1 ABSTRACT: Objective: To compare intermittent catheterization, sling mobilization and sling transection for treatment of urinary retention after mid-urethral sling surgery. Methods: Data registered in the Norwegian Female Incontinence Registry from 1998 to 2013 were analyzed in this cohort study to compare subjective and objective outcomes after intermittent catheterization, sling mobilization and sling transection as management of postoperative urinary retention after mid urethral sling surgery. Subjective outcomes were degree of symptom bother and the percentage of women stating very satisfied at the postoperative follow-up. The objective outcome was leakage at a cough-jump pad stress test. Results: Intervention due to urinary retention was performed in 585 of 18772 women (3.1 %). Women who had their sling mobilized or had intermittent catheterization, scored better on all postoperative outcomes than those who had their sling transected (P < 0.001). Sling transection was more often needed after intermittent catheterization than after mobilization (P =0.023). No outcome differences were found between intermittent catheterization only and sling mobilization only. Intention to treat analysis showed that women who underwent sling mobilization as the primary procedure significantly more often had a negative stress test (P = 0.033) and were more often very satisfied with the treatment (P =0.006) than those who were primarily catheterized. Conclusion: Sling mobilization is a more successful treatment than intermittent catheterization or transection for urinary retention after mid-urethral sling surgery.
2 Clinical Trial Registration: Clinical trial registration was not applicable because this study is based on an analysis of anonymous data from The Norwegian Female Incontinence Registry.
3 Introduction: Since Ulmsten and Petros introduced the tension-free vaginal tape (TVT) procedure in 1996, the minimally invasive mid-urethral sling (MUS) operations have rapidly become the procedures of choice in the operative management of stress urinary incontinence. 1 Numerous studies have shown good short- and long-term subjective and objective outcomes both for stress urinary incontinence and stress predominant mixed urinary incontinence. 2-4 Although serious complications are rare, the prevalence of incomplete bladder emptying has been reported from 2.3-26 % following these procedures. 5,6 Incomplete bladder emptying ranges from spontaneous micturition with high post-void residuals to complete urinary retention. Incomplete bladder emptying exposes the women to severe bother that decreases her quality of life, may cause urinary tract infections and pain and may necessitate prolonged intermittent catheterization. 6 Proposed risk factors for this unwanted outcome are old age, concomitant surgery, vaginal vault prolapse, poorly performed intraoperative cough test, low preoperative urinary flow rate and low detrusor contractility. 6 Postoperative urinary retention after MUS procedures may in many cases resolve after a few days with intermittent catheterization. This is therefore often the preferred method. 7-10 If the postoperative urinary retention is not resolved with intermittent catheterization, transection of the sling has commonly been performed. 7,8 However, transection may lead to recurrence of stress incontinence, reported in up to 61% of the women. 11,12 Operative mobilization of the sling is an intervention in which the vaginal wound is opened and the sling pulled down 5 10 mm. 13,14 If adequate voiding is not obtained, the procedure may be repeated. Sling mobilization has been reported to successfully resolve postoperative urinary retention after MUS operations in 87-100% of the cases. 13-18
4 In Norway 30 of 41 gynecological community and tertiary care departments performing female incontinence surgery have since 1998 reported continuously pre- and postoperative subjective and objective data, type of incontinence procedure and complications to the Norwegian Female Incontinence Registry. Operative mobilization of the sling to relieve urinary retention has been reported to the Registry since 2008. The aim of this study was to compare postoperative subjective stress incontinence, treatment satisfaction and objective leakage in women who had been treated with intermittent catheterization, sling transection or sling mobilization for urinary retention after mid-urethral sling surgery. Materials and Methods: Pre- and postoperative data registered in the Norwegian Female Incontinence Registry for women who had undergone an MUS operation for stress or mixed urinary incontinence in 30 gynecological community and tertiary care departments were analyzed in this cohort study. The operations were performed between September 1 st 1998 and December 5 th 2013. Women diagnosed with urinary retention after MUS surgery were treated with intermittent catheterization, sling mobilization or sling transection. We used preoperative, operative and 6-12 months postoperative follow-up data recorded in the Registry for this study. The subjective outcomes chosen were treatment satisfaction and stress incontinence index, and the objective outcome was leakage during a cough-jump pad stress test. 19,20 The MUS operations in use were TVT, TVT-O (tension-free vaginal sling-obturator), TOT (transobturator sling) and AJUST (adjustable single-incision sling). Women with concomitant pelvic organ prolapse surgery were not included.
5 Participating departments that report data to the Registry use a validated short-form urinary incontinence disease-specific questionnaire for subjective data. 19 The data are registered preoperatively and at the 6 12 months follow-up visit as well as at later visits. The questionnaire generates a stress incontinence index from 0 to 12 that indicates degree of subjective symptom bother. A high score signifies more severe bother than a low score. The questionnaire also contains a question about treatment satisfaction. The choices are: very satisfied, moderately satisfied, neither satisfied nor dissatisfied, moderately dissatisfied or very dissatisfied. Our objective outcome was a standardized and reproducible cough-jump pad stress test used by all contributing departments before surgery and at subsequent follow-ups. This stress test quantifies grams of leakage by the increase in pad weight after 3 coughs and 20 jumping jacks with 300 ml bladder volume. 20 Urinary retention is not defined in the Registry. For the purpose of this study we defined that a woman had urinary retention when the surgeon found indication to perform bladder catheterization, sling mobilization or sling transection after an MUS procedure. When postoperative retention occurred, intermittent catheterization was initiated in most cases. If the retention was not relieved, sling transection was performed when the surgeon found it appropriate. However, from 2008 information on sling mobilization was added to the Registry. Under local or general anesthesia the vaginal incision was opened and the sling pulled down 5 10 mm to relieve the obstruction. The mobilization was in most cases performed within the first postoperative week. 13,14 The mobilization procedure could be repeated. Catheterization is registered in the Registry when it was performed for more than 1 week. Date of mobilization and date of start or ending of catheterization is not registered in the Registry. The date of tape transection is registered in the Registry, but not the method used, i.e. midline or lateral transection.
6 The three interventions for treating urinary retention after MUS operations were compared and analyzed in two different ways. First they were compared and analyzed according to the ultimate method (i.e. women who had the sling transected after initial catheterization or mobilization, were counted as transected, Fig 1). Secondly an intention-to-treat analysis was performed in which the methods were compared according to the first method used (i.e. women who had the sling transected following a period of intermittent catheterization were counted as catheterized). The Norwegian Female Incontinence Registry is approved and regulated by The Norwegian Data Protection Authority. Written, informed consent is obtained from all participating women, and clinical data are exported and stored in de-identified format in the Registry. Approval of the clinical data studies is given from the Department Heads and institutional personal data officers of all the collaborating hospitals. The Regional Committee for Medical and Health Research Ethics of South East Norway considered our study to represent a quality assurance measure for treatment given and therefore without need of formal committee evaluation and approval. Statistics: SPSS version 22 was used and descriptive statistics calculated for the aims above. Many of the data were not normally distributed and are therefore presented as median and 5 and 95 percentiles. Mann-Whitney U-test was used for continuous variables and chi square test for dichotomous variables. We used a significance level of 5 %. To adjust for potential imbalance in preoperative values when analyzing continuous variables, a linear regression analysis was performed, using the preoperative values as covariates. Variables with markedly skewed distributions were dichotomized before being analyzed in a regression analysis. To adjust for potential imbalance in preoperative values when analyzing dichotomous variables, logistic regression analysis was performed, using the preoperative values as covariates. Each outcome variable was calculated as a per protocol analysis in which the
7 denominator was obtained by subtracting women with missing data from the total number of women. Results: From September 1 st 1998 to December 5 th 2013, 18772 stress and mixed incontinent women underwent 15190 TVT, 2140 TVT-O, 889 TOT and 553 AJUST operations without concomitant pelvic organ prolapse surgery and were reported to The Norwegian Female Incontinence Registry. Five hundred and eighty-five of the 18772 women (3.1%) were registered as having had one or more of the three interventions for postoperative urinary retention and were thus included in this study. The median time to follow-up was 9 months, 5% and 95% percentiles were 5 and 19 months. Intermittent catheterization was performed in 328 women, 121 had the sling primarily transected, and 136 underwent sling mobilization (Fig 1). Intermittent catheterization, sling mobilization and sling transection were performed in respectively 27, 18 and 20 centers. No significant differences in any of the postoperative subjective and objective outcomes were found between women who were catheterized as their only treatment (i.e. not in need of later transection of the sling) and those who had the sling mobilized as their only treatment (i.e. not in need of later transection of the sling) (Table I, A vs B). However, the total group of women who had the sling transected (n=197) and the women who only had a sling transection (n=121), leaked significantly more during the postoperative pad stress test, had a higher postoperative stress
8 incontinence index and were less satisfied than the women in the other two groups (Table I, A vs. C,B vs. C, A vs. D and B vs. D). Subsequent sling transection was performed more often in the women who primarily underwent intermittent catheterization 62 of 328, (19%) than in those who had the sling primarily mobilized 14 of 136, (10%) (P = 0.027). Median time from the MUS operation to transection was 24.5 weeks (5% and 95% percentiles 2 and 188 weeks). In the intention-to-treat analysis the women who had sling mobilization as primary treatment leaked significantly less during the postoperative pad stress test (P = 0.033) and were significantly more often very satisfied (P = 0.006) with their treatment than those who were primarily catheterized. (Table II). A subgroup analysis done on the 12 centers performing all three procedures found similar results for all except one outcome variable. When women who were catheterized and not transected were compared with women who had the tape mobilized and not transected, a significant difference (p=0.05) favoring tape mobilization was found for percentage very satisfied. Table III show the rates of recurrence of stress incontinence in the different treatment groups when the results from the postoperative cough-jump pad stress test is dichotomized into negative or positive (defined as 0 grams). This differs from the analyses presented in Tables I, and II in which the results from the stress test were analyzed as a continuous variable. Recurrence of stress incontinence was seen significantly less often after tape mobilization than after tape transection (P = 0.001) or intermittent catheterization (P = 0.023). There was no significant difference in recurrence of stress incontinence between the women who had intermittent catheterization and those who had tape transection. (P = 0.114).
9 Discussion: Urinary retention was found in 3.1% of women after a MUS procedure. No differences were found in postoperative satisfaction or recurrence of stress incontinence in women who had intermittent catheterization and those who had tape mobilization when there was no need for a later tape transaction. However, women who had intermittent catheterization significantly more often underwent subsequent tape transection than those who had tape mobilization. Women who had their sling transected had significantly more leakage on the 6-12 months cough-jump pad stress test, were less satisfied and had a higher postoperative stress incontinence index than the women in the other two groups. Compared with tape mobilization recurrence of stress incontinence, defined as a positive stress test (> 0 gram), was found significantly more often after both transection and after catheterization. In the intention-to-treat analysis the women who had sling mobilization as their primary treatment had significantly less leakage during the postoperative pad stress test and were more often very satisfied with their treatment than women who had intermittent catheterization as their primary treatment. The median post-operative stress test is zero in all groups. Nevertheless, there is a significant difference between both catheterization and mobilization versus transection. One might therefore debate whether this difference is clinically significant. However, calculating the percent postoperative positive stress tests (> 0 gram) in women catheterized not transected, mobilized not transected, total transected and only transected, the results were 7.4, 2.6, 34.4 and 24.7 respectively. We therefore find that the observed differences are clinically significant.
10 When performed correctly mobilization of the sling relieves the retention and the woman remains continent. In our experience it is easiest and probably best performed within the first few postoperative days, but Glavind found the procedure to be feasible up to 21 days after the primary operation. 18 The time between the MUS operation and sling mobilization is not registered in the Registry. Glavind performed sling mobilization on average 11.6 days (range 2-21), while Rautenberg performed the procedure usually within 1 3 days. 14,18 Probably due to the impairment of the hammock effect, more involuntary leakage is seen after tape transection than after intermittent catheterization or tape mobilization. Our retention rate of 3.1% is comparable to the studies by Kuuva (2.3%) and Glavind (3.9%), but not the study by Takacs where urinary retention was found as high as 26%. 5,6,18 Recurrence of stress incontinence occurred in 25 % after sling transection only in our study, while others have reported up to 61%. 11,12 After sling mobilization a recurrence rate of 5.9 % was found by Glavind, corresponding well with 7 % in our study (Table III). 18 Intermittent catheterization and sling mobilization are believed to be successful alternatives to sling transection for treatment of postoperative urinary retention after MUS surgery. Sling mobilization has been reported with success rates of up to 100 %. 13-18. However, many surgeons still prefer to use intermittent catheterization for simplicity and fear of recurrence of incontinence after sling mobilization. Hong et al. found that normal spontaneous voiding was resumed in 87.5 % after intermittent catheterization, corresponding with 81 % in our study (Fig I). 8 The fear of incontinence seems based on our study to be unwarranted. We found that only 10% of the women who had the sling mobilized underwent sling transection at a later date and that there was significantly less recurrence of stress incontinence in the mobilization group compared to the women who had the sling transected.
11 The major strength of our study is that it is based on a large national database comprising 80% of MUS operations in Norway performed on an unselected patient population by a variety of experienced and inexperienced surgeons. To our knowledge, this is the only published study with sufficient sample size to detect statistically significant differences between these three postoperative interventions for urinary retention after MUS procedures. Registry data from unselected and heterogeneous patients and doctors has the advantage of reporting the conditions in a real clinical setting. Registry studies can therefore have a high external validity. It has been reported that combined pelvic organ prolapse and incontinence surgery increase the risk of postoperative urinary retention. 21 Not including women with concomitant pelvic organ prolapse surgery strengthens the conclusion of our study. Urinary retention is not defined in the Registry. For this study our definition of urinary retention was that the patient had undergone bladder catheterization, sling mobilization or sling transection after an MUS procedure. Because the indications to perform these interventions varied from department to department and from surgeon to surgeon; this may have contributed to selection bias. We recognize that if some departments systematically use transection of the sling as treatment for total retention and one of the other two interventions for partial urinary retention, this might impact the results and complicate interpretations. However, as indication for intervention is not registered we had no means to adjust for this fact in the analyses of the current study. Another potential selection bias is the fact that not all centers perform all three interventions. However, when outcomes from 12 centers performing all three procedures were compared with outcomes from all centers, similar results were found for all, except one outcome variable. The
12 marked differences in interventions performed by different centers, may still have contributed to selection bias, but not in our opinion to a serious selection bias The surgical technique by which a sling was mobilized and whether it was transected laterally or in the midline are not registered in the Registry. This also is a limitation of the study. Furthermore, we realize the fact that the imbalance in patient numbers registered as receiving either catheterization or mobilization (more than twice as many having had catheterization than mobilization) may have influenced the results. The imbalance was due to the fact that mobilization was only registered in the Registry after 2008. The use of registry data always carries the possibility of inaccuracies in the individual entries and the results must therefore always be interpreted with this in mind. We would have liked to analyze possible differences between different sling systems. However, the numbers of TVTO, TOT and AJUST procedures were not sufficient to obtain adequate test power. A subgroup analysis performed on the TVTs (which constituted 80 % of the procedures) generated similar results as for the total cohort. Our study indicates that mobilization of the sling should be preferred to transection of the sling or prolonged intermittent catheterization in the case of urinary retention after MUS procedures. Intermittent catheterization may lead to physical and social discomfort and it is impossible to know when voiding will be normalized. Normalized voiding is however, quickly achieved in most cases by mobilization. Our study found that nearly twice as many women needed tape transection after intermittent catheterization (19%) than after mobilization (10%), and that the women with tape transection had a higher risk of experiencing recurrent stress incontinence. This suggests that tape mobilization should be the method of choice when faced with urinary retention
13 after MUS surgery. We acknowledge the limitations of a registry-based study, but believe that our large, unselected database strongly supports this view. The best way of comparing different interventions is by randomizing women prospectively in controlled trials (RCTs) but this would require a very large multicenter study. To our knowledge no such RCT has so far been published. We believe it is ethically sound to encourage surgeons to prefer sling mobilization rather than long term intermittent catheterization for postoperative urinary retention after MUS operations, despite lacking RCTs. Concluding message: Our findings indicate that sling mobilization is a more successful treatment than intermittent catheterization or sling transection for urinary retention after MUS surgery. Acknowledgments: (written permissions on file): The authors thank Tomislav Dimoski, MBA, system developer at the Norwegian Female Incontinence Registry for data base and IT assistance. We thank the statisticians Leiv Sandvik at the Unit of Biostatistics and Epidemiology, Oslo University Hospital and Petter Mowinckel at the Pediatric Department, Oslo University Hospital, for statistical support Part of the work has been presented in a short abstract form at the 44 th Annual Meeting of the International Continence Society (ICS 2014), October 20-24, Rio, Brazil, at the 29 th Bi-Annual meeting of the Nordic Urogynecological Association (NUGA 2015), January 29-31, Stockholm,
14 Sweden and at the Annual Meeting of the Norwegian Association for Obstetricians and Gynecologists, October 23-25, 2014, Trondheim, Norway.
15 References: 1. Ulmsten U, Henriksson L, Johnson P, Varhos G. An ambulatory surgical procedure under local anesthesia for treatment of female urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct. 1996;7(2):81-85; discussion 85-86. 2. Nilsson CG, Palva K, Aarnio R, Morcos E, Falconer C. Seventeen years' follow-up of the tensionfree vaginal tape procedure for female stress urinary incontinence. Int Urogynecol J. 2013;24(8):1265-1269. 3. Ogah J, Cody JD, Rogerson L. Minimally invasive synthetic suburethral sling operations for stress urinary incontinence in women. Cochrane Database Syst Rev. 2009(4):CD006375. 4. Svenningsen R, Staff AC, Schiotz HA, Western K, Kulseng-Hanssen S. Long-term follow-up of the retropubic tension-free vaginal tape procedure. Int Urogynecol J. 2013;24(8):1271-1278. 5. Kuuva N, Nilsson CG. A nationwide analysis of complications associated with the tension-free vaginal tape (TVT) procedure. Acta Obstet Gynecol Scand. 2002;81(1):72-77. 6. Takacs P, Medina CA. Tension-free vaginal tape: poor intraoperative cough test as a predictor of postoperative urinary retention. Int Urogynecol J Pelvic Floor Dysfunct. 2007;18(12):1445-1447. 7. Croak AJ, Schulte V, Peron S, Klingele C, Gebhart J, Lee R. Transvaginal tape lysis for urinary obstruction after tension-free vaginal tape placement. J Urol. 2003;169(6):2238-2241. 8. Hong B, Park S, Kim HS, Choo MS. Factors predictive of urinary retention after a tension-free vaginal tape procedure for female stress urinary incontinence. J Urol. 2003;170(3):852-856. 9. Shukla A, Paul SK, Nishtar A, Bibby J. Factors predictive of voiding problems following insertion of tension-free vaginal tape. Int J Gynaecol Obstet. 2007;96(2):122-126. 10. Bailey C, Matharu G. Conservative management as an initial approach for post-operative voiding dysfunction. Eur J Obstet Gynecol Reprod Biol. 2012;160(1):106-109. 11. Klutke C, Siegel S, Carlin B, Paszkiewicz E, Kirkemo A, Klutke J. Urinary retention after tensionfree vaginal tape procedure: incidence and treatment. Urology. 2001;58(5):697-701. 12. Viereck V, Rautenberg O, Kociszewski J, Grothey S, Welter J, Eberhard J. Midurethral sling incision: indications and outcomes. Int Urogynecol J. 2013;24(4):645-653. 13. Chang WC, Sheu BC, Huang SC, et al. Postoperative transvaginal tape mobilization in preventing voiding difficulty after tension-free vaginal tape procedures. Int Urogynecol J. 2010;21(2):229-233. 14. Rautenberg O, Kociszewski J, Welter J, Kuszka A, Eberhard J, Viereck V. Ultrasound and early tape mobilization--a practical solution for treating postoperative voiding dysfunction. Neurourol Urodyn. 2014;33(7):1147-1151. 15. Nguyen JN. Tape mobilization for urinary retention after tension-free vaginal tape procedures. Urology. 2005;66(3):523-526. 16. Glavind K, Glavind E. Treatment of prolonged voiding dysfunction after tension-free vaginal tape procedure. Acta Obstet Gynecol Scand. 2007;86(3):357-360. 17. Price N, Slack A, Khong SY, Currie I, Jackson S. The benefit of early mobilisation of tension-free vaginal tape in the treatment of post-operative voiding dysfunction. Int Urogynecol J Pelvic Floor Dysfunct. 2009;20(7):855-858. 18. Glavind K, Shim S. Incidence and treatment of postoperative voiding dysfunction after the tension-free vaginal tape procedure. Int Urogynecol J. 2015. 19. Kulseng-Hanssen S, Borstad E. The development of a questionnaire to measure the severity of symptoms and the quality of life before and after surgery for stress incontinence. BJOG. 2003;110(11):983-988.
16 20. Berild GH, Kulseng-Hanssen S. Reproducibility of a cough and jump stress test for the evaluation of urinary incontinence. International Urogyecological Journal. 2012;2012 april 19. 21. Wang KH, Neimark M, Davila GW. Voiding dysfunction following TVT procedure. Int Urogynecol J Pelvic Floor Dysfunct. 2002;13(6):353-357; discussion 358.