To report long term results and complications of artificial urinary sphincter (AUS) in children for stress urinary incontinence (SUI) due to intrinsic sphincter deficiency (ISD).

We performed a retrospective, monocentric study. All patients under 18 years old that underwent an AUS (AMS 800) implantation between 1986 and october 2018 were included. 
Pre-operative data (underlying disease, severity of SUI) were reported, as well as per and post operative complications. Complete continence was defined as 0 pad/day. Explantation-free and revision-free device survival was estimated by the Kaplan-Meier method.

Overall 37 patients with a median age of 12 years [IQR 9-16] were included (15 females, 22 males). Median follow up was 18,7 years (IQR 0,3-31,4). 
25 had a neurogenic ISD, mostly from spinal dysraphism, 5 had a congenital sphincter agenesis, 3 had a post-prostatic rhabdomyosarcoma ISD and 2 an epispadias. 1 patient had a gunshot pelvic trauma and 1 a pelvic surgery for Hirschsprung syndroma, both resulting in ISD.
In patients with neurogenic bladder, the main preoperative voiding mode was spontaneous voiding (n=35, 95%), often with abdominal thrust (or Crédé). 2 patients practiced clean
intermittent catheterization (self-catheterization or by caregiver).
All patients had a preoperative urodynamic evaluation assessing the ISD before device implantation.
48% had previously undergone surgery, mostly for vesico-ureteral reflux or cryptorchidism. 
All implantation were performed in open approach except for one young female patient (robotic approach).
In male, the cuff implantation was pericervical in 2/3 cases (n=25). For the remaining patients, the cuff was located in a bulbar position. For all female patients, the cuff was implanted around the bladder neck. 
The median size if the cuff was 6cm (IQR 4-7,5). In 81%, a 61-70 cmH2O balloon was implanted.
5 patients experienced early complications (< 30 days after implantation) : 3 patients had a urinary tract infection, 1 leading to device explantation 21 days after implantation. 2 had acute urinary retention, resolved after few days of cathetirization.

At last follow-up, 83,3% (n=30) still had a functional AUS in place.
10 (27%) patients underwent explantation (appendix 1) due to erosion or infection. Only 4 patients had a second device implantation after explantation of the first one. Among those patients, 3 underwent explantation of the second device. 
16 (43%) underwent revision (appendix 2), with a median delay of 114 months (IQR 0-223).
10 patients have had an augmentation cystosplasty for compliance degradation non-responding to anticholinergic treatments.
18 (49%) patients were spontaneously voiding, 17 patients were practicing self-catheterization, 1 underwent an ileal urinary derivation and 1 had an indwelling catheter.

At last follow up, the overall continence was satisfaying, with 30 patients (81%) requiring 0 pad/day, 4 patients requiring 1 pad/day and only one who was fully incontinent.
As for the patients still having a functional AUS, 27 (90%) patients wore 0 pad/day and 3 (10%) patients wore 1 pad/day.

AUS implantation remains a trigering intervention, aspecially for children. To our knowledge, this study reports the longest follow-up of patients with AUS implanted during childhood. 
Erosion was the most common complication as reported in litterature.
We observed a relatively high rate of revision, that may be related to the very long follow up of our patients.

Artificial urinary sphincter AMS 800 is a long term effective surgical treatment for SUI related to ISD in children. Its implantation remains challenging especially in this population of patients who often had undergone previous bladder neck surgery.
However, AUS provides a long term high rate of continence, and often a high rate of satisfaction, even if revisions are often necessary.