Artificial urinary sphincter (AUS) implantation is the gold-standard surgical treatment for stress urinary incontinence (SUI) caused by intrinsic sphincter deficiency (ISD) in women. Robot-assisted AUS implantation has been recently described and several techniques have been reported with anterior or posterior approach of the bladder neck(1,2). This technique still remains challenging in particular regarding bladder-neck dissection. The objective of this study is to report the early experience of a modified technique of robot-assisted AUS implantation in women with a posterior approach to the bladder neck and intraoperative real-time cystoscopic control.

We reviewed the records of patients who underwent a first robot-assisted AUS implantation between 2017 and 2019. All of the operations were performed with the Da Vinci Xi®, by transperitoneal approach and by posterior dissection of the bladder neck, in order to avoid the "blind" anterior dissection. Dissection begins between the bladder and the vagina, using a vaginal valve to expose the anterior vaginal fornix. When the dissection was low enough, a lateral dissection was initiated. The next step was to open the umbilical-retropubic space anteriorly and carry out the dissection until the right and left endopelvic fascia. Only a thin plane remained between the anterior and posterior dissection. Using the Maryland or the Cadiere forceps, a passage was made from back to front to the left and then to the right in order to exactly position the cuff sizer around the bladder neck. Real-time intraoperative cystoscopic monitoring was systematically carried out to check the correct level of dissection (sub-trigonal, at the bladder neck) and to avoid any injury to the bladder neck when moving from back to front. This endoscopy was performed with a 70° cystoscope, and monitoring was made possible by the simultaneous display of the endoscopy and laparoscopy in the surgeon console.
Perioperative and intraoperative data, functional outcomes and complications were reported. Continence was defined as 0 to 1 “confident” pad per day. Functional outcomes were also assessed with the USP-questionnaire, the PGI-I score and the satisfaction rate.

During the period of this study, 25 patients were included. The median age was 67 years (IQR:61-74). 24 patients (96%) had a history of pelvic surgery (80% SUI surgery, 56% prolapse surgery). The median preoperative maximum urethral closure pressure was 26cmH2O (19-37). 
The median operative time was 263min (221-349). 2 conversions (8%) to open surgery were required: 1 for adhesion and significant pelvic fibrosis and 1 for poor respiratory tolerance of pneumoperitoneum. 9 intraoperative complications were reported: 5 vaginal injuries (20%), 2 bladder neck injuries (8%) (including one leading to non-implantation of AUS) and 2 urethral injuries (8%). A cystotomy was performed in 5 cases (20%), in order to better visualize the bladder neck and ureteral meatus. The median size of the implanted cuff was 7.5cm (7.5-8). The median hospital stay was 3 days (2-7). 
With a median follow-up period of 15 months (13-26), 21 patients (84%) had their initial AUS in place, 3 AUS (12%) were removed for infection and/or erosion (at 1 month, 3 months and 12 months), and 1(4%) was revised due to persistent SUI (implantation of a smaller cuff).
Among the patients with an AUS in place (n=21), 15 patients (80%) were continent. The median satisfaction rate was 90% (80-100). The median USP score for SUI was 0 (0-1.5) for SUI, 5 (1.-7.7) for overactive bladder and 1 (0-3) for voiding symptoms.  A PGI-I score of 1 was reported in 80% of patients.

Whatever the surgical approach (open or robotic), the circumferential dissection of the bladder neck still remains challenging, and if damage is done to it early erosion may result. The posterior approach, recently described(2), provides a safe dissection, without any “blind” passage at the posterior side of the bladder neck, and frees the surgeon from having to rely on the digital vaginal control of an experienced assistant. However, with this posterior approach, it may sometimes be difficult to locate the lower point of the dissection, but can be helped with intraoperative cystoscopic monitoring.
Thanks to this technique, we observed a low rate of bladder-neck injury, only one of which obliged us to stop the procedure, and no ureteral injury occurred. 
However, 3 explantations related to erosion occurred. 
This study has several limitations: its retrospective design and the short follow up period.

We have reported here on our early experience with robot-assisted AUS implantation in women with the posterior approach to the bladder neck and with real-time cystoscopic monitoring.  This procedure appears feasible, reproducible and ensures proper cuff positioning and with a lower risk of bladder or cervical injury. Medium-term functional results appear satisfactory. Further studies with long-term results seem necessary to evaluate the benefits of this compared to robot-assisted implantation with anterior bladder neck approach or open surgery.

Peyronnet B, Capon G, Belas O et al. Robot-assisted AMS-800 artificial urinary sphincter bladder neck implantation in female patients with stress urinary incontinence. Eur Urol 2019;75:169-175
Gontran-Tellier B, Boissier R, Baboudjian M. Robot-assisted implantation of an artificial urinary sphincter, the AMS-800, via a posterior approach to the bladder neck in women with intrinsic sphincter deficiency. BJU Int 2019;124(6):1077-1080