The main complications related to surgery with mesh are the exposure of the material and the functional complications such as pain, the voiding obstruction, and dyspareunia. These complications can have a serious functional and psychological impact and lead to surgical revisions. The UK authorities have recently decided to suspend the use of meshs for incontinence or  genital prolapse surgery.
Materiovigilance can not accurately and comprehensively estimate the incidence of complications related to the use of these medical devices because of a lack of knowledge of the number and type of surgeries with mesh and the other hand of probable underreporting of the latest complications; some exposures are observed more than 5 years after. Randomized trials that rely on the most trained teams have the risk of underestimating the occurrence of rare complications.
Since 2017, the VIGI-MESH Observatory has been providing a prospective and detailed collection of the use of these medical devices in the surgery of urinary incontinence and prolapse as well as complications. The objective of this first analysis is to describe the method of collecting information and to report our first results about the incidence of complications.

Participation was offered to any woman operated for stress urinary incontinence, or pelvic organ prolapse. We included mid-urethra sling (MUS) or colposuspension, vaginal repair surgery with or without mesh, and abdominal or laparoscopic repair surgery. Each women included was informed and gave their consent.
Each surgeon had to declare his own surgical procedures using a specific sheet detailing the characteristics of the patient) and the surgical procedures implemented. The complications were also the subject of a specific declaration using another sheet describing the complication and its management. The exhaustive nature of the collection was controlled by the nominal delivery declarations of the devices, and the surgical procedure codes collected by the medical information department (MID) of each establishment. 
The severity of complications was graded according to Clavien-Dindo's classification. Only serious complications (grade III and above) have been considered. Failure or recurrence of incontinence or prolapse was not considered as a serious complication.
We chose considered five surgical groups: MUS alone, MUS associated with surgical treatment of prolapse, vaginal repair with use of mesh, vaginal repair without mesh, and laparoscopy with mesh (with or without associated Burch colposuspension). A Kaplan-Meier survival curve was constructed to estimate the incidence of complication by surgical group.
The complications were described according to the initial surgical group, the nature of the complication, the delay since surgery, and the management of the complication.

Between February 2017 and August 2018, we included prospectively 1873 women. Index surgery (at baseline) was MUS only in 658 women, prolapse surgery in 1062 (including 888 with at least one mesh), and incontinence associated with prolapse surgery in 153 women (all with at least one mesh); associated hysterectomies (total or subtotal) were not counted as a prolapse correction procedure (Table).
Mean age was 62 (29 to 93), 75.9% were postmenopausal; 45.9% were sexually active; mean BMI was 26 kg/m² (16 to 61); 9.2% were smoker, 6.8% were diabetic, 17.5% has already undergone an hysterectomy, and 20.4% a surgery for SUI or POP; the physical status (ASA score) was rated 1 for 628 women (33.5%), 2 for 950 (50.6%), 3 for 143 (7.6%), and 4 for 2 (0.1%). Mean length of follow-up was 7.3 months (0 to 18). 
Among the 1873 women included in this analysis, 52 (2.78%) experimented a serious complication during the surgery or the first months of follow-up. The incidence of serious complication 6 months after the surgical procedure was estimated around 0.99% [0.12-1.86] after laparoscopy with mesh (blue curve), 3,54% [95% CI: 2.03-5.04] after MUS alone (red curve), 7.03% [2.79-11.28] after MUS associated with POP surgery (green curve), 1.73% [0-3.77] for vaginal repair without mesh (purple curve), and 2.80% [0.86-4.65] after vaginal repair with mesh (orange curve; Figure). Most were Grade IIIb complications (surgical revision under general anaesthesia), one case need resuscitation, and no death occurred. The most frequent complication was obstructive micturition (18 cases all related to MUS, 2.22%) and the second one mesh exposure (9 cases related to MUS, 1.11%; 5 to vaginal mesh, 1.28%; and 2 to laparoscopic mesh, 0.33%).
Thirteen complications occurred within the first 48 hours, all related to vaginal surgery: bladder (eight cases) or urethral (two cases) wounds, bleeding or thrombus (four cases), and complete urinary retention. The vaginal placement of the mesh was discontinued eight times because of the complication. Three hemorrhages required a return to the operating room for secondary hemostasis. The mesh was removed on day 1 vaginally in one case because of bladder exposure. One MUS was vaginally loosened on day 1 due to bladder retention in another case.
Twenty-four complications were treated between 2 days and 2 months after index surgery. It was necessary to loosen the MUS in twelve cases, and to cut it in two cases. In three cases (one sacrocolpopexy by laparoscopy and two vaginal surgery with mesh) it was necessary to manage a painful ureteral obstruction. In three cases it was necessary to evacuate a hematoma or resume hemostasis in the operating room. In two cases it was necessary to partially remove the MUS because of early vaginal exposure. In two cases the mesh was removed by laparoscopy due to pelvic infection (one sacrocolpopexy and one rectopexy).
Fifteen complications were treated between 2 and 12 months postoperatively due to late bladder retention on MUS (one case), vaginal exposure (six MUS, four vaginal repair with mesh, and two sacrocolpopexy), urethral exposure of MUS (two cases), and painful subcutaneous swelling on a retropubic MUS (one case). Complications requiring a division of the MUS (one case), a partial resection of the MUS (seven cases) or mesh (one vaginal mesh, and two related to sacrocolpopexy), and in the four other vaginal trimming.
Taking into account all revisions for complication, 11 MUS out of 811 were partially or totally removed (1.36%), as well as 2 mesh among 391 vaginal repair (0.51%), and 4 mesh among 611 laparoscopic prrocedures (0.65%). We observed 29 surgical revisions (1.55%) for urinary incontinence (24 cases) or prolapse (5 cases); the most frequent wass the placement of a MUS after sacrocolpopexy by laparoscopy (14 cases or 3.59%). Finally, 69 women (3.68%) were reoperated at least once for complication, urinary incontinence or prolapse.

Intraoperative complications were mainly intraoperative urinary wounds during vaginal surgery leading to discontinuation of mesh placement. Half of early complications were voiding dysfunction after MUS. And later complications are mostly vaginal exposures of MUS or mesh that have been placed vaginally or, less often, by laparoscopy. Complications are more common in case of surgery that combinate procedure for urinary incontinence and procedure for prolapse.

It seems essential to continue a long-term vigilance after surgical pelvic floor repair when a mesh is used. 
Our observatory will allow us to make comparisons between the different types of materials and different surgical methods, to identify those who may pose problems. These issues are of interest for patients, clinicians, manufacturers, and health authorities.