Role of endoscopic management in synthetic sling/mesh erosion following previous incontinence surgery: a systematic review from European Association of Urologists Young Academic Urologists (YAU) and Uro-technology (ESUT) groups

A systematic literature search was conducted according to the Cochrane review and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol [22, 23]. Medical subject headings (MeSH) terms used, but not limited to, were: ‘endoscopic’, ‘laser’, ‘holmium’, ‘mesh’, ‘erosion’, ‘sling’, ‘TVT’, ‘tension-free vaginal tape’, ‘transobturator tape’, ‘TOT’, ‘urethral’, ‘transurethral’, ‘resection’, ‘removal’, ‘incontinence’, ‘surgical’, ‘incision’ and ‘intervention’. Boolean operators (AND, OR) were used to refine the search. The search strategy was conducted to find relevant articles from CINAHL (1996–December 2018), EMBASE (1996–December 2018), Ovid Medline (1996–December 2018), Cochrane Library (2018), Scopus (1996–December 2018), Clinicaltrials.​gov, Google Scholar and individual urological journals. The search was confined to all English language articles between January 1996 and December 2018. All original studies were included and two reviewers (S.S and B.K.S.) identified all studies independently; discrepancies were resolved by mutual consensus. Where additional information or clarification was needed, the primary authors of the studies were contacted directly.

The PICO statement for this review is as follows: the population examined was adults who underwent synthetic sling procedures for incontinence. There was no comparative group, and the outcomes were to ascertain the results of endoscopic management following sling erosion.

The following variables were extracted from each study: journal and year of publication, study type, time period, presenting symptom, procedure type, time to presentation following initial surgery, location of mesh erosion, complications and their management, success rate and follow-up. We used the standard IUGA/ICS terminology for reporting mesh-associated complications in this manuscript [24]. Data were collated using Microsoft Excel (Microsoft Corp., Redmond, WA, USA), version 12.2.4.). There was a substantial degree of heterogeneity between studies in both design and reporting, and hence our analysis was limited to narrative synthesis and pooled analysis of mean results.

All of the included articles were retrospective case series and reported on 3–23 patients each. The mean age across studies varied from 48 to 68 years, and 149 (75%) had tension-free vaginal tapes (TVT) or transobturator tapes (TOT) as their initial procedure. The journal, year of publication, level of evidence and time to presentation after their original surgery are reported in Table 1.

Voiding lower urinary symptoms (LUTS) was the most commonly reported symptom (17 studies). Other symptoms included recurrent UTIs (11 studies), haematuria (9 studies), SUI (6 studies), urge incontinence (UI) (5 studies), dyspareunia (7 studies), dysuria (4 studies) and pelvic pain (3 studies). Mesh-associated stones were a common finding across the literature in patients with urinary mesh erosion. The median time to presentation after their initial surgery was 34.5 months (range: 1–247 months).

The site of mesh erosion was the bladder in 134 patients (68%), of which 12 (6%) were in the bladder neck. Urethral mesh erosion was seen in 63 patients (32%) across all studies.

One hundred eight (55%) patients were treated exclusively with laser incision (Table 2). The initial and final success rate with this technique was 67% (72 patients) and 92% (99 patients) respectively. Of these patients, 27 needed further laser procedures, with 18 (17%) needing one additional procedure, 6 (5.6%) needing two additional procedures and 3 (2.8%) needing multiple (> 2) procedures. Of the remaining nine patients, eight (7%) needed vaginal surgery to remove a mesh remnant and one needed open cystotomy to remove the eroded mesh.

Post-operative complications were seen in 26 (24%) patients. These included SUI in 23 (21%), urinary tract infection in 2 (1.9%) and voiding difficulty in 1 patient. There were no other reported major complications such as fistula formation or intra-operative perforation.

Ninety (45%) patients were managed using TUR with an electrode loop or laparoscopic scissors/forceps. The initial and final success rates with this technique were 80% (72 patients) and 98% (88 patients) respectively. Of these patients, 16 needed further endoscopic procedures, with 10 (11%) needing one additional procedure, 4 (4%) needing two additional procedures and 2 (2%) needing multiple (> 2) procedures. The remaining two patients needed open surgery to remove the eroded mesh.

Post-operative complications were seen in 25 (28%) patients. These included SUI in 19 (21%), vesicovaginal fistula in 3 (3%), intra-peritoneal bladder perforation needing a further procedure in 2 (2%) and retroperitoneal bladder perforation in 1 patient.

While there is increasing concern about the growing numbers of mesh or sling erosions, all types of surgical intervention carry a risk of morbidity. Bearing in mind the limitations of our review, which is based on retrospective case series, endourological techniques show excellent success rates of > 90% with endoscopic techniques using laser excision or endoscopic resection, although 20–25% patients may need more than one endoscopic procedure to achieve this. Considering the lack of an open comparator group and reporting bias, the real success rate could be much lower. While the complication rates seem to be 24–28%, one would argue that the majority of these were SUI, which is expected as an outcome of mesh removal irrespective of whether it is done endoscopically or as an open approach, and this may not really count as a complication but probably as an expected outcome of the mesh removal procedure.

Polypropylene meshes play a huge role in the minimally invasive repair of pelvic organ prolapse (POP) and SUI [28]. However, this also presents complications of mesh misplacement, migration or erosion [4]. FB erosion following urinary incontinence surgery can be a late complication and may be associated with the material used, operating surgeon’s expertise and technical difficulties at the time of initial surgery [4, 13]. Unrecognized bladder perforation during trocar insertion can also cause mesh erosion and may present as a late complication [3].

Often presenting with voiding LUTS, recurrent UTIs and haematuria, other symptoms identified in our review included SUI, urge incontinence (UI), dysuria and dyspareunia. The median time of presentation after surgery was 34 months, and a low index of suspicion is necessary in these patients with a previous history of incontinence-related procedures. This delay in presentation can be due to vague non-specific symptoms and a lack of standardized follow-up [29]. A high index of suspicion is often needed, and a detailed evaluation is necessary often with a cystoscopy to examine the LUT, which can help to confirm the diagnosis. Mesh-associated stones are not uncommon and the presence of stones should raise a strong suspicion of mesh erosion.

Our review of the literature, which is the largest to date, shows that endoscopic management is a feasible and effective option in managing FB erosion. Holmium laser was associated with an overall success rate of 92% and endoscopic management with resectoscope/laparoscopic scissors had a success rate of 98%. The most common complication was SUI in both groups; however it was largely unclear whether this was de novo or a pre-existing complaint. SUI following removal of mesh surgery can be due to a failed incontinence procedure but equally mesh erosion can damage the urethra and sphincter resulting in scarring. The management of this can be complicated and may need reconstructive surgery with urethroplasty to help improve functional outcome [4]. Endoscopic management with resectoscope and electrode loop and the use of laparoscopic scissors was associated with vesicovaginal fistula in three patients and bladder perforation in two patients. It is worth noting that the laser approach allowed for decreased post-operative morbidity and complications, decreased inpatient hospital stay and good functional outcomes [7]. However, the outcomes of the two approaches may not be directly comparable because of different populations, variation in techniques and different surgical expertise.

The duration of follow-up varied across all studies and ranged from 1 to 134 months. We suspect it is important for all patients to have long-term outpatient-based follow-up and perhaps have a vaginal speculum examination along with a flexible cystoscopy 6–12 months post-operatively to pick up any early signs of mesh erosion [30]. This could be repeated if there are any obvious symptoms or a clinical suspicion of erosion. Presence of bladder or urethral stones especially if adherent to the musoca suggests an erosion. The patients should also be counselled to return for a review if they develop persistent or worsening urinary symptoms. Patients at high risk of potential mesh-related complications seem to be those who have a combined mesh for POP repair and SUI mesh sling [30]. In a study of 41,604 women in the US who underwent transvaginal mesh repair for POP with or without concurrent sling for SUI, transvaginal POP repair without mesh but with concurrent sling or a sling for SUI only and a follow-up period of 1-year, the risk of erosion was 2.7% and the risk of repeated surgery with concomitant erosion was 2.1% in the POP repair with mesh plus sling group [30]. The SUI sling group on its own had an erosion rate of 1.5% and a risk of repeated surgery with concomitant erosion of 1.6%. Similarly, concomitant hysterectomy and hypertension were also associated with mesh erosions [31]. In high-risk patients the index of FB erosion and early counselling and appropriate management should be made available.

In this review we adhere to the methodological approach of the Cochrane guidelines and PRISMA checklist. It summarizes the role of endoscopic management of mesh erosions including the use of both laser and transurethral resection and gives an overview of the success rate and complications with this technique, although all papers included in our review were retrospective case series and hence prone to bias. Furthermore, there was a lot of heterogeneity in the reported studies making it challenging to do a formal meta-analysis. There are some limitations to our systematic review. Individual studies did not have a standardized management or follow-up, and this was left to the individual surgeon and centre based on their expertise and patient presentation. Given that data collection and reporting were not standardized either, it was difficult to compare or combine outcomes.

A recent study looked at the reporting outcome measures in trials on synthetic mesh procedures for POP and concluded that urgent action is needed to improve the quality of research in this field [32]. Of the 71 randomized trials, 24 different types of mesh were identified. These trials reported on 110 different outcomes and 60 outcome measures. Clinically important measures such as erosion, pain and dyspareunia were reported in 40, 29 and 25 trials respectively. They recommend developing and implementing a minimum standardized data set, which forms the core outcomes of this procedure. Another systematic review on the type of synthetic material used found that polyester sling material caused the highest rates of vaginal erosion [33]. The US Food and Drug Administration (FDA) has ordered all manufacturers of surgical mesh intended for transvaginal repair of prolapse (cystocoele) to stop selling and distributing the product [34]. Given the scale of mesh erosions and the public outcry associated with it, perhaps a national registry of all sling or mesh procedures should be the way ahead, where every case is registered and the core outcome measures including the type of material used, surgical technique, complications and follow-up are mandatory. This should be protocol based and have outcome measures which are standardized and hence comparable.