intermittent catheterization (CIC). It is essential to counsel patients about the possibility of needing CIC before treatment. PVR should be assessed 2 to 4 weeks after injection. Urinary tract infections (UTIs) occur in 15 to 25% of patients, a higher rate compared to other third-line therapies. Hematuria is usually transient and occurs in 2 to 5% of cases. Pain at the injection sites is minimal when adequate local anesthesia is used. Systemic effects are extremely rare at the 100-unit dose.

### Patient Selection

Ideal candidates for Botox injections are those willing and able to perform self-catheterization if necessary, have good manual dexterity, are not on anticoagulation (a relative contraindication), and do not have an active UTI. Contraindications include urinary retention, active infection, myasthenia gravis, pregnancy, and allergy to botulinum toxin.

<image>Cystoscopic view during onabotulinumtoxinA injection showing needle placement into the detrusor at standardized injection sites across the bladder wall</image>

## Sacral Neuromodulation (SNM)

### Mechanism of Action

Sacral neuromodulation involves electrical stimulation of the S3 sacral nerve root, which modulates afferent signaling to the pontine micturition center. This modulation restores the balance between excitatory and inhibitory neural pathways, thereby inhibiting abnormal detrusor contractions. SNM is also effective for non-obstructive urinary retention and fecal incontinence, although its exact mechanism remains incompletely understood.

### Procedure (Two-Stage)

The SNM procedure is typically performed in two stages. The first stage is a test phase, known as peripheral nerve evaluation, during which a tined lead electrode is placed at the S3 foramen under fluoroscopic guidance. This lead is connected to an external pulse generator, and the patient undergoes a test period lasting 2 to 4 weeks. Success is defined as at least a 50% improvement in symptoms based on voiding diary comparisons, with success rates during this phase ranging from 60 to 80%. If the test phase is successful, the second stage involves permanent implantation of an implantable pulse generator (IPG) in a subcutaneous pocket in the upper buttock. The IPG is connected to the permanent lead, which may be the same lead used during the test phase. The device settings are programmable and can be adjusted externally by both the patient and clinician.

### Devices

Several devices are available for SNM. The Medtronic InterStim II is a non-rechargeable device with a battery life of 5 to 7 years, requiring replacement surgery upon depletion. The Medtronic InterStim Micro is a smaller, rechargeable device with a battery life exceeding 15 years, necessitating weekly recharging. The Axonics System is also rechargeable, offers a 15-year battery life, and is fully MRI conditional at 3 Tesla. Regarding MRI compatibility, the InterStim II is conditional only for 1.5 Tesla head and extremity scans, whereas the InterStim Micro and Axonics devices allow full-body MRI at both 1.5 and 3 Tesla.

### Efficacy

During the test phase, 60 to 80% of patients respond favorably. Long-term success rates are approximately 60 to 70% at five years, with improvements noted in urgency incontinence, frequency, urgency, and urinary retention. SNM is also FDA-approved for fecal incontinence, providing a dual indication. However, revision surgery is common, occurring in 25 to 40% of patients within five years, primarily due to lead migration, pain, or loss of efficacy.

### Complications

Common complications include pain at the implant site (15 to 20%), lead migration (5 to 10%), and infection (2 to 5%), which may necessitate device explantation. Patients may experience undesirable stimulation sensations in the leg or perineal region. Device malfunction and battery depletion (in non-rechargeable devices) are additional concerns. The overall revision or explantation rate is 25 to 40% at five years.

### Contraindications

SNM is contraindicated in patients unable to operate the device, pregnant women (the device should be turned off during pregnancy), those requiring frequent MRI scans with older non-MRI conditional devices, and individuals with sacral anatomic abnormalities that prevent lead placement.

<image>Fluoroscopic image of S3 tined lead placement for sacral neuromodulation, and photograph of the implantable pulse generator with patient programmer device</image>

## Percutaneous Tibial Nerve Stimulation (PTNS)

### Mechanism of Action

PTNS is a form of peripheral neuromodulation targeting the posterior tibial nerve, which contains mixed sensory and motor fibers originating from spinal segments L4 to S3. Electrical stimulation of this nerve travels retrograde to the sacral plexus, modulating bladder reflex pathways. This mechanism shares a neurological basis with sacral neuromodulation but accesses the nervous system peripherally rather than via direct sacral nerve stimulation.

### Technique

The procedure involves inserting a 34-gauge needle electrode approximately 3 cm cephalad to the medial malleolus, posterior to the tibia. A surface electrode serves as a ground pad and is placed on the ipsilateral foot. Stimulation parameters typically include a frequency of 20 Hz, a pulse width of 200 microseconds, and an adjustable current ranging from 0.5 to 9 milliamps. Correct placement is confirmed by observing a motor response such as toe flexion or fanning. Each session lasts about 30 minutes. The standard treatment protocol consists of 12 weekly induction sessions followed by maintenance treatments every 2 to 4 weeks indefinitely.

### Efficacy

The SUmiT trial demonstrated a 54.5% response rate with PTNS compared to 20.9% in the sham group at 12 weeks. The OrBIT trial showed PTNS to have comparable efficacy to tolterodine extended-release. PTNS improves urgency, frequency, urgency incontinence, and nocturia. However, ongoing maintenance treatments are required to sustain benefits. Its initial response rate is generally lower than that of Botox or SNM.

### Advantages

PTNS is minimally invasive, office-based, and does not require anesthesia or implantation of a device. It carries no risk of urinary retention and has a low side-effect profile. This makes it suitable for patients who decline injections or implants.

### Disadvantages

The main drawbacks include the need for weekly office visits during the induction phase and ongoing maintenance indefinitely, which can be burdensome in terms of time and cost. Its efficacy is somewhat lower than Botox or SNM, and it may not be practical for patients with mobility or transportation challenges.

### Implantable Tibial Nerve Stimulator (eCoin)

An emerging technology is the eCoin device, a subcutaneously implanted stimulator targeting the posterior tibial nerve. This device eliminates the need for office visits, but clinical trial data are still pending.

<image>Demonstration of PTNS technique showing needle electrode placement near the posterior tibial nerve at the medial ankle with surface electrode and stimulator device</image>

## Comparative Data: Third-Line Therapies

When comparing the three third-line therapies, Botox is administered via cystoscopic injection under local anesthesia, SNM requires a surgical implant with local anesthesia or sedation, and PTNS is performed with a needle in the office without anesthesia. Response rates are approximately 60 to 70% for Botox, 60 to 80% during the SNM test phase, and around 55% for PTNS. Botox effects last 6 to 9 months and require repeat injections, SNM provides continuous therapy with battery replacement every 5 to 15 years depending on the device, and PTNS requires maintenance treatments every 2 to 4 weeks. Botox carries a 5 to 8% risk of urinary retention, whereas SNM and PTNS do not. UTI risk is highest with Botox (15 to 25%), lower with SNM (2 to 5%), and minimal with PTNS. Revision surgery is not required for Botox or PTNS but occurs in 25 to 40% of SNM patients at five years. MRI compatibility varies by device, with newer SNM devices offering full-body MRI conditionality.

| Feature | OnabotulinumtoxinA (Botox) | Sacral Neuromodulation (SNM) | PTNS |
|---|---|---|---|
| Administration | Cystoscopic injection | Surgical implant (2-stage) | Office-based needle electrode |
| Anesthesia | Local | Local/sedation | None |
| Response rate | 60-70% | 60-80% (test phase) | ~55% |
| Duration of effect | 6-9 months (repeat injections) | Continuous (battery 5-15 yr) | Requires maintenance q2-4 weeks |
| Retention risk | 5-8% (may need CIC) | None | None |
| UTI risk | 15-25% | 2-5% | Minimal |
| Revision surgery | None | 25-40% at 5 years | None |
| MRI compatibility | N/A | Device-dependent (newer: full-body 3T) | N/A |
| Best for | Patients willing to self-catheterize | Continuous therapy without repeat procedures | Least invasive; no anesthesia/retention risk |

### ROSETTA Trial (NEJM 2016)

The ROSETTA trial compared 200 units of Botox to SNM for urgency urinary incontinence. Botox showed a greater reduction in urgency incontinence episodes at six months but was associated with a higher UTI rate (35% versus 11%) and a higher rate of CIC (8% versus 2%). Both therapies were effective, and the choice should be guided by patient preference and willingness to perform catheterization.

## Sequencing and Patient Selection

There is no definitive algorithm for the sequence of third-line therapies. The AUA/SUFU guidelines consider all three options equivalent and recommend shared decision-making. Practically, patients willing to self-catheterize and seeking a rapid response may prefer Botox. Those desiring continuous therapy without repeat procedures might choose SNM. Patients who want the least invasive option without anesthesia or risk of retention may opt for PTNS. Failure of one therapy does not preclude success with another, and sequential trials of different third-line therapies are reasonable.

## Clinical Pearls

All three third-line therapies—Botox, SNM, and PTNS—demonstrate comparable efficacy, with response rates around 55 to 70%. The primary counseling point for Botox is the 5 to 8% risk of urinary retention requiring clean intermittent catheterization; therefore, ensuring patient willingness and ability to self-catheterize is crucial before offering this treatment. SNM has evolved to include rechargeable, MRI-conditional devices such as Axonics and InterStim Micro, which allow full-body 3 Tesla MRI scans and should be preferred when available. PTNS is the least invasive option but demands the greatest time commitment due to weekly office visits and ongoing maintenance, making patient compliance a significant barrier. Importantly, failure of one third-line therapy does not predict failure of another, so sequential trials are appropriate. In cases of neurogenic detrusor overactivity, such as spinal cord injury or multiple sclerosis, the Botox dose is increased to 200 units, and SNM efficacy may vary depending on the lesion level.

## References
- AUA/SUFU Guideline on Diagnosis and Treatment of Overactive Bladder, 2019 (amended 2023)
- EAU Guidelines on Urinary Incontinence, 2024 Update
- Amundsen CL, et al. "OnabotulinumtoxinA vs sacral neuromodulation for urgency urinary incontinence" (ROSETTA Trial). *NEJM*. 2016;375(12):1445-1454.
- Peters KM, et al. "Randomized trial of percutaneous tibial nerve stimulation versus sham" (SUmiT Trial). *J Urol*. 2010;183(4):1438-1443.
- Peters KM, et al. "Percutaneous tibial nerve stimulation vs tolterodine ER" (OrBIT Trial). *J Urol*. 2009;182(3):1055-1061.
- Campbell-Walsh-Wein Urology, 12th Edition, Chapter on Neuromodulation