Introduction

Stress urinary incontinence (SUI) is estimated to affect 50 % of women presenting with urinary incontinence (UI) [1]. Patients with SUI are treated with conservative measures such as pelvic floor exercises initially. Those who do not respond to conservative measures are offered surgery. Midurethral slings (MUS) are the primary gold standard procedure for treating SUI with long-term cure rates of 77–90 % [2]. Recurrent SUI after failed incontinence surgery is a challenging problem.

The reoperation rate for incontinence within 5 years following surgery for prolapse and incontinence has been reported to be 8 % [3]. There is no consensus on the procedure of choice for treating recurrent SUI. There are variable data available regarding the cure rate of SUI with MUS after persistence or recurrence of SUI following previous surgery. We aimed to evaluate the effectiveness of MUS in women with persistent or recurrent SUI after previous surgery. We also wanted to study the complication rates following repeat surgery.

Methods

Sources

All prospective studies describing effectiveness of MUS in women with recurrent SUI were included in this review. They were identified by searching the MEDLINE, EMBASE, CINAHL, CENTRAL (The Cochrane Library, Issue 2, 2009), National Library for Health, metaRegister of Controlled Trials and Google Scholar (up to August 2011). The conference proceedings and abstracts from the International Continence Society and the International Urogynecological Association annual meetings for the last 3 years were also searched. The following keywords were used for the search as text word or subject headings without language restriction: ‘stress urinary incontinence, midurethral sling, tension-free tape, transobturator tape and recurrent stress urinary incontinence’. Hand searches of the bibliographies and citation list of all relevant articles were also carried out. The first or corresponding authors of included trials were contacted for additional information if necessary. No ethical approval was sought for this study, as it was a systematic review and meta-analysis of published manuscripts.

Study selection

Randomised controlled trials (RCTs) and prospective studies of women who underwent retropubic and transobturator vaginal tapes [TVT, outside in transobturator tape (TOT), inside out transobturator tape (TVT-O)] for recurrent SUI following previous surgeries were included. Two reviewers (AP and PJ) independently selected the articles for inclusion by assessing the eligibility of full papers against the review inclusion criteria, with disagreements resolved by discussion, if necessary, with a third reviewer (PL) per a priori protocol. In cases of duplication, the study that reported most recent data was included.

  • Participants: women with recurrent SUI following failed stress continence surgery

  • Intervention: synthetic MUS TVT/ TOT/TVT-O

  • Outcome measures: subjective/objective cure/improvement

Data extraction and quality assessment

Two independent reviewers (AP and PJ) extracted data for quality and results independently. Data included characteristics of participants, intervention, comparisons if any, outcome assessment tools and results of the studies. The Jadad score was used to assess quality of RCT and the Newcastle-Ottawa Scale was used for the prospective cohort studies.

Statistical analysis

Regression rates from individual prospective studies were meta-analysed using a random effects model. Statistical analysis was performed using Stata 8.0 (StataCorp, College Station, TX, USA).

Results

We included 12 studies in total; Fig. 1 summarises the flow for study selection in this systematic review. There was 1 RCT [4] and 11 prospective studies [5–15] with a total of 430 participants between 23 and 87 years of age (see Table 1 for details of included studies). There were nine non-comparative studies, eight of which described effectiveness of TVT and one used TOT (see Table 1). Three comparative studies that described comparison between different sling procedures in the management of recurrent SUI included TVT vs TOT [8], TVT-O vs TOT [4] and TVT vs TOT vs intravaginal sling (IVS) [14]. The interval between primary and repeat surgery ranged from 6 weeks to 48 months. Previous surgery included Burch colposuspension, Kelly plication, Marshall-Marchetti-Krantz procedure (MMK), Stamey needle suspension, Pereyra’s operation, MUS, conventional slings and paraurethral collagen injection in eight studies[4–11], while four studies [12–15] included women with previous failed MUS only.

Fig. 1
figure 1

Selection process for the systematic review of MUS surgery in recurrent SUI

Full size image
Table 1 Details of studies included in the systematic review of MUS for the treatment of recurrent SUI
Full size table

Methodological quality

For the RCT, the Jadad score was 3, randomisation was done using opaque sealed envelope and follow-up was adequate. Power calculation was done and blinding was described.

The quality of the non-randomised prospective studies is shown in Fig. 2. The recruitment was consecutive in two studies [5, 9]. The diagnosis was ascertained both clinically and urodynamically in ten studies, and clinically [with urodynamics (UDS) if history not suggestive] in one study [9]. The instruments used to assess outcomes were satisfactorily described in ten studies. The outcome was assessed as overall cure in the majority of studies [5–10, 14, 15], while subjective and objective components were assessed separately in some studies [4, 11–13]. The follow-up was described in all of them and ranged from 1 month to 12 years and it was adequate (>80 %) in all 11 studies.

Fig. 2
figure 2

Quality of prospective studies in the meta-analysis of MUS in recurrent SUI

Full size image

Outcomes

The studies used different parameters for subjective/objective cure and improvement. Subjective parameters used were bladder diary, patient satisfaction, subjective symptoms analysis, quality of life (QOL) evaluation, King’s Health Questionnaire (KHQ), Pelvic Organ Prolapse/Urinary Incontinence Sexual Function Questionnaire (PISQ-12), Patient Global Impression of Improvement (PGI-I), International Consultation on Incontinence Questionnaire-Short Form (ICIQ-SF), Incontinence Severity Index (ISI), Urogenital Distress Inventory (UDI), Incontinence Impact Questionnaire (IIQ), Urinary Incontinence Severity Score (UISS), visual analogue scale (VAS) and absence of complaint of leakage.

The objective outcomes measured in these studies were UDS, cough stress test and pad test (which included 1-h pad test <2 g/h, <1 g/h, 24-h pad test <10 g/24 h).

Criteria used for improvement were 1-h pad test < 5 g, 24-h pad test showed >80 % reduction in leakage, reduction in urine leakage to less than 50 % and QOL improvement >75 % but less than <90 %.

Overall subjective cure

On meta-analysis the subjective cure rate following MUS for recurrent SUI after any previous surgery was found to be 78.5 % [95 % confidence interval (CI) 69–88] at the follow-up of 29.72 ± 29.49 months (Fig. 3).

Fig. 3
figure 3

Meta-analysis for MUS after recurrent SUI with any previous incontinence surgery

Full size image

There were four studies looking at cure rate of MUS after previous MUS surgery for SUI [12–15]. The subjective cure rate was 73.3 % (95 % CI 55–97) at 15.7 ± 7.7 months follow-up (Fig. 4).

Fig. 4
figure 4

Meta-analysis for MUS after recurrent SUI with previous MUS

Full size image

Outcome after TVT

The subjective cure rate following retropubic MUS (TVT) was 79.8 % (95 % CI 71–89) at a follow-up of 27.26 ± 26.85 months (Fig. 5). There were ten studies looking at outcome after retropubic MUS following any previous failed incontinence surgery. Three of these ten studies looked at outcome with repeat retropubic MUS after previous failed MUS [13–15]. Seven of these ten studies had a success rate of over 80 %. One study reported a cure rate of 61 % and two others had a cure rate ≥ 70 %.

Fig. 5
figure 5

Meta-analysis for TVT after recurrent SUI with any previous incontinence surgery

Full size image

Outcome after TOT

There were four studies looking at outcome after TOT following any previous surgery and the collated subjective cure was 54 % (Fig. 6). Two of these four looked at outcome of repeat TOT after previous failed MUS [12, 14]. The highest cure rate reported after TOT was 66.7 % (95 % CI 25–100) [8]. One study had a cure rate of 33.3 % [14] and the other two studies had cure rates of 52.6 % [12] and 55.6 % [4].

Fig. 6
figure 6

Meta-analysis for TOT after recurrent SUI with any previous incontinence surgery

Full size image

Complications

In this systematic review, there were 28 urinary tract injuries during 363 retropubic sling procedures (7.7 %) during repeat continence surgery. This compares favourably with a previous systematic review showing bladder injury in 10 of 238 procedures (4.2 %) during primary retropubic sling procedures [16]. The incidence of de novo urgency was 5.5–20 %. Voiding dysfunction was found in 4–16 % and urinary tract infection in 5.9–13 %. Excess bleeding (1.7–6.7 %) and urinary retention needing loosening of tape (9 %) were the other complications seen in some studies [7, 9, 13].

Discussion

The studies report good cure rates for recurrent SUI with MUS following previous failed incontinence surgery (62–100 %). There seems to be a lower cure rate with TOT (33–67 %) compared to the retropubic tape (61–100 %) for recurrent SUI after previous surgery for this condition. The rate of urinary tract injuries is around 7.7 % with retropubic slings.

The search was thorough and not restricted by language. The manuscript adheres to the 27-item PRISMA checklist [17]. The latest joint International Urogynecological Association/International Continence Society terminology for female pelvic floor dysfunction was used [18]. The definition of cure varied between studies and this can also account for the variable success rates described. The drawbacks of this piece of work are not weaknesses of the systematic review itself but the dearth of primary RCTs in this area. The long-term success of MUS has been reported between 78.9 and 84.7 % [19, 20]. The common reason for failure of initial sling is either anatomical or functional [21]. A correctly placed MUS mimics the effect of the pubourethral ligament. [22].

Repeat continence surgery is often difficult as a result of scarring and adhesions with the distortion in anatomy from previous surgeries, thereby increasing intraoperative complications [5]. Retropubic MUS has a good success rate after any type of previous surgery for SUI including previous MUS surgery with bladder perforation rates quoted between 2.9 and 19 %. Our view is that, as long as bladder perforation is recognised and the tape is repositioned, it may not have long-term implications. A retropubic approach is preferred over a transobturator approach for repeat surgery [12]. The axis of the transobturator is less acute to the urethral axis as compared to the retropubic route, thereby creating a platform sling with less circumferential compression of the urethra [23]. Another reason could be the need for wider dissection of the periurethral area with the transobturator approach resulting in migration of the tape [24].

The retropubic sling has been used successfully to treat recurrent SUI due to intrinsic sphincter deficiency (ISD) [15]. The rationale proposed for this is the greater acute angle which may be more obstructive than the transobturator and, therefore, more successful in treating incontinence due to ISD. The main mechanism of MUS relies on dynamic kinking of the proximal urethra and compression of the midurethra at the level of the sling. This procedure requires adequate mobility of the urethra. The success rate of repeat surgery was 90 % in women with sufficient preoperative mobility of the urethra, while it was only 33 % in women with a fixed urethra [7].

It would be ideal to collect data on repeat surgery on failed incontinence surgery prospectively via national databases (e.g. British Society of Urogynaecology, BSUG in the UK) to build up a further evidence base and perform individual patient data meta-analysis.