Two routes of transobturator tape procedures in stress urinary incontinence: a meta-analysis with direct and indirect comparison of randomized trials


  • Pallavi M. Latthe,

    1. Department of Obstetrics and Gynaecology, Birmingham Women’s Hospital, *Goodhope Hospital, Birmingham, and Southmead Hospital, Bristol, UK
    Search for more papers by this author
  • Pinki Singh,

    1. Department of Obstetrics and Gynaecology, Birmingham Women’s Hospital, *Goodhope Hospital, Birmingham, and Southmead Hospital, Bristol, UK
    Search for more papers by this author
  • Richard Foon,

    1. Department of Obstetrics and Gynaecology, Birmingham Women’s Hospital, *Goodhope Hospital, Birmingham, and Southmead Hospital, Bristol, UK
    Search for more papers by this author
  • Philip Toozs-Hobson

    1. Department of Obstetrics and Gynaecology, Birmingham Women’s Hospital, *Goodhope Hospital, Birmingham, and Southmead Hospital, Bristol, UK
    Search for more papers by this author

Pallavi Latthe, Department of Obstetrics and Gynaecology, Birmingham Women’s Health Care NHS Trust, Edgbaston, Birmingham B15 2TG, UK.


Study Type – Therapy (meta-analysis)
Level of Evidence 1a


To assess the effectiveness and complications of transobturator tape (inside-out and outside-in, TOT) by means of a systematic review of direct and indirect randomized controlled trials (RCTs).


MEDLINE, EMBASE, CINAHL, LILIACS (up to December 2008), CENTRAL (The Cochrane Library, Issue 1, 2009), MetaRegister of Controlled Trials, The National Library for Health, the National Research Register and Google Scholar were searched using various relevant search terms. The citation lists of review articles and included trials were searched and contact with the Correspondence of each included trials was attempted. RCTs which compared the effectiveness of synthetic transobturator (inside-out tape TVTO, or outside-in TOT) with TVT by the retropubic route (Gynecare, Ethicon, Inc., or similar tape by a different company) or with each other for the treatment of stress urinary incontinence (SUI), and in all languages, were included. Two reviewers extracted data on participants’ characteristics, study quality, intervention, cure and adverse effects independently. The data were analysed using Review Manager 5 software.


There were 12 RCTs that compared TOT with TVT, and 15 that compared TVTO vs TVT for treating SUI. There were four direct comparison RCTs of TVTO vs TOT. When compared at 1–44 months, the subjective (odds ratio 1.16; 95% confidence interval 0.83–1.6) and objective (0.94; 0.66–1.32) cure of TOT was similar to TVT. For TVTO, the subjective (1.06, 0.85–1.33) and objective cure (1.03, 0.77–1.39) was also similar to TVT. Adverse events such as bladder injuries (TOT, odds ratio 0.11, 0.05–0.25; TVTO, 0.15, 0.06–0.35) and haematomas (0.06, 0.01–0.30) were less in the TOT than TVT. Voiding difficulties (TOT, odds ratio 0.61, 0.35–1.07); TVTO, 0.81, 0.48–1.31) were slightly lower in TOT but this was not statistically significant. Groin/thigh pain (TVTO, odds ratio 8.05, 3.78–17.16) and vaginal injuries (TOT, 5.82, 1.85–18.3; TVTO, 1.69, 0.73–3.91) were more common in the transobturator tapes. Mesh erosion in TVTO (0.77, 0.22–2.72) and TOT (1.11, 0.54–2.28) was similar to TVT. The effectiveness data over 6 months available from four direct comparison studies of TVTO vs TOT suggested equivalent results for objective cure (1.06, 0.65–1.73) and subjective cure (1.37, 0.93–2.00). When compared indirectly, TVTO has similar subjective (1.23, 0.83–1.82) and objective cure (0.97, 0.62–1.52) to TOT. On indirect comparison, the de novo risk of urgency was similar in the two groups but voiding difficulties seemed to be less in the inside-out group.


The evidence for the equivalent effectiveness of TOT and TVTO when compared with each other is established over the short-term. Bladder injuries and voiding difficulties seem to be less with inside-out tapes on indirect comparison. An adequate long-term follow-up of the RCTs is desirable to establish the long-term continued effectiveness of transobturator tapes.


stress urinary incontinence


urodynamic SUI


transobturator tape (outside-in)


tension-free vaginal tape (obturator


randomized controlled trial


intention to treat


odds ratio.


Stress urinary incontinence (SUI) is estimated to affect up to a third of women aged ≥18 years, with a median age of 45 years [1]. The tension-free vaginal tape (TVT) is a standard minimally invasive procedure used to treat SUI, since 1995 when it was first described by Ulmsten et al.[2,3]. It was shown to have similar effectiveness in the medium term to colposuspension, but with fewer complications [4]. Although success rates are reported to be 84–95%, complications described include bladder, bowel and major blood vessel injury, as well as postoperative voiding difficulties [5] and de novo urgency and urge incontinence [6].

In 2001 Delorme [7] described a new method of inserting the tape which passes through the obturator foramen, thus theoretically avoiding some of the complications such as bladder perforation. In the transobturator tape with the outside-in technique (TOT), after the initial anterior vaginal incision and dissection, the tape is introduced from the skin on the obturator foramen and comes out in the vaginal incision. In the TVTO technique the needle is passed in a reverse route, i.e. in from vaginal incision and out through the obturator foramen (inside-out) [8]. In the preliminary study, Delorme showed that there was a high success rate, no bladder perforations and few perioperative complications via the transobturator route and this procedure was subsequently widely adopted and has a good evidence base [9–11]. The path traversed by the needle in the inside-out technique is slightly different to the outside-in technique [12]. There was a suggestion of slightly inferiority in a recent systematic review comparing the TOT and TVTO to retropubic TVT, although this was not statistically significant [13].

The aim of the present study was to assess the effectiveness and complications of tapes in SUI using an up-to-date systematic review and meta-analysis of randomized controlled trials (RCTs) between TVTO and TOT, by direct and indirect comparison.


A prospective peer-reviewed protocol for this review was prepared a priori. All reports which describe (or might describe) RCTs of mid-urethral slings by the obturator route in the treatment of SUI were obtained. MEDLINE (1966 to December 2008), EMBASE (1980 to December 2008), CINAHL (1982 to December 2008), LILIACS (up to December 2008), the Cochrane Incontinence Group Trials Register (19 December 2008), CENTRAL (The Cochrane Library, Issue 3, 2006), National Library of Health, Google Scholar, MetaRegister of Controlled Trials, the citation lists of review articles and included trials were searched and contact with the correspondent of each included trial abstract was attempted. The following keywords were used for the search as text words or subject headings without language restriction using OVID software: ‘stress urinary incontinence’, ‘transobturator tape’, ‘continence surgery’, ‘tension free tape’, ‘mid urethral sling’. In addition, hand searches of the bibliographies and citation lists of all relevant reviews and primary studies to identify articles not captured by electronic searches, as well as the proceedings of the International Urogynaecological Association and ICS of the last 3 years (2006–2008) were performed. In some cases, the first author or correspondent of included trials were contacted for additional information.

All prospective RCTs comparing synthetic vaginal tape by the obturator route (TOT or TVTO) with TVT (retropubic tape below-upwards) or with each other for women with SUI were selected. In the suprapubic arch sling, the needles are passed from suprapubic incision into the vagina via retropubic route (above-downwards). It was not included as an alternative control for TVT due to controversy about differences in complications such as vaginal erosion, return to theatre for loosening of tape, as well as lower subjective cure rates than TVT [14–16]. Two authors (P.L. and P.S.) independently scrutinized the electronic searches and obtained full papers of all citations that were likely to meet the predefined selection criteria. Final inclusion or exclusion decisions were then made after examining these reports. In cases of two or more publications from same trial group, we selected the most recent and complete version for quality assessment and then extracted relevant data from both the manuscripts. We also asked the authors for recent most data wherever available.

All assessments of the quality of trials and data extraction were done independently by two authors (P.S. and P.L.) using forms designed according to Cochrane guidelines. Reports not in English were assessed by one person who had command of the language, to allow data extraction from these reports. We resolved any disagreements about inclusion or exclusion by consensus or arbitration by a third reviewer (R.F.). Data on characteristics of the study participants including details of previously administered treatments, interventions used and methods used to measure success (or definitions of cure/failure) and adverse events were extracted. The following quality criteria and methodological details were assessed: method of randomization; quality of allocation concealment until randomization; blinding to treatment allocation after randomization; power calculation, intention to treat (ITT) analysis and numbers lost to follow up or withdrawals.

The studies used different objective measures such as negative cough stress test on urodynamics and 1-day or 24-h pad test of <1–5 g for cure. Different subjective measures like visual analogue scales, simple questionnaires and validated questionnaires were also used. The subjective cure rates (wherever given) and objective cure rates were pooled separately.

The results were analysed statistically according to the statistical guidelines of the Cochrane Collaboration [17] using RevMan 5 software. Data from ITT analyses were used where available. For the dichotomous data, results of each study were expressed as an odds ratio (OR) with 95% CI and combined for meta-analysis using the Mantel-Haenszel method [18]. We used the results of TVTO vs TVT and TOT vs TVT to generate pooled log ORs and their variances for making adjusted indirect comparisons of TVTO vs TOT. This was done to confirm the results available from direct comparison studies but from a larger sample size.

The log OR (ln ORAB) of the adjusted indirect comparison of TVTO (A) vs TOT (B) was estimated by:

ln ORAT − ln ORBT

and its variance was:

Var(ln ORAB) = Var(ln ORAT) + Var(ln ORBT)

where ln ORAT is the log OR of the direct comparison of TVTO vs TVT and ln ORBT is the log OR of direct comparison of TOT vs TVT. The square root of Var(ln ORAB) gave the standard error, from which the 95% CI for the estimated ln ORAB was calculated [19].


Figure 1 summarizes the flow for study selection in the systematic review and meta-analyses. We found 31 RCTs (13 abstracts) in total, with 4796 women analysed (details of patient characteristics, inclusion and exclusion criteria and various outcome measures are given in detail in Table S1) in these trials. The TVT and TVTO made by Gynecare (Ethicon, Inc., Johnson & Johnson, USA) and various different TOT brands (Monarc; Obtape/Uratape-Mentor Porges; unknown) were used in these studies. The numbers of women randomized by type of tape were TVTO, 1530; TOT, 1144; and TVT, 2122. Fifteen trials compared TVTO with TVT [20–34] and 12 compared TOT to TVT [5,35–45], and four trials compared TVTO with TOT [46–49]. The trials included women with mean ages of 50–60 years, and the all of the study participants appear to have had urodynamic studies to confirm urodynamic SUI (USI). Seven studies included women with stress-predominant mixed UI as well [22,29,36,43,46,48,49], while 22 studies included women with only USI. The surgery was performed using standard techniques [7,8,50]. Cystoscopy was used in both groups in all women in eight trials [5,30,31,33,35,40,41,47]. The primary outcome in most trials was ‘cure’ and this was measured and reported in various ways. Most but not all studies reported on subjective cure rates and some also defined objective cure (supplementary Table S1). In one trial, the primary endpoint was a reduction in bladder injury [35] and in another it was operating time and complications [43]. Outcomes were assessed at various periods (3–44 months) after surgery, except one study, reported in an abstract, where the outcomes were reported at 1 month [28]. About half of all studies reported on changes in quality of life (as recommended by the ICS) [22,24,25,29–31,33,35,37,40–42,45,46,48]. Other studies reported on some other measures of patient satisfaction, like a simple questionnaire [5,21,28,38,47], or visual analogue scale [27,36,43].

Figure 1.

The study selection process for systematic review of effectiveness of TVTO and TOT in SUI.

Figure 2 gives details of various quality criteria assessed for this systematic review. Nine of the 31 trials reported adequate concealment before allocation. Thirteen studies had computer-generated randomization, two had telephone randomization, while five were quasi-randomized [28,29,32,39,47], Of the 31 studies, 21 reported ‘blinding’ either the patient or assessor or the statistician. Power calculation and ITT analysis was done in 17 of 31 studies. In 28 of 31 trials, <15% of randomized participants withdrew or were lost to follow-up. Overall, the quality of trials was variable.

Figure 2.

The quality of studies included in the systematic review and meta-analysis of the TVTO and TOT.

When compared at 1–44 months, the subjective and objective cure of TVTO and TOT was similar to TVT (Table 1 and Fig. 3). When a subgroup analysis of subjective cure of TOT in USI vs mixed incontinence was done, there was no statistical difference in the two groups (OR 0.83, 0.66–1.03). There was no difference between inside-out and outside-in tapes in USI vs mixed incontinence on both direct and indirect comparison (Table 2). Adverse events such as bladder injuries were less common in the transobturator tapes. Voiding difficulties were similar (Fig. 4). Groin/thigh pain (TVTO, OR 8.05, 3.78–17.16) and vaginal injuries (TOT, OR 5.82, 1.85–18.3; TVTO, OR 1.69, 0.73–3.91) were more common in the transobturator tapes. Mesh erosion was similarly common in tapes inserted by transobturator and retropubic route. Figure 4 quantifies the risks of various complications by the type of tapes.

Table 1.  Indirect and direct comparison between TVTO vs TOT with the common comparator, TVT
ComparisonsOR (95% CI)
Subjective cureObjective cureBladder injuryVoiding difficultiesDe novo urgency
  • *

    chi-square test for heterogeneity (P < 0.05) statistically significant for the comparison.

Direct comparison     
 TVTO vs TVT1.06 (0.85, 1.33)*1.03 (0.77, 1.39)0.11 (0.05, 0.25)0.81 (0.48, 1.31)2.23 (1.30, 3.82)
 TOT vs TVT1.16 (0.83, 1.60)0.94 (0.66, 1.32)0.15 (0.06,0.35)0.61 (0.35, 1.07)0.87 (0.50, 1.51)
 TVTO vs TOT1.37 (0.93, 2.00)1.06 (0.65, 1.73) 1.23 (0.27, 5.69)0.98 (0.42, 2.31)
Indirect comparison     
 TVTO vs TOT1.23 (0.83, 1.82)0.97 (0.62, 1.52)0.17 (0.005, 0.05)0.49 (0.24–1.04)1.74 (0.82,3.69)
Figure 3.

The meta-analysis of cure with TOT vs TVT.

Table 2.  Subjective cure in the TVTO and TOT routes in USI and mixed incontinence
ComparisonsSubjective cure (SUI only)No. of studiesSubjective cure (mixed)No. of studies
Direct comparison
TVTO vs TVT1.03 (0.81, 1.31)81.29 (0.70, 2.37)2
TOT vs TVT0.98 (0.68, 1.41)62.42 (1.10, 5.30)1
TVTO vs TOT1.20 (0.49, 2.94)11.41 (0.92, 2.15)3
Indirect comparison    
TVTO vs TOT0.94 (0.69, 1.27)140.99 (0.46, 2.12)3
Figure 4.

A meta-analysis of complications with TOTs.

As shown in Table 1, the effectiveness data over 6 months available from four direct comparison studies of TVTO vs TOT suggests equivalent results on objective cure and subjective cure. When compared indirectly, TVTO had similar subjective and objective cure, rates and complications as TOT. There is a slightly higher risk of UTI in the TVTO group (OR 1.85, 0.51–6.68) although this was not statistically significant. Interestingly, on indirect comparison, the bladder injury and voiding difficulties seem to be less with inside-out than with outside-in tapes (Table 1), although the risk of de novo urgency is similar.


The evidence for the equivalent effectiveness of inside-out and outside-in tapes and TVT with each other is established over a short- to medium-term follow up. Bladder injuries, voiding difficulties and haematomas are less common but the risk of groin pain and vaginal injury is higher with transobturator tapes.

To the best of our knowledge, this systematic review, with the data that was used, summarizes most of the existing RCTs on this topic. The two techniques of inserting vaginal tape by obturator route were not significantly different in effectiveness, both on direct and indirect meta-analysis. The TVTO and TOT had fewer complications of bladder injury but more groin pain and de novo urgency than TVT. The higher rate of mesh erosion (though statistically insignificant) could be a sequel of higher recognized and unrecognized vaginal injuries in this group due to its path of insertion.

There are several strengths of this review. The search was thorough and systematic, with no language restrictions. Two reviewers independently selected the studies and extracted data, to minimize errors. We attempted to contact authors of published and unpublished studies to obtain further details. We adhered to the QUOROM checklist while reporting the meta-analyses [51]. We used network meta-analysis or indirect comparison [52] for subjective and objective cure, and this supports the results from the existing few direct comparison trials on cure and complication rates by the two obturator routes which traverse slightly different paths [53]. We took care to ensure that the included trials were similar, showed consistent results and included the direct comparison data to ensure the robustness of our indirect comparison [54].

A potential objection by some might be the inclusion of one of the studies that was retracted after publication, for ethical reasons [5]. However, we feel that, as the patients who were involved gave permission at the time, it would be unethical to ignore the data collected to guide future treatment and research. Another potential criticism could be that we have not assessed abstracts of other meetings like the American Urogynecological Society or the AUA. In this developing field there is still poor standardization of what constitutes anatomical and functional cure. The definition and duration at assessments of cure and adverse events such as bladder injury, voiding difficulties/urinary retention, urgency or detrusor activity were different in the included studies. It is difficult to explain a plausible reason for TVTO in causing lesser voiding difficulty than TOT, as found on the indirect comparison, although this has not been confirmed in direct studies and it was not well defined in all studies. In one study the postvoid residual was significantly (statistical rather than clinical) higher in the inside-out procedure than the outside-in procedure (18.9 vs 5.3 mL) [48], whereas there were no differences found in a non-randomized study [55]. The bladder injuries were reported by various methods, in that some trials used cystoscopy in both groups [6], whereas some used the indigo-carmine test [5], and some did not use cystoscopy in the TOT group, with the potential of missing a bladder or urethral injury. The cadaver studies suggest that perioperative cystoscopy is not required [12], although there are reports of this complication [56,57], and in the included trials [20,36,43]. Nevertheless, the heterogeneity for the various adverse events was not significant. The funnel plot [58] indicated a symmetrical distribution of the studies, indicating a low likelihood of publication or reporting bias (Fig. 5). We could not obtain more details on some of the abstracts despite attempting to contact the authors.

Figure 5.

A Funnel plot of studies included in the meta-analysis to study subjective cure of SUI with TOT or TVT.

The trend towards less voiding difficulty and significantly less risk of bladder injury might theoretically make the TVTO and TOT preferable in women with borderline flow rates and who have had previous operations such as colposuspension, where the bladder could be tethered. Clinicians can enter the outcomes on a national/international database/registry to collect robust observational data over long periods. We also feel that it is advisable to consider intraoperative cystoscopy during a TVTO or TOT procedure to identify bladder/urethral injury and rectify this at the time of operation.

To help resolve the issue of long-term effectiveness and complications, there is a need for a long-term follow-up of patients already randomized in the trials. The main issues are sample size, trial methods and standardization of definitions of outcome. An individual patient-data meta-analysis might address the uncertainty of definition of cure and different complications by combining raw data from various studies included in this review, as well as the data from ongoing studies.


The authors acknowledge the helpful comments of Professor K. Khan and Mr A. Coomarasamy.


Pallavi M. Latthe and Philip Toozs-Hobson have received a small grant from UKCS. Source of Funding: UKCS.


The TOT is one of the more pre-eminent procedures for the minimally invasive treatment of female SUI. Currently there are insufficient data on the superiority of one way (outside-in) vs the other (inside-out). This meta-analysis provides a tool to better understand the real efficacy of TOT, considering the outcomes and complications. It is often difficult when assessing previous reports to understand the real value of an innovative procedure. The merit of this paper, after an honest and good review, is to allow the reader to improve the quality of cure, taking care of the quality of life of patients that currently request procedures that are more effective but also safer and free from complications.

Mauro Cervigni

Urogynaecology Unit, San Carlo-IDI Hospital, Rome, Italy