- Email: [email protected]
Can we predict urinary stress incontinence by using demographic, clinical, imaging and urodynamic data?
European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Can we predict urinary stress incontinence by using demographic, clinical, imaging and urodynamic data? Edyta Wlaz´lak a,*, Grzegorz Surkont a, Ka L. Shek b, Hans P. Dietz b a b
Clinic of Operative and Oncologic Gynecology, Medical University of Lodz, Wilenska 37, 94029 Lodz, Poland Sydney Medical School Nepean, University of Sydney, Nepean Hospital, Penrith, Sydney, NSW 2750, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 21 February 2015 Received in revised form 5 May 2015 Accepted 24 July 2015
Objective: It has been claimed that urethral hypermobility and resting urethral pressure can largely explain stress incontinence in women. In this study we tried to replicate these findings in an unselected cohort of women seen for urodynamic testing, including as many potential confounders as possible. Study design: This study is a retrospective analysis of data obtained from 341 women. They attended for urodynamic testing due to symptoms of pelvic floor dysfunction. We excluded from the analysis women with a history of previous anti-incontinence and prolapse surgery. All patients had a standardised clinical assessment, 4D transperineal pelvic floor ultrasound and multichannel urodynamic testing. Urodynamic stress incontinence (USI) was diagnosed by multichannel urodynamic testing. Its severity was subjectively graded as mild, moderate and severe. Candidate variables were: age, BMI, symptoms of prolapse, vaginal parity, significant prolapse (compartment-specific), levator avulsion, levator hiatal area, Oxford grading, midurethral mobility, maximum urethral pressure (MUP), maximum cough pressure and maximum Valsalva pressure reached. Results: On binary logistic regression, the following parameters were statistically significant in predicting urodynamic stress incontinence: age (P = 0.03), significant rectocele (P = 0.02), max. abdominal pressure reached (negatively, P < 0.0001), midurethral mobility (P = 0.0004) and MUP (negatively, P < 0.0001). On multivariate analysis, accounting for multiple interdependencies, the following predictors remained significant: max. abdominal pressure reached (negatively, P < 0.0001), cough pressure (P = 0.006), midurethral mobility (P = 0.003) and MUP (negatively, P < 0.0001), giving an R2 of 0.24. Conclusions: Mid-urethral mobility and MUP are the main predictors of USI. Demographic and clinical data are at best weak predictors. Our results suggest the presence of major unrecognised confounders. ß 2015 Published by Elsevier Ireland Ltd.
Keywords: Stress urinary incontinence 3D/4D ultrasound Urodynamic testing Pelvic floor function Levator ani
Introduction Stress urinary incontinence (SUI) is the most common type of urinary incontinence in women [1]. A first clinical description of SUI was published in 1912 [2]. Kelly described the open vesical neck as a possible cause of stress urinary incontinence (SUI). After introducing an operation that, in his opinion, cured SUI he postulated that the success of his procedure resulted from narrowing of the vesical neck. In 1922 Bonney proposed loss of urethral support as the main cause of stress incontinence. In his opinion, Kelly’s operation cured SUI through improving urethral support [2]. Jeffcoate and Roberts in 1949 showed that many stress incontinent women exhibited a loss of the urethrovesical angle
* Corresponding author. Tel.: +48 501587964; fax: +48 426860471. E-mail address: [email protected] (E. Wlaz´lak). http://dx.doi.org/10.1016/j.ejogrb.2015.07.012 0301-2115/ß 2015 Published by Elsevier Ireland Ltd.
[2]. Krantz postulated that the pubourethral ligaments, fibromuscular structures connecting the urethra to the pelvic sidewall, affect pressure transmission [3]. In 1960, Enhorning claimed that transmission of abdominal pressure was reduced in women with stress incontinence because the urethra descended below the abdominal pressure zone [2]. Another related pathophysiological concept is the ‘‘hammock hypothesis’’, which suggests that the vagina dorsal to the urethra provides a backboard against which increasing intra-abdominal forces compress the urethra [4]. The idea that primary urethral weakness could cause urinary incontinence independent of vaginal weakness appeared in a proposed classification by Blaivas et al. [5]. In their classification, they named this Type III incontinence to distinguish it from Types I and II, each of which showed movement, while Type III did not. Type III has now been largely replaced by the term intrinsic sphincter deficiency (ISD) to focus attention on urethral factors, which appeared to be independent of vaginal
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
position and mobility. There is a growing clinical impression that some degree of ISD may exist in many patients who, until recently, were thought to have only hypermobility as a cause of their incontinence [2]. Progress made in the field of imaging, especially using translabial ultrasound, now allows quantifying different aspects of urethrovesical function [6–8]. As a result it should be possible to advance understanding of the pathophysiology of urinary incontinence in women through the simultaneous analysis of anatomical and functional findings. Lately it has been claimed that urethral hypermobility and resting urethral pressures can largely explain stress incontinence in women, with maximal urethral closure pressure, not urethral support, likely to be the factor most strongly associated with stress incontinence [9]. In this study we tried to replicate these findings in an unselected cohort of women seen for urodynamic testing, including as many potential confounders as possible. We also performed the same analysis for the diagnosis of urodynamic stress incontinence (USI) for comparison. Materials and methods This study is a retrospective analysis of data obtained on urodynamic testing at a tertiary urogynaecology unit. Between January 2009 and April 2010, 454 women attended the unit for urodynamic testing due to symptoms of pelvic floor dysfunction. We excluded 113 women with a history of previous antiincontinence and prolapse surgery, leaving 341 data sets. All subsequent analysis pertains to those 341 patients. All patients had a standardised in-house, nonvalidated interview including symptoms of stress urinary incontinence as defined by the International Continence Society [10], a clinical assessment using the ICS POP-Q [11], along with a 4D transperineal pelvic floor ultrasound [12] using GE Kretz Voluson 730 expert and Voluson i systems, and multichannel urodynamic testing (Neomedix Acquidata, Neomedix, Sydney, Australia). Significant prolapse was defined as a prolapse stage 2 (ICS POP-Q system). Levator contraction strength was assessed digitally, using the Modified Oxford Grading (MOS) [13]. Urethral sphincter function was assessed with urethral profilometry. Maximal urethral pressure (MUP) was obtained with a 5F single-lumen water-perfused catheter with a freehand pull-through technique. Maximum abdominal pressure and maximum cough pressure were obtained both in the standing position on Valsalva manoeuvre and during coughing. Urodynamic stress incontinence (USI) was diagnosed in women with a positive stress test during multichannel urodynamic examination. Its severity was subjectively graded as mild, moderate, marked and severe depending on the amount of urine loss and the abdominal pressure at which this occurred. We used maximum cough pressure and maximum Valsalva pressure obtained during urodynamic testing to measure the efficacy of the response to increased demand on the continence mechanism [9,10]. All ultrasound images were analysed offline using proprietary software (4D View v 10), with the computer operators (GS and EW) blinded against all clinical data. Urethral mobility was described by vectors of movement from rest to a maximum Valsalva of 6 equidistant points marked along the length of the urethra from bladder neck (point 1) to external urethral meatus (point 6) in the midsagittal plane, as previously described [6]. For midurethral mobility analysis we used vector 4. Hiatal dimensions at rest and on Valsalva were measured in the plane of minimal hiatal dimensions, as previously described [14]. Levator trauma was identified by tomographic ultrasound (TUI) as previously described [15]. Binary outcomes of stress incontinence (SI) and USI were analysed using binary logistic regression. Both univariate and
115
multivariate models were developed for SI and USI. A backwards elimination approach was used to optimise models. Candidate variables were age, BMI, symptoms of prolapse, vaginal parity, significant prolapse (compartment-specific), avulsion, hiatal area, Oxford grading, midurethral mobility, MUP, maximum cough pressure and maximum Valsalva pressure reached. Ethical approval was obtained from the local human research ethics committee (SWAHS HREC 09-42). Results A test-retest series of urethral mobility vectors performed by two authors (EW and KLS) showed excellent interrater reliability, with an Intraclass Correlation Coefficient for single measures of 0.839 (95% CI 0.745–0.897) and 0.870 (95% CI 0.811–0.911) for segmental urethral mobility. A test-retest series for hiatal biometry also demonstrated excellent repeatability, consistent with multiple previous reports, with an ICC for single measures of 0.854 (CI 0.764–0.911) and 0.950 (CI 0.915–0.970) (for hiatal diameters), and 0.985 (CI 0.966–0.992) for area on Valsalva. All continuous parameters were tested for normality (histogram and Kolmogorov–Smirnov testing) and were found to be normally or near-normally distributed. The mean age of 341 women included in the analysis was 54 (19–89) years and BMI was 29 (17–59). Mean vaginal parity was 2.5 (0–10), and 89% were vaginally parous. A Vacuum or Forceps was reported by 25%. 23% of analysed women had had a hysterectomy. 75% of patients complained of stress urinary incontinence (SI), 68% of urge incontinence, 29% of frequency, 47% of nocturia, 28% of symptoms of voiding difficulty (hesitancy, straining and stop-start voiding), and 42% of symptoms of prolapse (lump or dragging sensation). On examination a cystocele stage 2 was found in 42%, significant central compartment prolapse in 9%, and a significant clinical rectocele (stage 2) in 43%. In 62% we detected any form of prolapse of stage 2 or higher. Mean Oxford grading was 2.4 (SD 1.2), mean bladder neck descent was 30.6 (SD 12.8) mm and mean midurethral mobility was 17.6 (SD 7.1) mm. Midsagittal hiatal diameter on Valsalva was 6.7 (1.2) cm and mean hiatal area on Valsalva 28.7 (SD 9.6) cm2. A total of 67 patients (20%) had an avulsion of the puborectalis muscle. Urodynamic findings were as follows: mean maximal urethral pressure (MUP) was 42 cmH2O (SD 20), mean maximum abdominal pressure was 87 cmH2O (SD 35) and mean maximum cough pressure was 110 cmH2O (SD 35). The urodynamic diagnosis was USI in 216 patients (65%). It was mild in 50, moderate in 68, marked in 34 and severe in 64 women. DO was diagnosed in 28%, voiding dysfunction in 31% of patients. On binary logistic regression analysis with SI as the outcome, the following parameters were (positively unless stated otherwise) predictive of SI: age (negatively, P = 0.0003), BMI (P = 0.0246), symptoms of prolapse (negatively, P = 0.0015), vaginal parity (P = 0.026), significant cystocele (negatively, P = 0.034), significant central compartment prolapse (negatively, P = 0.003), levator avulsion (negatively, P = 0.0001) and midurethral mobility (P = 0.018). MUP was not significantly related to SI. Multivariable analysis, accounting for multiple interdependencies, showed the following parameters as still significant: age (negatively, P < 0.0001), BMI (0.032), symptoms of prolapse (negatively, P = 0.0038), vaginal parity (P < 0.0001) and maximum abdominal pressure reached (negatively, P = 0.01). In Table 1 we included results of two different models used, with only statistically significant variables presented. On binary logistic regression, the following parameters were predictive of USI: age (P = 0.03), significant rectocele (P = 0.02), max. abdominal pressure reached (negatively, P < 0.0001), midurethral mobility (P = 0.0004) and MUP (negatively, P < 0.0001).
116
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
Table 1 Results of binary logistic regression modelling and multivariate modelling (accounting for multiple interdependencies) with SI as the outcome. Parameter
Binary logistic regression
Multivariable analysis
Age BMI Symptoms of prolapse Vaginal parity Cystocele stage 2 Centr. comp. stage 2 Levator avulsion Midurethral mobility Max. p abd (Valsalva)
0.0003 (neg) 0.0246 (pos) 0.0015 (neg) 0.026 (pos) 0.034 (neg) 0.003 (neg) 0.0001 (neg) 0.018 (pos) n.s.
<0.0001 (neg) 0.032 (pos) 0.0038 (neg) <0.0001 (pos) n.s. n.s. n.s. n.s. 0.01 (neg)
pos, positive association; neg, negative association; Centr. comp., central compartment; Max. p abd (Valsalva), maximum abdominal pressure reached on Valsalva during urodynamic exam.
Multivariable analysis, accounting for multiple interdependencies, showed the following parameters as remaining significant: MUP (negatively, P < 0.0001), rectocele (P = 0.01), max. abdominal pressure (negatively, P < 0.0001), max cough pressure (P = 0.003), avulsion (negatively, P = 0.03), and midurethral mobility (P = 0.0005). On univariate linear modelling with USI grade (0– 4) as outcome, the following parameters were significantly associated with USI: age (P < 0.0001), vaginal parity (P = 0.04), maximum abdominal pressure (negatively, P < 0.001), maximum cough pressure (negatively, P = 0.007), midsagittal hiatal diameter on Valsalva (P = 0.03), midurethral mobility (P = 0.0013) and MUP (negatively, P < 0.001). On multivariate analysis, accounting for multiple interdependencies (e.g. between significant cystocele, BND and midurethral mobility), the following predictors remained significant: maximal abdominal pressure (negatively, P < 0.0001), cough pressure (P = 0.006), midurethral mobility (P = 0.003) and MUP (negatively, P < 0.0001), giving an optimal R2 of 0.24. In Table 2 we included results of two different models used, with only statistically significant variables presented. Comment Despite progress made in stress urinary incontinence (SUI) operative treatment over the last two generations, cure rates have not risen substantially over 80% [9]. The mechanism of action of anti-incontinence procedures as well as reasons for failure are still poorly understood [7,16–18], as is the pathophysiology and aetiology of stress urinary incontinence [2]. Over the last century, several hypotheses have been proposed to explain stress urinary incontinence. These theories are based on clinical observations and focus primarily on the causative role of urethral support loss and urethral quality [2]. These hypotheses have been tested by comparing measurements of urethral support and function in women with primary stress urinary incontinence to the same measurements obtained in asymptomatic volunteers Table 2 Results of binary logistic regression modelling and multivariate modelling (accounting for multiple interdependencies) with USI as the outcome. Parameter
Binary logistic regression
Multivariable analysis
Age Rectocele stage 2 Max. p abd (Valsalva) Midurethral mobility Maximal urethral pressure Levator avulsion
0.03 (neg) 0.02 (pos) <0.0001 (neg) 0.0004 (pos) <0.0001 (neg) n.s.
n.s. 0.01 (pos) <0.0001 (neg) 0.0005 (pos) <0.0001 (neg) 0.003 (neg)
pos, positive association; neg, negative association; Max. p abd (Valsalva), maximum abdominal pressure reached on Valsalva during urodynamic exam.
who were recruited to be similar in age, race, and parity [9]. In this study maximal urethral closure pressure was the parameter that differed the most between groups being 43% lower in women with stress incontinence than asymptomatic women. Although operations that provide differential support to the urethra are effective, this study showed that urethral support may not be the predominant cause of stress incontinence [9]. In this retrospective study we tried to take into consideration as many factors as possible, which may have a confounding influence on clinical stress incontinence (SI) and urodynamic stress incontinence (USI). We analysed three main factors involved in the mechanism of stress continence: (1) urethral function as measured by maximal urethral pressure obtained with a 5F single-lumen fluid-perfused catheter; (2) urethral support, as determined by urethral mobility vectors (6) and (3) the efficiency of the response to increased demands on the continence mechanism, by assessing the observation of urethral loss relative to maximal cough and abdominal pressure. We have again confirmed that urethral resting pressure and mid-urethral mobility are predictors of USI, although we have been unable to replicate the findings reported by DeLancey [9]. While his model reached a Nagelkerke R2 of over 60%, ours at best reached about 24%. This very likely is due to differences between methods and populations. Our results suggest the presence of major unrecognised confounders in this real-world population of patients referred for urodynamic testing. One of them may be urethral kinking, that is, deformation of the urethra due to differential mobility of urethral segments, a topic that clearly requires further research. There may be other factors influencing the efficiency of continence mechanisms for example a reflex activation of urethral rhabdosphincter, levator ani and other fibromuscular structures on coughing [19]. Reflex contraction of the levator ani and external perineal muscles can be observed on translabial ultrasound during sudden increases in intra-abdominal pressure. These reflex contractions are almost universally present in nulliparous pregnant women. There seems to be a reduction in reflex contraction magnitude after childbirth, and this reduction may be associated with vaginal delivery. Their magnitude may be associated with postpartum stress urinary incontinence. The clinical significance of this finding is uncertain but this finding is consistent with the growing evidence of pelvic floor damage, both macroscopic and functional, or possibly ultrastructural, attributable to vaginal delivery [20]. The effect of pregnancy remains poorly defined, although it is likely that hormonal effects of pregnancy have an effect on urethral support and hence pressure transmission [21]. During a cough, normal PFM function is likely to produce timely compression of the urethra and vagina and additional external support to the urethra, reducing displacement, velocity, and acceleration. In women with SUI, who are likely to have weaker urethral attachments, this shortening contraction may not occur; consequently, the urethras of women with SUI may move further and faster for a longer duration [19]. Snooks and Swash [22,23] first brought attention to the importance of urethral denervation after childbirth and its possible contribution to urinary and faecal incontinence. Stress incontinence is frequently associated with a decline in the electrophysiological function of the pudendal nerve [21], the striated urethral sphincter [21], and the pelvic floor muscles [21,24]. Most recent studies continue to support the finding of prolonged pudendal nerve terminal motor latency in SI [21]. Urethral vasculature has also been postulated to play a role in the continence mechanism and different Doppler parameters have been studied to evaluate correlation with SI, although results are conflicting [25]. There are a number of weaknesses of this study. It was retrospective and conducted in a symptomatic cohort of largely
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
Caucasian women who were referred for urodynamic testing. Hence, our results may only be applicable to similar populations. While we utilised a sophisticated measure of mid-urethral mobility rather than simpler tests such as the Q-tip test, we employed urethral pressure measurements obtained with a single lumen perfused catheter. A double-lumen catheter with simultaneous registration of bladder pressures may have been superior but would also have required a stiffer catheter that would be more likely to introduce a splinting artefact. We did not obtain any electrophysiological information, with concentric needle electromyography probably the most informative technique. However, this would require a prospective study design and place far greater demands on patients. Finally, we did not collect information of muscular reflex activity which can be obtained noninvasively by imaging; however, reflex activity has not been shown to be a major factor in the pathophysiology of USI [26]. Despite all those shortcomings this study has shown however that measures obtained with current clinically available technology are insufficient to predict or explain urodynamic stress incontinence. We conclude that mid-urethral mobility and urethral resting pressure are the main predictors of USI. Demographic and clinical data are at best weak predictors. Our results suggest the presence of major unrecognised confounders not currently investigated during urodynamic testing supported by translabial ultrasound imaging. References [1] Sandvik H, Hunskaar S, Vanvik A, Bratt H, Seim A, Hermstad R. Diagnostic classification of female urinary incontinence: an epidemiological survey corrected for validity. J Clin Epidemiol 1995;48:339–43. [2] DeLancey JO. Why do women have stress urinary incontinence? Neurourol Urodyn 2010;29:13–7. [3] Krantz KE. The anatomy of the urethra and anterior vaginal wall. Am J Obstet Gynecol 1951;62:374–86. [4] DeLancey JO. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713– 20 [discussion 1720–3]. [5] Blaivas JG, Olsson CA. Stress incontinence: classification and surgical approach. J Urol 1988;139:727–31. [6] Shek KL, Dietz HP. The urethral motion profile: a novel method to evaluate urethral support and mobility. Aust NZ J Obstet Gynecol 2008;48:337. [7] Chantarasorn V, Shek KL, Dietz HP. Sonographic appearance of transobturator slings: implications for function and dysfunction. Int Urogynecol J 2011;22: 493–8.
117
[8] Dietz HP. Pelvic floor ultrasound in incontinence: what’s in it for the surgeon? Int Urogynecol J 2011;22:1085–97. [9] DeLancey JO, Trowbridge ER, Miller JM, et al. Stress urinary incontinence: relative importance of urethral support and urethral closure pressure. J Urol 2008;179:2286–90. [10] Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21: 167–78. [11] Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7. [12] Dietz HP. Pelvic floor ultrasound: a review. Am J Obstet Gynecol 2010;202: 321–34. [13] Schuessler B, Laycock J, Norton P, Stanton S, editors. Pelvic floor re-education: principles and practice. 1st ed., London: Springer; 1994. [14] Dietz HP, Shek KL, Clarke B. Biometry of the pubovisceral muscle and levator hiatus by three-dimensional pelvic floor ultrasound. Ultrasound Obstet Gynecol 2005;25:580–5. [15] Dietz HP, Bernardo M, Kirby A, Shek KL. Minimal criteria for the diagnosis of avulsion of the puborectalis muscle by tomographic ultrasound. Int Urogynecol J 2010;22:699–704. [16] Ward KL, Hilton P, UK and Ireland TVT Trial Group. A prospective multicenter randomized trial of tension-free vaginal tape and colposuspension for primary urodynamic stress incontinence: two-year follow-up. Am J Obstet Gynecol 2004;190:324–31. [17] Surkont G, Wlaz´lak E, Kowalski A, Suzin J. Burch colposuspension – analysis of complications on our own material. Ginekol Pol 2007;78:288–92. [18] Surkont G, Wlazl´ak E, Kazimierak W, Dunicz-Sokolowska A, Suzin J. The influence of SUI operative treatment with use of IVS tape on quality of life. Ginekol Pol 2007;78:381–3. [19] Lovegrove Jones RC, Peng Q, Stokes M, Humphrey VF, Payne C, Constantinou CE. Mechanisms of pelvic floor muscle function and the effect on the urethra during a cough. Eur Urol 2010;57:1101–10. [20] Dietz HP, Bond V, Shek KL. Does childbirth alter the reflex pelvic floor response to coughing? Ultrasound Obstet Gynecol 2012;39:569–73. [21] Abrams P, Cardozo L, Khoury S, Wein A, editors. Incontinence. 5th ed., Paris: ICUD-EAU; 2013. [22] Snooks SJ, Barnes PR, Swash M. Damage to the innervation of the voluntary anal and periurethral sphincter musculature in incontinence: an electrophysiological study. J Neurol Neurosurg Psychiatry 1984;47:1269–73. [23] Swash M, Snooks SJ, Henry MM. Unifying concept of pelvic floor disorders and incontinence. J R Soc Med 1985;78:906–11. [24] Gunnarsson M, Mattiasson A. Female stress, urge, and mixed urinary incontinence are associated with a chronic and progressive pelvic floor/vaginal neuromuscular disorder: an investigation of 317 healthy and incontinent women using vaginal surface electromyography. Neurourol Urodyn 1999;18:613–21. [25] Yang JM, Yang SH, Huang WC. Functional correlates of Doppler flow study of the female urethral vasculature. Ultrasound Obstet Gynecol 2006;28:96–102. [26] Dietz HP, Erdmann M, Shek KL. Reflex contraction of the levator ani in women symptomatic for pelvic floor disorders. Ultrasound Obstet Gynecol 2012;40:215–8.
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Can we predict urinary stress incontinence by using demographic, clinical, imaging and urodynamic data? Edyta Wlaz´lak a,*, Grzegorz Surkont a, Ka L. Shek b, Hans P. Dietz b a b
Clinic of Operative and Oncologic Gynecology, Medical University of Lodz, Wilenska 37, 94029 Lodz, Poland Sydney Medical School Nepean, University of Sydney, Nepean Hospital, Penrith, Sydney, NSW 2750, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 21 February 2015 Received in revised form 5 May 2015 Accepted 24 July 2015
Objective: It has been claimed that urethral hypermobility and resting urethral pressure can largely explain stress incontinence in women. In this study we tried to replicate these findings in an unselected cohort of women seen for urodynamic testing, including as many potential confounders as possible. Study design: This study is a retrospective analysis of data obtained from 341 women. They attended for urodynamic testing due to symptoms of pelvic floor dysfunction. We excluded from the analysis women with a history of previous anti-incontinence and prolapse surgery. All patients had a standardised clinical assessment, 4D transperineal pelvic floor ultrasound and multichannel urodynamic testing. Urodynamic stress incontinence (USI) was diagnosed by multichannel urodynamic testing. Its severity was subjectively graded as mild, moderate and severe. Candidate variables were: age, BMI, symptoms of prolapse, vaginal parity, significant prolapse (compartment-specific), levator avulsion, levator hiatal area, Oxford grading, midurethral mobility, maximum urethral pressure (MUP), maximum cough pressure and maximum Valsalva pressure reached. Results: On binary logistic regression, the following parameters were statistically significant in predicting urodynamic stress incontinence: age (P = 0.03), significant rectocele (P = 0.02), max. abdominal pressure reached (negatively, P < 0.0001), midurethral mobility (P = 0.0004) and MUP (negatively, P < 0.0001). On multivariate analysis, accounting for multiple interdependencies, the following predictors remained significant: max. abdominal pressure reached (negatively, P < 0.0001), cough pressure (P = 0.006), midurethral mobility (P = 0.003) and MUP (negatively, P < 0.0001), giving an R2 of 0.24. Conclusions: Mid-urethral mobility and MUP are the main predictors of USI. Demographic and clinical data are at best weak predictors. Our results suggest the presence of major unrecognised confounders. ß 2015 Published by Elsevier Ireland Ltd.
Keywords: Stress urinary incontinence 3D/4D ultrasound Urodynamic testing Pelvic floor function Levator ani
Introduction Stress urinary incontinence (SUI) is the most common type of urinary incontinence in women [1]. A first clinical description of SUI was published in 1912 [2]. Kelly described the open vesical neck as a possible cause of stress urinary incontinence (SUI). After introducing an operation that, in his opinion, cured SUI he postulated that the success of his procedure resulted from narrowing of the vesical neck. In 1922 Bonney proposed loss of urethral support as the main cause of stress incontinence. In his opinion, Kelly’s operation cured SUI through improving urethral support [2]. Jeffcoate and Roberts in 1949 showed that many stress incontinent women exhibited a loss of the urethrovesical angle
* Corresponding author. Tel.: +48 501587964; fax: +48 426860471. E-mail address: [email protected] (E. Wlaz´lak). http://dx.doi.org/10.1016/j.ejogrb.2015.07.012 0301-2115/ß 2015 Published by Elsevier Ireland Ltd.
[2]. Krantz postulated that the pubourethral ligaments, fibromuscular structures connecting the urethra to the pelvic sidewall, affect pressure transmission [3]. In 1960, Enhorning claimed that transmission of abdominal pressure was reduced in women with stress incontinence because the urethra descended below the abdominal pressure zone [2]. Another related pathophysiological concept is the ‘‘hammock hypothesis’’, which suggests that the vagina dorsal to the urethra provides a backboard against which increasing intra-abdominal forces compress the urethra [4]. The idea that primary urethral weakness could cause urinary incontinence independent of vaginal weakness appeared in a proposed classification by Blaivas et al. [5]. In their classification, they named this Type III incontinence to distinguish it from Types I and II, each of which showed movement, while Type III did not. Type III has now been largely replaced by the term intrinsic sphincter deficiency (ISD) to focus attention on urethral factors, which appeared to be independent of vaginal
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
position and mobility. There is a growing clinical impression that some degree of ISD may exist in many patients who, until recently, were thought to have only hypermobility as a cause of their incontinence [2]. Progress made in the field of imaging, especially using translabial ultrasound, now allows quantifying different aspects of urethrovesical function [6–8]. As a result it should be possible to advance understanding of the pathophysiology of urinary incontinence in women through the simultaneous analysis of anatomical and functional findings. Lately it has been claimed that urethral hypermobility and resting urethral pressures can largely explain stress incontinence in women, with maximal urethral closure pressure, not urethral support, likely to be the factor most strongly associated with stress incontinence [9]. In this study we tried to replicate these findings in an unselected cohort of women seen for urodynamic testing, including as many potential confounders as possible. We also performed the same analysis for the diagnosis of urodynamic stress incontinence (USI) for comparison. Materials and methods This study is a retrospective analysis of data obtained on urodynamic testing at a tertiary urogynaecology unit. Between January 2009 and April 2010, 454 women attended the unit for urodynamic testing due to symptoms of pelvic floor dysfunction. We excluded 113 women with a history of previous antiincontinence and prolapse surgery, leaving 341 data sets. All subsequent analysis pertains to those 341 patients. All patients had a standardised in-house, nonvalidated interview including symptoms of stress urinary incontinence as defined by the International Continence Society [10], a clinical assessment using the ICS POP-Q [11], along with a 4D transperineal pelvic floor ultrasound [12] using GE Kretz Voluson 730 expert and Voluson i systems, and multichannel urodynamic testing (Neomedix Acquidata, Neomedix, Sydney, Australia). Significant prolapse was defined as a prolapse stage 2 (ICS POP-Q system). Levator contraction strength was assessed digitally, using the Modified Oxford Grading (MOS) [13]. Urethral sphincter function was assessed with urethral profilometry. Maximal urethral pressure (MUP) was obtained with a 5F single-lumen water-perfused catheter with a freehand pull-through technique. Maximum abdominal pressure and maximum cough pressure were obtained both in the standing position on Valsalva manoeuvre and during coughing. Urodynamic stress incontinence (USI) was diagnosed in women with a positive stress test during multichannel urodynamic examination. Its severity was subjectively graded as mild, moderate, marked and severe depending on the amount of urine loss and the abdominal pressure at which this occurred. We used maximum cough pressure and maximum Valsalva pressure obtained during urodynamic testing to measure the efficacy of the response to increased demand on the continence mechanism [9,10]. All ultrasound images were analysed offline using proprietary software (4D View v 10), with the computer operators (GS and EW) blinded against all clinical data. Urethral mobility was described by vectors of movement from rest to a maximum Valsalva of 6 equidistant points marked along the length of the urethra from bladder neck (point 1) to external urethral meatus (point 6) in the midsagittal plane, as previously described [6]. For midurethral mobility analysis we used vector 4. Hiatal dimensions at rest and on Valsalva were measured in the plane of minimal hiatal dimensions, as previously described [14]. Levator trauma was identified by tomographic ultrasound (TUI) as previously described [15]. Binary outcomes of stress incontinence (SI) and USI were analysed using binary logistic regression. Both univariate and
115
multivariate models were developed for SI and USI. A backwards elimination approach was used to optimise models. Candidate variables were age, BMI, symptoms of prolapse, vaginal parity, significant prolapse (compartment-specific), avulsion, hiatal area, Oxford grading, midurethral mobility, MUP, maximum cough pressure and maximum Valsalva pressure reached. Ethical approval was obtained from the local human research ethics committee (SWAHS HREC 09-42). Results A test-retest series of urethral mobility vectors performed by two authors (EW and KLS) showed excellent interrater reliability, with an Intraclass Correlation Coefficient for single measures of 0.839 (95% CI 0.745–0.897) and 0.870 (95% CI 0.811–0.911) for segmental urethral mobility. A test-retest series for hiatal biometry also demonstrated excellent repeatability, consistent with multiple previous reports, with an ICC for single measures of 0.854 (CI 0.764–0.911) and 0.950 (CI 0.915–0.970) (for hiatal diameters), and 0.985 (CI 0.966–0.992) for area on Valsalva. All continuous parameters were tested for normality (histogram and Kolmogorov–Smirnov testing) and were found to be normally or near-normally distributed. The mean age of 341 women included in the analysis was 54 (19–89) years and BMI was 29 (17–59). Mean vaginal parity was 2.5 (0–10), and 89% were vaginally parous. A Vacuum or Forceps was reported by 25%. 23% of analysed women had had a hysterectomy. 75% of patients complained of stress urinary incontinence (SI), 68% of urge incontinence, 29% of frequency, 47% of nocturia, 28% of symptoms of voiding difficulty (hesitancy, straining and stop-start voiding), and 42% of symptoms of prolapse (lump or dragging sensation). On examination a cystocele stage 2 was found in 42%, significant central compartment prolapse in 9%, and a significant clinical rectocele (stage 2) in 43%. In 62% we detected any form of prolapse of stage 2 or higher. Mean Oxford grading was 2.4 (SD 1.2), mean bladder neck descent was 30.6 (SD 12.8) mm and mean midurethral mobility was 17.6 (SD 7.1) mm. Midsagittal hiatal diameter on Valsalva was 6.7 (1.2) cm and mean hiatal area on Valsalva 28.7 (SD 9.6) cm2. A total of 67 patients (20%) had an avulsion of the puborectalis muscle. Urodynamic findings were as follows: mean maximal urethral pressure (MUP) was 42 cmH2O (SD 20), mean maximum abdominal pressure was 87 cmH2O (SD 35) and mean maximum cough pressure was 110 cmH2O (SD 35). The urodynamic diagnosis was USI in 216 patients (65%). It was mild in 50, moderate in 68, marked in 34 and severe in 64 women. DO was diagnosed in 28%, voiding dysfunction in 31% of patients. On binary logistic regression analysis with SI as the outcome, the following parameters were (positively unless stated otherwise) predictive of SI: age (negatively, P = 0.0003), BMI (P = 0.0246), symptoms of prolapse (negatively, P = 0.0015), vaginal parity (P = 0.026), significant cystocele (negatively, P = 0.034), significant central compartment prolapse (negatively, P = 0.003), levator avulsion (negatively, P = 0.0001) and midurethral mobility (P = 0.018). MUP was not significantly related to SI. Multivariable analysis, accounting for multiple interdependencies, showed the following parameters as still significant: age (negatively, P < 0.0001), BMI (0.032), symptoms of prolapse (negatively, P = 0.0038), vaginal parity (P < 0.0001) and maximum abdominal pressure reached (negatively, P = 0.01). In Table 1 we included results of two different models used, with only statistically significant variables presented. On binary logistic regression, the following parameters were predictive of USI: age (P = 0.03), significant rectocele (P = 0.02), max. abdominal pressure reached (negatively, P < 0.0001), midurethral mobility (P = 0.0004) and MUP (negatively, P < 0.0001).
116
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
Table 1 Results of binary logistic regression modelling and multivariate modelling (accounting for multiple interdependencies) with SI as the outcome. Parameter
Binary logistic regression
Multivariable analysis
Age BMI Symptoms of prolapse Vaginal parity Cystocele stage 2 Centr. comp. stage 2 Levator avulsion Midurethral mobility Max. p abd (Valsalva)
0.0003 (neg) 0.0246 (pos) 0.0015 (neg) 0.026 (pos) 0.034 (neg) 0.003 (neg) 0.0001 (neg) 0.018 (pos) n.s.
<0.0001 (neg) 0.032 (pos) 0.0038 (neg) <0.0001 (pos) n.s. n.s. n.s. n.s. 0.01 (neg)
pos, positive association; neg, negative association; Centr. comp., central compartment; Max. p abd (Valsalva), maximum abdominal pressure reached on Valsalva during urodynamic exam.
Multivariable analysis, accounting for multiple interdependencies, showed the following parameters as remaining significant: MUP (negatively, P < 0.0001), rectocele (P = 0.01), max. abdominal pressure (negatively, P < 0.0001), max cough pressure (P = 0.003), avulsion (negatively, P = 0.03), and midurethral mobility (P = 0.0005). On univariate linear modelling with USI grade (0– 4) as outcome, the following parameters were significantly associated with USI: age (P < 0.0001), vaginal parity (P = 0.04), maximum abdominal pressure (negatively, P < 0.001), maximum cough pressure (negatively, P = 0.007), midsagittal hiatal diameter on Valsalva (P = 0.03), midurethral mobility (P = 0.0013) and MUP (negatively, P < 0.001). On multivariate analysis, accounting for multiple interdependencies (e.g. between significant cystocele, BND and midurethral mobility), the following predictors remained significant: maximal abdominal pressure (negatively, P < 0.0001), cough pressure (P = 0.006), midurethral mobility (P = 0.003) and MUP (negatively, P < 0.0001), giving an optimal R2 of 0.24. In Table 2 we included results of two different models used, with only statistically significant variables presented. Comment Despite progress made in stress urinary incontinence (SUI) operative treatment over the last two generations, cure rates have not risen substantially over 80% [9]. The mechanism of action of anti-incontinence procedures as well as reasons for failure are still poorly understood [7,16–18], as is the pathophysiology and aetiology of stress urinary incontinence [2]. Over the last century, several hypotheses have been proposed to explain stress urinary incontinence. These theories are based on clinical observations and focus primarily on the causative role of urethral support loss and urethral quality [2]. These hypotheses have been tested by comparing measurements of urethral support and function in women with primary stress urinary incontinence to the same measurements obtained in asymptomatic volunteers Table 2 Results of binary logistic regression modelling and multivariate modelling (accounting for multiple interdependencies) with USI as the outcome. Parameter
Binary logistic regression
Multivariable analysis
Age Rectocele stage 2 Max. p abd (Valsalva) Midurethral mobility Maximal urethral pressure Levator avulsion
0.03 (neg) 0.02 (pos) <0.0001 (neg) 0.0004 (pos) <0.0001 (neg) n.s.
n.s. 0.01 (pos) <0.0001 (neg) 0.0005 (pos) <0.0001 (neg) 0.003 (neg)
pos, positive association; neg, negative association; Max. p abd (Valsalva), maximum abdominal pressure reached on Valsalva during urodynamic exam.
who were recruited to be similar in age, race, and parity [9]. In this study maximal urethral closure pressure was the parameter that differed the most between groups being 43% lower in women with stress incontinence than asymptomatic women. Although operations that provide differential support to the urethra are effective, this study showed that urethral support may not be the predominant cause of stress incontinence [9]. In this retrospective study we tried to take into consideration as many factors as possible, which may have a confounding influence on clinical stress incontinence (SI) and urodynamic stress incontinence (USI). We analysed three main factors involved in the mechanism of stress continence: (1) urethral function as measured by maximal urethral pressure obtained with a 5F single-lumen fluid-perfused catheter; (2) urethral support, as determined by urethral mobility vectors (6) and (3) the efficiency of the response to increased demands on the continence mechanism, by assessing the observation of urethral loss relative to maximal cough and abdominal pressure. We have again confirmed that urethral resting pressure and mid-urethral mobility are predictors of USI, although we have been unable to replicate the findings reported by DeLancey [9]. While his model reached a Nagelkerke R2 of over 60%, ours at best reached about 24%. This very likely is due to differences between methods and populations. Our results suggest the presence of major unrecognised confounders in this real-world population of patients referred for urodynamic testing. One of them may be urethral kinking, that is, deformation of the urethra due to differential mobility of urethral segments, a topic that clearly requires further research. There may be other factors influencing the efficiency of continence mechanisms for example a reflex activation of urethral rhabdosphincter, levator ani and other fibromuscular structures on coughing [19]. Reflex contraction of the levator ani and external perineal muscles can be observed on translabial ultrasound during sudden increases in intra-abdominal pressure. These reflex contractions are almost universally present in nulliparous pregnant women. There seems to be a reduction in reflex contraction magnitude after childbirth, and this reduction may be associated with vaginal delivery. Their magnitude may be associated with postpartum stress urinary incontinence. The clinical significance of this finding is uncertain but this finding is consistent with the growing evidence of pelvic floor damage, both macroscopic and functional, or possibly ultrastructural, attributable to vaginal delivery [20]. The effect of pregnancy remains poorly defined, although it is likely that hormonal effects of pregnancy have an effect on urethral support and hence pressure transmission [21]. During a cough, normal PFM function is likely to produce timely compression of the urethra and vagina and additional external support to the urethra, reducing displacement, velocity, and acceleration. In women with SUI, who are likely to have weaker urethral attachments, this shortening contraction may not occur; consequently, the urethras of women with SUI may move further and faster for a longer duration [19]. Snooks and Swash [22,23] first brought attention to the importance of urethral denervation after childbirth and its possible contribution to urinary and faecal incontinence. Stress incontinence is frequently associated with a decline in the electrophysiological function of the pudendal nerve [21], the striated urethral sphincter [21], and the pelvic floor muscles [21,24]. Most recent studies continue to support the finding of prolonged pudendal nerve terminal motor latency in SI [21]. Urethral vasculature has also been postulated to play a role in the continence mechanism and different Doppler parameters have been studied to evaluate correlation with SI, although results are conflicting [25]. There are a number of weaknesses of this study. It was retrospective and conducted in a symptomatic cohort of largely
E. Wlaz´lak et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 193 (2015) 114–117
Caucasian women who were referred for urodynamic testing. Hence, our results may only be applicable to similar populations. While we utilised a sophisticated measure of mid-urethral mobility rather than simpler tests such as the Q-tip test, we employed urethral pressure measurements obtained with a single lumen perfused catheter. A double-lumen catheter with simultaneous registration of bladder pressures may have been superior but would also have required a stiffer catheter that would be more likely to introduce a splinting artefact. We did not obtain any electrophysiological information, with concentric needle electromyography probably the most informative technique. However, this would require a prospective study design and place far greater demands on patients. Finally, we did not collect information of muscular reflex activity which can be obtained noninvasively by imaging; however, reflex activity has not been shown to be a major factor in the pathophysiology of USI [26]. Despite all those shortcomings this study has shown however that measures obtained with current clinically available technology are insufficient to predict or explain urodynamic stress incontinence. We conclude that mid-urethral mobility and urethral resting pressure are the main predictors of USI. Demographic and clinical data are at best weak predictors. Our results suggest the presence of major unrecognised confounders not currently investigated during urodynamic testing supported by translabial ultrasound imaging. References [1] Sandvik H, Hunskaar S, Vanvik A, Bratt H, Seim A, Hermstad R. Diagnostic classification of female urinary incontinence: an epidemiological survey corrected for validity. J Clin Epidemiol 1995;48:339–43. [2] DeLancey JO. Why do women have stress urinary incontinence? Neurourol Urodyn 2010;29:13–7. [3] Krantz KE. The anatomy of the urethra and anterior vaginal wall. Am J Obstet Gynecol 1951;62:374–86. [4] DeLancey JO. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713– 20 [discussion 1720–3]. [5] Blaivas JG, Olsson CA. Stress incontinence: classification and surgical approach. J Urol 1988;139:727–31. [6] Shek KL, Dietz HP. The urethral motion profile: a novel method to evaluate urethral support and mobility. Aust NZ J Obstet Gynecol 2008;48:337. [7] Chantarasorn V, Shek KL, Dietz HP. Sonographic appearance of transobturator slings: implications for function and dysfunction. Int Urogynecol J 2011;22: 493–8.
117
[8] Dietz HP. Pelvic floor ultrasound in incontinence: what’s in it for the surgeon? Int Urogynecol J 2011;22:1085–97. [9] DeLancey JO, Trowbridge ER, Miller JM, et al. Stress urinary incontinence: relative importance of urethral support and urethral closure pressure. J Urol 2008;179:2286–90. [10] Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21: 167–78. [11] Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7. [12] Dietz HP. Pelvic floor ultrasound: a review. Am J Obstet Gynecol 2010;202: 321–34. [13] Schuessler B, Laycock J, Norton P, Stanton S, editors. Pelvic floor re-education: principles and practice. 1st ed., London: Springer; 1994. [14] Dietz HP, Shek KL, Clarke B. Biometry of the pubovisceral muscle and levator hiatus by three-dimensional pelvic floor ultrasound. Ultrasound Obstet Gynecol 2005;25:580–5. [15] Dietz HP, Bernardo M, Kirby A, Shek KL. Minimal criteria for the diagnosis of avulsion of the puborectalis muscle by tomographic ultrasound. Int Urogynecol J 2010;22:699–704. [16] Ward KL, Hilton P, UK and Ireland TVT Trial Group. A prospective multicenter randomized trial of tension-free vaginal tape and colposuspension for primary urodynamic stress incontinence: two-year follow-up. Am J Obstet Gynecol 2004;190:324–31. [17] Surkont G, Wlaz´lak E, Kowalski A, Suzin J. Burch colposuspension – analysis of complications on our own material. Ginekol Pol 2007;78:288–92. [18] Surkont G, Wlazl´ak E, Kazimierak W, Dunicz-Sokolowska A, Suzin J. The influence of SUI operative treatment with use of IVS tape on quality of life. Ginekol Pol 2007;78:381–3. [19] Lovegrove Jones RC, Peng Q, Stokes M, Humphrey VF, Payne C, Constantinou CE. Mechanisms of pelvic floor muscle function and the effect on the urethra during a cough. Eur Urol 2010;57:1101–10. [20] Dietz HP, Bond V, Shek KL. Does childbirth alter the reflex pelvic floor response to coughing? Ultrasound Obstet Gynecol 2012;39:569–73. [21] Abrams P, Cardozo L, Khoury S, Wein A, editors. Incontinence. 5th ed., Paris: ICUD-EAU; 2013. [22] Snooks SJ, Barnes PR, Swash M. Damage to the innervation of the voluntary anal and periurethral sphincter musculature in incontinence: an electrophysiological study. J Neurol Neurosurg Psychiatry 1984;47:1269–73. [23] Swash M, Snooks SJ, Henry MM. Unifying concept of pelvic floor disorders and incontinence. J R Soc Med 1985;78:906–11. [24] Gunnarsson M, Mattiasson A. Female stress, urge, and mixed urinary incontinence are associated with a chronic and progressive pelvic floor/vaginal neuromuscular disorder: an investigation of 317 healthy and incontinent women using vaginal surface electromyography. Neurourol Urodyn 1999;18:613–21. [25] Yang JM, Yang SH, Huang WC. Functional correlates of Doppler flow study of the female urethral vasculature. Ultrasound Obstet Gynecol 2006;28:96–102. [26] Dietz HP, Erdmann M, Shek KL. Reflex contraction of the levator ani in women symptomatic for pelvic floor disorders. Ultrasound Obstet Gynecol 2012;40:215–8.