Multifilament polypropylene mesh for urinary incontinence: 10 cases of infections requiring removal of the sling


Dr A. Bongain, Department of Gynecology and Obstetrics, University Hospital of Nice, Hôpital de l'Archet 2, CHU de Nice, BP 3079, 06202 Nice cedex 3, France.


One hundred and forty-nine women received a suburethral woven polypropylene mesh insertion for urodynamic stress incontinence. Eleven patients (7.5%) presented between 4 and 17 months after surgery (mean 10 months) with symptoms of retropubic infection. Ten patients required surgical exploration, drainage of the collection and removal of the tape either unilaterally (six) or bilaterally (four). Bacterial cultures were negative in nine women. Following removal of the tape, eight patients reported residual incontinence, either stress leakage (five) or de novo urge incontinence (three). Multifilament polypropylene meshes appear to be more at risk of infection, usually requiring sling removal, which may be related to pore size of the mesh. Our data reinforce the importance of proper evaluation of new surgical procedures by rigorously designed randomised trials with adequate follow up to identify rare but potentially serious complications.


Suburethral slings of autologous fascia or artificial materials (expanded polytetrafluoroethylene (Gore-tex), polyethylene (Mersilene), polypropylene are an established treatment for the treatment of stress urinary incontinence. The tension-free vaginal tape (Gynecare) is a minimally invasive polypropylene monofilament mesh sling which is well established as an effective treatment.1 Other suburethral slings using polypropylene in either a multifilament mesh (the intravaginal slingplasty [IVS], Tyco) or unwoven (Uratape, Porgès) are available.

Infection is an extremely rare complication of the tension-free vaginal tape, occurring in approximately 0.7% of cases, and usually resolves with medical treatment.2 Here, we report 11 cases of purulent collections in the retropubic space related to the use of a multifilament polypropylene mesh sling (IVS, Tyco).


We retrospectively reviewed the cases of all patients who had had a multifilament mesh sling (IVS, Tyco) inserted between February 2001 and April 2002. Particular reference was made to those who presented with evidence of infection.

All patients underwent a pre-operative clinical evaluation to confirm the diagnosis of urodynamic stress incontinence by urodynamic studies. A urinary flow rate was measured to identify evidence of voiding dysfunction (a maximum flow rate below 15 mL/second and postmicturition residual above 20% of the urine volume (always above 150 mL). Dual-channel cystometry was performed to confirm the diagnosis an identify evidence of detrusor overactivity, and static and dynamic urethral pressure profiles were obtained to identify intrinsic sphincter deficiency (a maximum urethral closure pressure below 30 cm of water).

The procedure was performed on patients with urodynamic stress incontinence due to urethral hypermobility or intrinsic sphincter deficiency. None of the patients underwent a prolapse cure procedure at the same time. Antibiotic prophylaxis with Cephalosporin (Kefandol 2 g) intravenously was given 30 minutes before the procedure.

The slings were implanted using the IVS device. The multifilament polypropylene tape, measuring 40 cm in length and 1 cm in width, was implanted as follows: first, a colpotomy of 1.5 cm below the mid-urethra was performed. With the scissors, the para-urethral ducts were drawn out towards the ipsilateral ischiopubic ramus. The tapes were placed with a needle on either side of the urethra that successively pass through the skin, from the abdominal wall to vaginal incision and as near as possible to the posterior side of the pubis, through the Retzius space and the urogenital diaphragm. Cystoscopic monitoring was performed after each needle pass in order to ensure the bladder and the urethra were not injured. In the event of perforation to the bladder, the material was removed and then placed again, and the bladder catheter was kept in place for three days. After filling the bladder with 300 mL of normal saline, tension to the tape was adjusted by making the patient cough. Tension was adjusted so that urine leakage stops upon coughing. The suburethral colpotomy and skin incisions were closed with absorbable stitches. The catheter was removed 3 hours after the operation and the patient was discharged if the postmicturition residual was below 100 mL.

Prophylactic antibiotic therapy for urinary infections, such as Nitrofurantoin (Furadantine), three capsules per day for 10 days, was prescribed.

Patients who presented with signs of infection were initially managed medically with antibiotics (oxacillin: 3 g daily) and washing of the vaginal wall or the skin with iodinated polyvidone for between 14 and 30 days. Where this was ineffective, surgery was performed to explore the infected area, obtain samples for bacteriological culture and drain any collection. The sling was removed from the site of the collection. When the purulent collection was strictly located to one side, only the tape on this side was removed. When the purulent collection affected both sides of the tape, we removed the whole device.

We devised a simple postal questionnaire to evaluate recurrence of stress incontinence symptoms or de novo urge incontinence and sent this to the patients at least five months after the clinical signs of infection had resolved.


One hundred and forty-nine patients were operated on during the study period. Eleven (7.4%) patients presented with an infection of the tape (Table 1). The interval between mesh implantation and onset of first functional signs of infection was an average of 10 months (from 4 to 17 months). The median age of patients at the time of surgery was 64 years (range: 36–91) and that of infected patients was 55 years (range: 38–72 years).

Table 1.  Characteristics and urinary incontinence and sexual activity outcomes of patients with IVS sling infection.
NoAgeTime of presentation (months)Clinical signsSurgical approachOutcome of surgery
15012Vaginal dischargeLaparotomyBilateral tape removal
2606Vaginal dischargeLaparotomyBilateral tape removal
3447Vaginal fistulaVaginalUnilateral tape removal
43817Vaginal dischargeVaginalUnilateral tape removal
5704Abdominal fistulaLaparotomyBilateral tape removal
6699Abdominal fistulaLaparotomyUnilateral tape removal
7729Abdominal fistulaLaparotomyUnilateral tape removal
84610Abdominal fistulaLaparotomyUnilateral tape removal
9759Vaginal dischargeNoneResolution at 3 months
105111Vaginal dischargeVaginalUnilateral tape removal
113713Vaginal dischargeVaginalBilateral tape removal

Ten of the patients failed to respond to antibiotics and required surgical intervention. Six patients had a laparotomy, and in four the collection was drained vaginally. Four patients required total removal while the other six patients had only unilateral removal because the collection was limited to just one branch of the sling.

After sling removal, the collection cleared rapidly without recurrence in any patient. Bacteriologic cultures were negative in nine patients and positive for a Staphylococcus aureus in one case. Ten patients completed the questionnaire. Eight reported urinary incontinence: five with a recurrence of the symptoms of stress urinary incontinence and three with de novo urge incontinence only.


Early synthetic slings were developed using polytetrafluoroethylene or polyethylene. Despite good efficiency (78% of patients cured and 12% of patients improved), these materials had a rate of chronic inflammation of 10%.1 They were very quickly abandoned in favour of polypropylene monofilament, which has a lower complication rate.

The efficacy of the tension-free vaginal tape (a monofilament woven polypropylene mesh protected by a plastic sheath which is removed at the end of the procedure) has led to the development of other tapes made in polypropylene including the multifilament woven polypropylene mesh. These newer meshes have the advantage of being more flexible and less extensible and are said to be easier to adjust once positioned under the urethra (contrary to the monofilament mesh which can no longer be moved once in place).

The main complications described with tension-free vaginal tape suburethral slings arise from needle injury (bladder perforation, vascular injury).2 To our knowledge, there is only one article that reports two cases of acute infection of the retropubic space.3 In our series, 8% of the patients had a late presentation of a retropubic collection, on average, 10 months after surgery.

These collections may be explained by bacterial infection of the multifilament mesh in relation to the ‘pore theory’. If the pore size (defined by the distance between the different filaments of a multifilament mesh) is less than 10 μm, then bacteria, which measure about 1 μm, can penetrate and colonise the mesh, whereas immune cells such as macrophages and neutrophils, which are larger, cannot.4 This would allow bacteria to escape from the action of the immune cells and colonise the mesh.

It is also possible that the collections represented a chronic inflammatory reaction: in the case of Dacron polyester mesh (also known as Mersilene), some authors have reported non-infectious collection due to allergic-like mechanisms.5

We consider that the collections in these patients were infectious. In 6 out of 10 patients, the infection was unilateral and the removal of the tape on the affected side only led to a complete recovery for all our patients. Had those collections been inflammatory in nature, one would expect the inflammation to have affected the other side also. The negative cultures might be explained by the pre-operative antibiotic treatment.

The delayed onset of infection and the absence of fever in all patients might be explained by the ability of some bacteria to produce ‘slime’. Bacterial slime interferes with the action of immune cells due to its low immunogenicity, thereby protecting the bacteria against phagocytosis by macrophages.6 The presence of slime, although unproven in this series, may explain the absence of fever, the lack of effect of antibiotics, and the need for surgical removal of the sling.

These chronic infections should be distinguished from acute retropubic infections, as described by Neuman.3 He described two cases of infection associated with severe systemic upset which had begun rapidly after surgery, and resolved quickly after surgical drainage and antibiotics, without removal of the slings.

In view of the data presented here, the occurrence of delayed infections has changed our post-operative management of the patients, who, in our opinion should be followed up to two years after the surgery. Thus, in the first year, we reviewed our patients after 1, 6, 12, 18 and 24 months.

In summary, we present data on chronic infection as a complication of woven polypropylene mesh presenting after a mean of 10 months. These infections were resistant to antibiotics and required surgical drainage with removal of the tape. The woven nature of the mesh may be implicated in the aetiology of the infection. Our data highlight the need for proper evaluation of new surgical techniques by long term randomised controlled trials to identify uncommon complications.

Accepted 5 July 2004