To investigate how urethral mobility and urethral closure pressure affect the outcome of tension-free vaginal tape (TVT) insertion for stress incontinence.
To investigate how urethral mobility and urethral closure pressure affect the outcome of tension-free vaginal tape (TVT) insertion for stress incontinence.
A total of 191 consecutive women with genuine stress urinary incontinence with or without intrinsic sphincter deficiency were evaluated prospectively with multichannel urodynamics, 24-h voiding diaries, clinical stress tests and introital ultrasound measurements preoperatively and 6 months after surgery. Additional introital ultrasound examinations were performed immediately after the operation, at 12 months and annually thereafter. 177/191 patients had completed a 36-month follow-up at the time of writing. Urethral mobility was described as linear dorsocaudal movement (LDM), with hypermobility being defined as LDM > 15 mm on sonography. Intrinsic sphincter deficiency was defined by a maximum urethral closure pressure (MUCP) of <20 cmH2O.
The overall cure rate at the 36-month follow-up was 89.5% (Kaplan-Meier estimator), with secondary cure (within 6 months of surgery) in 10.5% of these patients. The operation failed in 4.2% of the women and recurrence was seen in 6.3% of the cases. Bladder neck mobility was significantly reduced at the 6-month follow-up (P < 0.001). Compared with primary cure, therapeutic failure and secondary cure were associated with a significantly lower postoperative bladder neck mobility (P < 0.05). Postoperative hypermobility reduced the risk of therapeutic failure. In addition, women with therapeutic failure or secondary cure had a significantly lower MUCP than did those with primary cure (P < 0.01).
The effectiveness of the TVT sling appears to depend on adequate postoperative urethral mobility and urethral closure pressure. Copyright © 2006 ISUOG. Published by John Wiley & Sons, Ltd.
Insertion of a tension-free vaginal tape (TVT) according to Ulmsten et al.1 has evolved to be the preferred surgical option for treating stress urinary incontinence since its introduction in the mid 1990s2, 3. Published results indicate an objective 7-year cure rate of 81.3% after primary intervention3. This cure rate is comparable to the outcome of the colposuspension procedure according to Burch, as has also been confirmed in randomized controlled trials2, 4, 5.
Bladder neck mobility following the TVT procedure has been evaluated by sonography6–10, magnetic resonance imaging11, lateral cystourethrography12, X-ray imaging13 and the Q-tip test14–17. TVT insertion aims to stabilize the mid-urethra1, 6, thus differing from the Burch colposuspension procedure, in which the bladder neck is repositioned in the zone of abdominal pressure transmission6. Unlike Burch colposuspension, TVT insertion is therefore expected to have only a small effect on mobility in the area of the bladder neck6, 7, 10. Nevertheless, some studies8, 9, 15, 16 have found significant reductions in bladder neck mobility after TVT.
The mechanism by which TVT induces urinary continence under stress is still not fully understood. Urodynamic measures of urethral function, such as the urethral pressure profile, leak point pressures, pressure transmission ratio and degree of urethral mobility, have been studied previously in attempts to characterize the effectiveness of polypropylene sling procedures. The objective of our study was to characterize prospectively the effect of TVT on urethral position and function using pre- and postoperative urodynamic and sonographic evaluation. In addition, mid-term (up to 36 months) objective and subjective cure rates were recorded. These investigations served to define and identify criteria for urethral mobility with the help of sonography and to establish perioperative parameters that predict which women will have unsuccessful TVT procedures.
From February 1999 to December 2004, a total of 191 consecutive patients undergoing TVT insertion according to Ulmsten et al.1, under local anesthesia, as the sole procedure for treating urodynamically proven genuine stress urinary incontinence were enrolled into this study. The study was exempt from ethics committee approval by the Institutional Review Board of the University of Goettingen.
Preoperatively, all study patients underwent the usual clinical and urodynamic examinations of a quality assurance program (cystometry, urethral pressure profile and uroflowmetry), as well as introital ultrasound examination of the urogenital tract18, 19 (Figures 1 and 2). A detailed history was taken for each patient and their status determined urogynecologically and neurologically. Patients whose symptoms had an adverse effect on quality of life and who had failed to respond to conservative measures were offered the TVT procedure. The presence of intrinsic sphincter deficiency (defined as a maximum urethral closure pressure (MUCP) of <20 cmH2O) was not considered a contraindication to surgery. Concomitant detrusor instability was not an absolute contraindication to surgery, provided it was very mild and had responded to bladder drill (modification of drinking and voiding habits) and anticholinergic therapy preoperatively. Voiding dysfunction was defined as symptoms of stranguria and a post-micturition residual volume of >100 mL at discharge. The examinations were performed and the results documented as described in earlier studies18, 19. Since only routine data were recorded, the ultrasound examinations were performed according to the guidelines issued by the German–Austrian–Swiss Study Group of Urogynecology20 and all definitions used are in accordance with the guidelines of the International Continence Society21, unless stated otherwise. All introital ultrasound examinations were performed using a Siemens Sonoline Adara SLC (Siemens AG, Erlangen, Germany) ultrasound machine with the usual technique as described elsewhere18–20. Sonography was performed by four examiners in three hospitals. Image evaluation and analysis was done by one independent observer. The patients were examined in the recumbent position as part of cystometry after filling of the bladder to a standardized volume of about 300 mL. Postoperative measurements were performed at a filling volume of 200–300 mL as determined by ultrasound. In addition, the ultrasound examination served to determine residual urine and to identify upper urinary tract dilatation. Two-dimensional linear dorsocaudal movement (LDM) and hypermobility of the bladder neck (LDM > 15 mm) were defined and evaluated as described previously19, 22 (Figure 1). In addition, bladder neck mobility was determined as one-dimensional vertical displacement of the bladder neck during straining (bladder neck descent, BND) as defined by Dietz et al.23.
The patients underwent the TVT procedure at three different hospitals (university medical centers in Goettingen (n = 69) and Witten (n = 45) and the German Red Cross Hospital in Alzey (n = 77)). A total of 191 women of 198 consecutive patients completed postoperative follow-up at 6 months and were included in the analysis. Seven patients were excluded from the investigation because they were lost to postoperative follow-up at 6 months. On clinical and ultrasound examination at discharge (on the second postoperative day), all patients except one (who had mild voiding dysfunction and residual volume of 60 mL) were continent and did not report postoperative complications. 177 of the 191 patients had completed their 36-month follow-up at the time of writing. Immediately after the operation, outcome was evaluated objectively and subjectively as well as at follow-up after 6, 12, 24, and 36 months. Postoperative subjective assessment included a condition-specific quality of life tool, the Kings Health Questionnaire (KHQ)24, the patient's history and 24 h voiding diaries. Outcome was objectively assessed by supine and standing cough stress test, clinical examination and ultrasound (introital and determination of residual urine) performed 2 days after surgery as well as 6, 12, 24 and 36 months postoperatively. Repeat multichannel urodynamic testing was performed at the 6 month follow-up or in case of recurrence. The numbers of patients seen at each time point are given in Table 1.
|LDM (mm, median (range))||13.7 (2–30)||9.0 (1–32)***||10.4 (1–28)***||11.2 (2–28)***||11.8 (1–32)***||12.1 (1–29)**|
|Hypermobility, LDM > 15 mm (n (%))||80 (41.9)||25 (13.1)***||32 (16.8)***||45 (23.8)***||46 (25.0)***||44 (24.9)***|
|BND (mm, median (range))†||12.0 (−3 to 28)||8.3 (−4 to 31)***||8.0 (−1 to 27)***||9.0 (0–27)***||8.9 (1–31)***||9.2 (1–29)***|
|MUCPrest (cmH20, median (range))||41 (12–102)||—||46 (10–109)||—||—||—|
In addition, the tape position was checked sonographically immediately after surgery and at the 6-month follow-up. The center of the TVT was defined as the point of reference for distance measurements. The distance between the tape and the urethra was measured in mm along a line perpendicular to the urethra. The tape position was also determined in relation to the sonographically measured length of the urethra. To this end, the length of the urethra was defined by percentages, in order to describe the position of the tape relative to the urethra; the bladder neck was assumed to be at 0% of the urethra, the middle of the urethra at 50%, and the external urethral orifice at 100%. The tape position was classified as orthotopic when the tape was seen at the level of the middle third of the urethra (33–67%), and as dystopic when it was identified adjacent to the upper or lower third.
Before and 6 months after TVT insertion, a urethral pressure profile was obtained to determine the MUCP at rest (MUCPrest). The pressure was measured by withdrawing the catheter from the bladder into the urethra at a defined and continuous rate of 1 mm/s18.
Treatment outcomes were classified as cure, failure or recurrence using a combination of subjective and objective outcome measures as described previously18, 19. Cure of incontinence was defined as a dry, symptom-free patient without objective urine loss during vigorous coughing and other provocative activities at a standard bladder filling of 300 mL and a demonstrable positive urethral closure pressure during stress provocation. Additional criteria were no episodes of stress or urge incontinence in the 24-h voiding diary and no post-void residual urine. Moreover, the definition of cure comprised assessment of subjective continence by means of a self-completed detailed urinary incontinence questionnaire and the patient's history. Failure was defined as not becoming continent within 6 months of surgery. The recurrence-free interval was defined as the period until incontinence recurred. Apart from distinguishing recurrence versus surgical failure, we identified two subsets of cured patients, those with primary cure who were continent immediately after TVT and those with secondary cure who became continent within the first 6 months of the intervention and remained so during the study period.
Statistical analysis was performed using STATISTICA, version 6.1.409 (StatSoft, Inc. Tulsa, OK, USA (2003)). For group comparisons of parameters being at least ordinal scaled the Wilcoxon–Mann–Whitney test was applied; for categorical parameters Pearson's chi-square test was used. Cure rates and the incidence of adverse events were determined using Kaplan–Meier estimators. Intraindividual changes of variables being at least ordinal scaled were analyzed using the sign test, while the McNemar test was used for categorical variables. Statistical significance was assumed at a P-value of <0.05.
The patients had a median age of 59 (range, 22–81) years at the time of surgery. Their median height was 165 (range, 149–182) cm and their median weight was 72 (range, 46–131) kg, resulting in a median BMI of 26.5 (range, 16.3–43.4). The median parity was 2 (range, 0–6). The mean and median follow-up was 34.6 months and 36 months, respectively, with a range of 6–36 months.
At the 36-month follow-up, the overall cure rate of the study population was 89.5% (Kaplan–Meier estimator), including 10.5% with secondary cure. The TVT procedure failed in 4.2% of the women. Five of the eight patients classified as treatment failures showed improvement compared with their preoperative findings, but were not continent. Recurrence was seen in 6.3% of the cases. There was full agreement between objective and subjective results with regard to the incontinence-free interval in 179 of the 191 patients (93.7%). The exact two-sided 95% CI for the agreement between objective and subjective cure rates was [89.3–96.7]. A comparison of the patient data by center revealed no significant differences.
Six months after the TVT procedure, 1% of the patients had voiding difficulties but these had disappeared at the next follow-up visit. The incidence of urge symptoms was 6.3% 6 months postoperatively and 1.6% at the 12-month follow-up. De novo urge incontinence was present in 0.5% of cases at the 6-month follow-up and in no patients after 1 year.
The median LDM of the bladder neck during straining decreased significantly from 13.7 (range, 2–30) mm preoperatively to 9.0 (range, 1–32) mm immediately after the TVT procedure and 10.4 (range, 1–28) mm at the 6-month follow-up (P < 0.001) (Table 1).
The median BND, i.e. vertical movement, was 12.0 (range, −3 to 28) mm before surgery and decreased significantly after TVT insertion to 8.3 (range, −4 to 31) mm immediately after surgery and 8.0 (range −1 to 27) mm at the 6-month follow-up (P < 0.001) (Table 1).
The median urethral hypermobility (LDM > 15 mm during straining) also decreased significantly from 41.9% (n = 80) preoperatively to 13.1% (n = 25) immediately postoperatively (P < 0.001) and 16.8% (n = 32) at the 6-month follow-up (P < 0.001) (Table 1). Immediately after surgery, of the 25 patients with an LDM > 15 mm only 18 also had a BND > 15 mm.
Neither TVT–urethra distance nor the tape position relative to the urethra seemed to have an association with hypermobility (data not shown). Table 2 summarizes the preoperative and postoperative sonographic parameters in relation to cure rates. It is noteworthy that patients with recurrence or secondary cure had a significantly lower postoperative bladder neck mobility than had the group of patients with primary cure (P < 0.05) (Table 2). Patients in whom postoperative bladder neck mobility was reduced to ≤10 mm had a 2.6-fold higher risk of therapeutic failure (Table 3), whereas postoperative hypermobility of the bladder neck (LDM > 15 mm) reduced the risk of failure (relative risk = 0.4; data not shown). Analysis of pre- and postoperative mobility and hypermobility in relation to the occurrence of postoperative complications revealed no correlations (data not shown).
|Parameter||Primary cure (n = 151)||Secondary cure (n = 20)||No cure (n = 20)||P|
|LDMpreop (mm, median (range))||13.9 (2–30)||13.3 (5–24)||11.2 (3–25)||0.3493|
|LDMpostop (mm, median (range))||10.0 (1–32)||7.5 (1–21)||7.0 (2–19)||0.0394*|
|Hypermobilitypreop (n (%))||67 (44.4)||7 (35.0)||6 (30.0)||0.3803|
|Hypermobilitypostop (n (%))||23 (15.2)||1 (5.0)||1 (5.0)||0.2333|
|MUCPpreop (cmH20, median (range))||44 (19–102)||32 (18–68)||30 (12–68)||0.0053**|
|Parameter||Cure (n (%)) (n = 171)||No cure (n (%)) (n = 20)||Failure rate (%)||RR||P|
|LDM ≤ 10 mm||99 (57.9)||16 (80.0)||13.9||2.6||0.0451*|
|LDM > 10 mm||72 (42.1)||4 (20.0)||5.3|
|MUCPrest < 30 cmH2O||26 (15.2)||8 (40.0)||23.6||3.1||0.0061**|
|MUCPrest ≥ 30 cmH2O||145 (84.8)||12 (60.0)||7.6|
|LDM ≤ 10 mm, MUCPrest < 30 cmH2O||15 (8.8)||7 (35.0)||31.8||4.1||0.0005***|
|LDM > 10 mm or MUCPrest ≥ 30 cmH2O||156 (91.2)||13 (65.0)||7.7|
There was no significant increase in MUCP after TVT insertion (Table 1). Six months after the TVT procedure, there was a median MUCPrest of 46 (range, 10–109) cmH2O. Patients with therapeutic failure or secondary cure had a significantly lower median MUCPrest than had patients with primary cure (P < 0.01) (Table 2). The risk of therapeutic failure was 8.1 times higher in patients with a hypotonic urethra compared with those with a MUCPrest ≥ 20 cmH2O and it was 3.1 times higher in patients with a MUCPrest < 30 cmH2O compared with those with a MUCPrest ≥ 30 cmH2O (Table 3). A MUCPrest < 20 cmH2O or < 30 cmH2O with a low postoperative bladder neck mobility ⩽ 10 mm was associated with an 11.3 and 4.1 times higher risk of surgical failure, respectively (Table 3). However, it must be borne in mind that only seven patients had an MUCPrest < 20 cmH2O, so this result must be interpreted with caution.
Patients with postoperative funneling of the bladder neck had a median MUCPrest of 33 (range, 12–75) cmH2O, which was significantly below the value of 43 (range, 18–102) cmH2O (P < 0.01) in patients without postoperative funneling. No association was found between MUCPrest and the occurrence of postoperative complications such as voiding problems, urge symptoms, or de novo urge incontinence.
TVT is designed to provide support at the level of the mid-urethra and is therefore not expected to have any significant effect on mobility in the area of the bladder neck1, 6, as has been confirmed by perineal6, 10 and introital7 ultrasound studies. Nevertheless, we identified a significant reduction of bladder neck mobility as well as hypermobility after TVT insertion. Masata et al.8 also found a significant decrease in mobility along the entire length of the urethra 2 months after TVT insertion. Significantly diminished mobility with stabilization of the bladder neck region by the TVT is also suggested by the observation that downward movement of the urethrovesical junction during Valsalva maneuver is significantly reduced9.
Other investigators report a decreased mobility of the proximal urethra after TVT insertion, as demonstrated by means of the Q-tip test at rest and during straining15, 16. Studies performed with the Q-tip test15, 17 or lateral cystourethrography12 suggest that some preoperative mobility of the urethra appears to have a beneficial effect on the outcome of surgery since patients with therapeutic failure exhibited significantly restricted urethral mobility compared with patients whose operations were successful. Reduced bladder neck mobility before treatment did not preclude surgical success in general, but markedly reduced the chances of postoperative cure.
In this study, we did not find an association between decreased preoperative bladder neck mobility and the likelihood of cure, but patients in whom surgery failed or cure was delayed did show significantly lower postoperative mobility. In contrast, postoperative hypermobility of the proximal urethra reduced the risk of failure and thus had a protective effect. These results suggest that not only reduced preoperative bladder neck mobility but also reduced postoperative mobility appears to be associated with a significantly poorer cure rate.
The success of TVT insertion does not depend on correction of urethral hypermobility7, 10, 14–17. Quite the opposite is true: good mobility of the proximal urethra improves the likelihood of successful outcome12, 15, 17. Some investigators15, 17 attribute their observations mainly to the mechanisms of action of the TVT. When intra-abdominal pressure increases, urethral closure occurs due to kinking of the proximal urethra around the TVT7, 10, 25, 26. Moreover, the urethra and the surrounding tissue are compressed between the TVT and the symphysis10, 26. Both dynamic kinking and compression appear to depend on good mobility of the proximal urethra15, 17.
In contrast, the aim of the colposuspension procedure according to Burch is to stabilize the bladder neck area and thus to eliminate urethral hypermobility, which appears to be the crucial determinant of postoperative success in this approach19; patients undergoing Burch colposuspension had significantly higher recurrence rates when postoperative urethral hypermobility was diagnosed19.
It appears that the effectiveness of the TVT sling does not depend on changes in MUCPrest27, 28; our results confirm this. Failure of the TVT procedure is more likely in patients with a hypotonic urethra, although no uniform definition exists2, 29. In some studies a hypotonic urethra is defined as MUCPrest < 20 cmH2O30 or < 30 cmH2O31, the definition used here. In others, it is defined as a Valsava leak point pressure of < 60 cmH2O28, 29. The cure rates reported under these circumstances are 73%30 and 74%31, which is below the rate in patients with normal urethral closure pressure.
In our study, patients in whom surgery failed or cure was delayed had a significantly lower MUCPrest compared with patients with primary postoperative cure. An MUCPrest < 20 cmH2O or < 30 cmH2O was associated with a markedly higher risk of surgical failure. However, these conclusions are limited by the fact that a small sample of only seven patients had an MUCPrest < 20 cmH2O.
Patients with little urethral mobility combined with a low urethral closure pressure appear to be particularly predisposed to therapeutic failure28, 30, 31. For instance, Liapis et al.31 found a significantly higher cure rate in patients with a hypotonic but mobile urethra as opposed to those with a hypotonic and ‘immobile’ urethra. In our study, the subsets of patients with bladder neck mobility ≤ 10 mm had 11.3 and 4.1 times higher risks of surgical failure when poor mobility was associated with an MUCPrest of <20 cmH2O and <30 cmH2O, respectively. Again, one must bear in mind that only a small number of patients had an MUCPrest of <20 cmH2O.
Postoperative funneling of the bladder neck is associated with a poorer outcome of TVT insertion. Patients with funneling had a significantly lower MUCPrest than did those without funneling. The association of funneling and a low urethral closure pressure has also been reported by other investigators29, 32. These observations suggest that an inadequate closure pressure of the urethra promotes the occurrence of bladder neck funneling as a morphological correlate.
The objective cure rates achieved in our patients treated by TVT insertion correspond to those reported in the literature. Published continence rates after an average of 3 years range from 86 to 90%33, 34 while 89.5% of our study population were cured after 36 months. The incidence of complications in our patients is likewise in the range described in the literature3, 4, 30, 33–35.
Introital ultrasound is a suitable diagnostic tool with which to evaluate the postoperative mobility of the urethra and identify hypermobility. The results of our study indicate a significant decrease in mobility and hypermobility of the bladder neck after TVT insertion, but also demonstrate that correction of increased mobility is not a prerequisite for the success of the intervention. On the contrary, it seems to be advantageous to have good postoperative mobility of the proximal urethra. A low MUCPrest appears to be associated with a poorer outcome after TVT insertion. Patients who have this risk factor in combination with reduced urethral mobility seem to fare especially poorly after TVT.