Complications of Synthetic mid-Urethral Slings
By Christian Twiss, MD and Shlomo Raz, MD
Issues in Incontinence: Spring/Summer 2008
Due to its efficacy, safety, and relative simplicity, the synthetic mid-urethral sling procedure has emerged as one of the mainstays
of surgical therapy for female stress urinary incontinence. The transobturator approach to placing mid-urethral slings has recently been marketed as safer than the retropubic approach due to avoidance of entry into the retropubic space. However, accumulated experience has demonstrated that significant complications are possible with both techniques. The purpose of this review is to summarize the rates, etiology, and management of the most common complications encountered with synthetic mid-urethral slings and to compare, based on recent evidence, complication rates of the retropubic and transobturator approaches to sling placement.
Bleeding
One of the difficulties in assessing bleeding complications of synthetic sling placement is that “bleeding” as a complication has no firm definition or criteria. There is great variability in the literature in reporting and assessing the degree of bleeding, including intraoperative versus postoperative bleeding, bleeding requiring reoperation, bleeding requiring transfusion, hematoma size, volume of blood loss, and major vascular injury. Bleeding can occur from the vaginal incision, from the periurethral dissection, from vessels in the retropubic space, or, in rare cases, from overt injury to the iliac or obturator vessels. When placing retropubic slings, a common source of troublesome bleeding is inadvertent deviation of the trocar within the retropubic space, either too close to the urethra, leading to bleeding from the periurethral venous complex, or too lateral, leading to venous bleeding from vessels on the pelvic sidewall. Careful control of trocar placement reduces the potential for bleeding from these sites.
One of the largest series to report bleeding after retropubic placement of syntheticslings is an Austrian registry of 5578 tensionfree vaginal tape (TVT) procedures,1 which showed an overall 2.7% rate of bleeding complications with 0.8% requiring reoperation, primarily via laparotomy. Other large TVT series 2-3 report similar rates of hemorrhage/hematoma, between 1.9% and 2.7%. However, when routine pelvic magnetic resonance imaging (MRI) was performed after retropubic sling procedures, 25% of patients were found to have clinically unsuspected retropubic hematomas.4 From these data we can conclude that significant bleeding is possible after retropubic sling placement, albeit that the reported rates in the large studies suggest that major, clinically overt bleeding is uncommon.
More recent evidence suggests that bleeding is significantly reduced with the transobturator approach to mid-urethral sling placement. Several large series 5-8 fail to report any significant bleeding complications with the technique. The meta-analysis by Sung et al 9 of randomized controlled trials comparing retropubic with transobturator slings found that the rate of significant hematoma was 1.6% for the retropubic approach compared to 0.08% for the transobturator approach. Similarly, in the recent metaanalysis by Novara et al, 10
the odds ratio of pelvic hematoma was 4.83 for retropubic as compared to transobturator slings. However, despite this evidence, significant bleeding, including iliac and obturator vessel injury, has been reported 11-12 with use of the transobturator technique. Thus, although there is a consensus that transobturator sling procedures carry significantly less bleeding risk compared to retropubic mid-urethral sling procedures, transobturator sling placements are by no means immune to bleeding complications, and careful attention must be paid to proper technique.
Perforations and erosions
Perforations occur when the surgeon inadvertently places the sling material into an unwanted anatomical position, typically into the urinary tract (bladder, bladder neck, or urethra) or through the vaginal wall. Perforations are typically considered an intraoperative complication, recognized and corrected at the time of surgery. Erosions are typically detected in the postoperative period when sling material is found in the urinary tract or noted to be exposed in the vagina. While erosions are often considered a postoperative complication, it is important to be cognizant that many postoperative “erosions” are actually the result of undetected perforations that occurred at the time of surgery.
Bladder perforation is one of the more common intraoperative problems encountered with retropubic midurethral sling placement, occurring in 2.7%–6% in large series. 2, 13-16 Bladder perforation during retropubic sling placement occurs more commonly in patients with a history of anti-incontinence procedures,13 likely due to scarring within the retropubic space. Bladder perforation occurs less frequently with the transobturator approach compared to the retropubic approach, 9,10,17 with reported rates of bladder perforation between 0% and 1.5% with this technique. 7,9,17,18
However, the majority of bladder perforations, when recognized intraoperatively, are relatively minor complications and are often corrected intraoperatively by repositioning and re-passage of the trocar and sling material,16 typically followed by a period of bladder drainage with a urethral catheter.
Urethral perforation remains rare for both retropubic and transobturator approaches, but it can occur with either.2,7,16,19 As opposed to the case of bleeding and bladder perforation, the evidence does not currently suggest that urethral perforation occurs less commonly with the transobturator approach compared to the retropubic approach. Intraoperative urethral injury has been reported in 0.07%–0.2% of retropubic sling cases 2,16 and 0.1%–2.5% of transobturator sling cases.7,20,21 Since the urethra is a high-pressure system, this complication typically requires a formal repair of the damaged area and is somewhat more serious than bladder perforation. After repair, the choice of placing or not placing a sling depends upon the severity of the perforation and the comfort level and judgment of the surgeon. Regardless, as with bladder injury, a period of urethral catheterization is recommended during the healing period.
When sling material is found in the urethra in the postoperative period, it is termed a urethral erosion. The risk factors for postoperative urethral erosion include undetected urethral perforation, excessive sling tension, and postoperative urethral dilations.19 Treatment involves removal of the offending material and formal urethral repair. Unfortunately, both in our experience and in that of others, recurrent incontinence (stress and/or urge incontinence) is common after intervention in these cases.19,22
Bowel perforation is a potentially lethal complication that typically occurs when the retropubic approach is utilized in patients with bowel adhesions in the retropubic space due to prior abdominal surgeries. 23,24 Bowel perforation is rare, reported to occur in 0.03%–0.7% of retropubic sling cases. 13,23,24 The risk of this complication is greatly reduced with use of the transobturator approach because it specifically avoids entry into the retropubic space. We agree with other authors that the transobturator approach should be considered in patients who are at particularly high risk for bowel adhesions in the retropubic space. Treatment for bowel perforations is directed at resolving associated sepsis, removal of the sling, and resection and/or repair of the damaged segment of bowel.
Postoperative vaginal erosions are another problem with synthetic slings. Metaanalyses 10,17 of randomized trials comparing retropubic with transobturator synthetic slings fail to demonstrate a significant difference in vaginal erosion rates between the two techniques. Large case series 2,23,25,26 of TVT and SPARC-sling procedures report that vaginal erosions are rare, occurring in 0.2%–1.8% of cases. While some authors 27 have suggested slightly higher rates of vaginal erosions with transobturator slings, this may reflect the high erosion rates (6.1%–20%) noted to occur with the Mentor ObTape kit, 28-30 which has subsequently been removed from the market. More recently, a large French registry 20 of 984 TVT-O (inside-out transobturator approach for TVT) procedures reported a postoperative vaginal erosion rate of 0.6%, a figure within the range found in the large TVT and SPARC series.
These results bring to light that the more important factor in preventing vaginal erosions appears to be choice of synthetic sling material. Soft, woven, monofilament polypropylene mesh with large (> 75 microns) pore size appears to be the mesh of choice, as it allows proper incorporation of the mesh into host tissue and facilitates immune surveillance. 31 Both multifilament meshes and monofilament meshes with small pore sizes allow passage of microbes but not host macrophages, thereby promoting infection and subsequent erosion. 31
Management of vaginal erosions consists of either complete excision of the mesh, excision of the exposed portion of mesh, or observation. While several authors 32-34 have reported on their experience with mesh erosions, there is currently no evidence-based consensus on the management of vaginal erosions. We typically attempt complete mesh removal in cases where the mesh is infected or traverses the urinary tract, the mesh composition (multifilament, small pore size) is unfavorable and likely to re-erode or become infected, or the mesh is causing significant pain due to infection or its location. We perform more conservative excision of the exposed area of mesh followed by closure of the vaginal wall in cases where the mesh is uninfected, non-painful, and composed of favorable material. Others 35 have adopted a similar management scheme for vaginal erosions. Some authors 34 have also reported successful conservative observation and spontaneous healing of small, non-infected erosions.
Groin and thigh complications
Groin and thigh complications are significantly more prevalent with transobturator slings than retropubic slings and can be lifethreatening in some cases. A meta-analysis 17 of randomized controlled trials comparing retropubic with transobturator slings found that the odds ratio of groin/thigh pain was 8.3 for transobturator as compared to retropubic slings, and the large French registry of TVT-O procedures 20 reported a 2.7% rate of residual pain lasting greater than 4 weeks duration. In our experience and in that recently published by others, 36 the groin and thigh pain encountered after transobturator sling placement can in some cases be unrelenting and require sling removal, a challenging task. Such pain can result from the passage of the sling through the adductor brevis, adductor magnus, and gracilis muscles (with subsequent myositis); infection and/or abscess; hematoma; or, rarely, obturator nerve entrapment. 12,37 More significant is that serious infectious complications resulting from transobturator slings have been reported,38 including groin and thigh abscesses, sepsis, and gangrene.
Managing groin and thigh complications raises the important issue of “invasiveness” with regard to transobturator slings. While transobturator sling procedures are often marketed as “less invasive” due to avoidance of the retropubic space, one must be cognizant that transobturator slings are placed into an anatomic region that is very difficult to access after the sling is placed. Removal of retropubic slings remains relatively straightforward, especially because urologists and gynecologists are familiar with the anatomy of the retropubic space and urethra. Conversely, removal of a transobturator sling remains challenging because it occupies a deeptissue space that is difficult to access, and the anatomy of this region is far less familiar to pelvic surgeons. Thus, both retropubic and transobturator sling procedures are “invasive,” and each sling carries its own set of problems associated with the anatomic region that it occupies.
Postoperative voiding dysfunction
Voiding dysfunction after sling placement typically includes postoperative urgency/urge incontinence symptoms and/ or obstructive symptoms. The occurrence of voiding difficulties in the postoperative period highlights the importance of determining and documenting the patient’s voiding pattern prior to surgical intervention. For example, persistent urinary urgency after sling placement is typically not considered a complication whereas de-novo urgency symptoms are considered a complication of the surgical intervention.
Recent evidence 17,39 suggests that de-novo urgency after sling placement is relatively common, occurring in 10%–15% of cases; however, the range in the literature 40-43 spans a tenfold difference, from 3.1% to 32%. Recent meta-analyses of randomized trials comparing retropubic to transobturator slings are conflicting. The meta-analysis of Latthe et al 17 suggests that there is no significant difference in de-novo urgency rates with respect to the retropubic as compared to the transobturator approach, whereas the study by Novara et al
suggests fewer storage-related lower urinary tract symptoms with the transobturator technique compared to the retropubic. A recent study by Botros et al 42 suggests that transobturator slings may carry a lower rate of de-novo urge incontinence when compared to retropubic slings.
One hypothesis 42,44 for this difference in postoperative urgency symptoms is that transobturator slings occupy a more horizontal plane as compared to the U-shape of retropubic slings, thereby making less contact with the urethra, resulting in a reduced incidence of de-novo urgency symptoms; however, the evidence supporting this hypothesis is preliminary.
Treatment of de-novo urgency after sling surgery is first directed at ruling out sling erosion into the urinary tract, followed by treatment of urge symptoms including first-line pharmacologic therapy (antimuscarinics) followed by possible second-line therapies (neuromodulation, botulinum toxin) for failures. One must also be mindful that partial obstruction resulting from the sling placement can also result in de-novo urgency symptoms. Thus, a careful history assessing for simultaneous obstructive symptoms (straining, hesitancy, positional voiding, incomplete emptying), examination of the urethra and sling (for evidence of excessive tension or malposition), and urodynamic assessment are important in the evaluation of de-novo urgency symptoms. 45
In cases where obstruction is determined to be the likely cause of de-novo urgency symptoms, intervention to loosen or incise the sling or formal urethrolysis may be required. Success rates for storage-symptom resolution after sling de-obstructing procedures vary widely from 12%–100%, but most studies report success rates in the 75%–85% range. 46 The wide range of success rates may result from the timing of surgical intervention on the obstructing sling; there is evidence 47,48 suggesting that longer periods of obstruction are associated with higher rates of persistent storage symptoms.
Postoperative urinary obstruction can be a serious complication after sling surgery, and the incidence remains difficult to assess primarily due to the fact that the diagnosis is not always straightforward. While complete urinary obstruction requiring indwelling or intermittent catheterization is obvious, the diagnosis of partial obstruction can be problematic because obstructive symptoms are often accompanied by urgency symptoms, making the clinical assessment difficult, and because there are no firm urodynamic criteria for diagnosing urinary obstruction in women. We agree with other authors 49 that the single most important element in making the diagnosis is the temporal relationship between the onset of new obstructive voiding symptoms and the sling surgery. These symptoms may or may not be accompanied by some objective evidence of partial obstruction such as elevated postvoiding residual volume. This highlights the importance of defining the patient’s voiding pattern prior to surgery so that postoperative alterations can be recognized.