Hypothesis / aims of study
Patients with certain types of urinary incontinence still have unmet clinical needs. These include patients with mixed urinary incontinence (MUI) failing first-line therapies and patients with urgency urinary incontinence (UUI) failing second line therapies (SNM ± botulinum toxin). Pudendal nerve stimulation (PNS) may be a new treatment for these patients. We present safety, clinical and urodynamic data of 13 patients participating in an ongoing exploratory study of adaptive pudendal nerve stimulation using the Amber UI system.
Study design, materials and methods
Adult women with refractory MUI to first-line treatments (n = 8) or severe refractory UUI (SNM ± botulinum toxin) (n = 5) were implanted with the Amber UI system incorporating an implantable pulse generator (IPG) and two quadripolar electrode leads. The system has an embedded sensor and algorithm that enables adaptive stimulation with voluntary and automated control. Under general anesthesia, leads were accurately placed on the trunk and anterior pudendal nerve (PN) using radiological guidance and intraoperative electromyography (EMG) of the pelvic floor muscles (PFM) and external anal sphincter (EAS). Monophasic (pulse duration: 200µs) stimulation at 15Hz (basal mode) could be adapted to 40Hz (adaptive mode) by an embedded patient-actuated tap (accelerometer) function. Outcomes at 6 months included: safety (primary outcome), surgical feasibility, physiological (inc. urodynamic testing) and pilot clinical efficacy (voiding diaries; questionnaires).
Results
All 13 patients were implanted without technical difficulties. Intraoperative PNS showed EMG activity of the PFM and EAS with slight variability between lead and electrode combinations. Stimulation on different parts of the PN showed isolated PFM activity, concurrent PFM and EAS, and EAS activity alone [Figure 1].
One patient is still in the trial and did not complete the six-month follow-up visit, and one patient exited the trial for reasons unrelated to the device or therapy. Six-month data were available for 11 patients (n = 5 rUUI; n = 6 MUI). In MUI patients, 5-day voiding diary data showed complete continence (dry) in 4 patients (baseline 5.2 ± 2.7 incontinence episodes/day). One patient reported a 90% improvement in incontinence episodes/day (baseline 4.2 vs. 0.4 at 6 months). The remaining patient had a good initial response but this dropped to 23% at 6 months after one lead was switched off due to discomfort. In the rUUI group, one patient was dry at final follow up visit (3.6 to 0) and two showed >50% reduction in incontinence episodes (5.7 ± 0.1 to 1.7 ± 1.3).
All patients engaged with the embedded IPG inertial detection function (accelerometer) with many using this many times during the day in response to feelings of urgency. Further, relevant physiological biomarkers of pelvic neuromuscular activity could be detected real-time, in vivo, and classified to execute a control policy for future implementation of a fully adaptive (closed-loop) stimulation algorithm. [Figure 2].
PNS during awake urodynamics at 6 months showed a significant increase in mean baseline maximum cystometric capacity (MCC) with basal stimulation, and further increments with adaptive stimulation: baseline MCC: 198.4ml, 95% CI [121,275]; basal stim: 326.8ml, 95% CI [214,440] (p<0.01); adaptive stim: 361.4 ml, 95% CI [245,478] (p=0.0017).
Patients reported clinically meaningful improvements in quality of life with a mean decrease of 51 ± 9 points on the QoL-OAB questionnaire (p=0.0003), and 10 ± 2 on the ICIQ-UI-SF questionnaire (p=0.0003). All patients reported a positive global impression of improvement (PGI-I) score (6 reported a score of 1: very much better).
Interpretation of results
Our study shows that PNS through an implantable neurostimulator is safe and feasible in humans. The successful implantation of the device without technical difficulties highlights its potential value as a future scalable treatment option. Specific EMG findings associated with PNS provide confirmation of correct placement and also confirm the historical principle (observed experimentally) that reflexes mediated by sphincter contraction have a key role in bladder control.
Unmet clinical needs in patients with MUI and rUUI were addressed and results were particularly promising in patient groups where existing treatment options had already been exhausted. The positive results from voiding diaries were confirmed by QoL questionnaires. The introduction of a patient activated adaptive stimulation mode shows promise and contributes to the positive clinical effect. The tap-induced activation gives a decrease in urgency and additional time to reach a toilet in time. Future fully closed-loop stimulation will be the next step to support patients at any time during the day.
In addition to its effect on incontinence, continuous PNS also shows that it has both immediate and sustained positive effects on bladder capacity. An increase in MCC has not been demonstrated in humans by other types of neuromodulation to date.