Must Colposuspension be Associated with Sacropexy to Prevent Postoperative Urinary Incontinence?

Transcription

1 european urology 51 (2007) available at journal homepage: Female Urology Incontinence Must Colposuspension be Associated with Sacropexy to Prevent Postoperative Urinary Incontinence? Elisabetta Costantini a, *, Alessandro Zucchi a, Antonella Giannantoni a, Luigi Mearini a, Vittorio Bini b, Massimo Porena a a Department of Medical-Surgical Specialties and Public Health, Section of Urology and Andrology, Unit of Urogynaecology, University of Perugia, Perugia, Italy b Department of Internal Medicine, University of Perugia, Italy Article info Article history: Accepted August 22, 2006 Published online ahead of print on September 5, 2006 Keywords: Colposuspension Prolapse Sacropexy Urinary stress incontinence Abstract Objectives: This prospective, randomised study investigated whether a prophylactic procedure, performed during colposacropexy for prolapse repair, prevents ex novo postoperative incontinence. Sixty-six consecutive continent patients with advanced prolapse were randomised into two groups: group A underwent sacropexy combined with a Burch colposuspension; no anti-incontinence procedure was performed in group B patients. Methods: Work-up included clinical assessment (Halfway System and International Continence Society [ICS] classification for prolapse and Ingelman Sunderberg scale for incontinence), the Urogenital Distress Inventory and Impact Incontinence Quality of Life questionnaires, urogynaecologic ultrasound scans, and complete urodynamic testing that included the urethral pressure profile and Valsalva leak point pressure with reduced prolapse. Check-ups were done at 3, 6, 12 mo postoperatively and then yearly. Mean follow-up time was 39.5 mo. Results: The mean age ( standard deviation) was 62 9 yr. All patients presented with grade (G) 3 4 prolapse. Postoperative incontinence was present in 12 of the 34 patients in group A: 7 G1; 4 G2, and 1 G3. Postoperative incontinence was present in 3 of the 32 patients in group B: 2 G1, 1 G3. The frequency of postoperative incontinence was significantly greater in patients who had undergone colposuspension ( p < 0.05). Conclusions: These preliminary data cast doubt on whether colposuspension should be performed during sacropexy for severe urogenital prolapse as prophylaxis for postoperative incontinence because it seems to emerge as overtreatment. Incontinence developed ex novo in 35% of continent patients treated with colposuspension combined with sacropexy. # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Urology Department, University of Perugia, Via Brunamonti 51, Perugia, Italy. Tel ; Fax: address: ecostant@unipg.it (E. Costantini) /$ see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi: /j.eururo

2 european urology 51 (2007) Introduction The onset of postoperative stress urinary incontinence (SUI) after repair of severe pelvic organ prolapse (POP) in previously continent women represents a challenge for urogynaecologists, and several prophylactic procedures have been developed to prevent it [1]. SUI in previously continent patients after prolapse surgery is frustrating for both the patient and the physician. Severe POP is not usually associated with subjective SUI, despite anatomic displacement of the urethrovesical junction and proximal urethra outside the urogenital diaphragm. The rotation of the urethra with straining may cause kinking or compression, effectively closing the urethra off. It is also likely that the accompanying large cystocele helps to dissipate the pressure placed on the bladder neck and preserve continence [2]. The relationship between occult incontinence and postoperative SUI has not been studied until recently. Despite lack of evidence, surgeons have generally performed anti-incontinence procedures at the time of prolapse repair. None of the previously published studies on the use of prophylactic procedures for occult incontinence controlled their studies with a nonintervention arm; therefore, the risk of occult incontinence in patients having postoperative SUI, if they did not have an additional incontinence procedure, remained unclear [1]. The Burch colposuspension is generally accepted as a reference procedure in the treatment of SUI. Because approximately 50% of women with SUI have associated vault prolapse, a Burch procedure can therefore be combined with a sacrocolpopexy to correct both conditions at once [3]. On the basis of our experience with sacropexy in patients with severe POP, we decided to perform a randomised, prospective study with a control arm to evaluate the impact of Burch colposuspension as anti-incontinence prophylaxis in continent patients undergoing abdominal surgery for severe prolapse. 2. Patients and methods 2.1. Procedure From 2000 to 2004, all clinically continent women with severe POP who had undergone colposacropexy were included in this study. The patients were randomised into two groups. In group A patients, a Burch colposuspension was performed in combination with sacropexy and in group B women, no prophylactic colposuspension was performed. Evaluation included history, Urogenital Distress Inventory (UDI) and Impact Incontinence Quality of Life (IIQ) questionnaires, voiding diary, urine culture, physical examination, pelvic ultrasound, and urodynamic assessment. Pelvic floor defects were classified using the Halfway System and the International Continence Society (ICS) system [4]. Measurements were made under maximum straining with the patient in the semirecumbent position. Urinary incontinence was clinically classified on the basis of the ICS definition and graded on the Ingelman Sunderberg scale. All patients underwent a stress test in the supine position at physiologic bladder capacity, before and after prolapse reposition both with the fingers and using a posterior blade of a Sims speculum placed in the anterior vaginal fornix [5]. The test was considered positive if leakage occurred with a cough or Valsalva manoeuvre. All patients underwent a urodynamic test that included uroflowmetry, cystometry, pressure/flow study, urethral profilometry, Valsalva leak point pressure (VLPP) performed at 200 cc with the patient in a semirecumbent position before and after prolapse reposition (terms and definition are according to ICS guidelines [4]). Patients with a negative stress test before and after prolapse reduction, no preoperative history of symptoms of urinary incontinence, a negative symptoms questionnaire, and no leakage during urodynamics tests were considered eligible for the study. All patients underwent colposacropexy with (n = 17) or without (n = 11) hysterectomy [6,7]. A standard Burch procedure was performed using four sutures, two on each side. The patients were assigned to surgical intervention according to a randomised block design [2]. All patients were informed about the trial aim and procedures and gave their informed consent. Follow-up visits, scheduled at 3, 6, and 9 mo and then annually, included a detailed urogynaecologic history, clinical examination, and stress test. The primary outcome measure was occurrence of ex novo urinary incontinence after surgery in each group. The primary outcome time point for measuring incontinence was at 6 mo. Patients were re-evaluated at 1 yr and then every year. Assessments were based on the stress test and subjective symptoms as measured by the UDI questionnaire Statistical analysis The sample size of 66 patients provides a statistical power (1 b) of about 80 85% at a = 0.05 for the detection of 25 30% difference in proportion of postoperative incontinence (or any other condition) between the two groups. The Mann-Whitney test was used to compare ordinal and nonnormally distributed continuous variables (deviation from Gaussian distribution were checked by using the Kolgomorov- Smirnov test with the Lilliefors method) [8]. For simplicity, data were reported as mean standard deviation. Categorical data were analysed by the McNemar test, X 2 test, or Fisher exact test, as appropriate. For adjusting predictive variables of postoperative incontinence, a multiple logistic regression model was applied with subjects subdivided according to their postoperative status at follow-up (0 = continent; 1 = incontinent). For fitting logistic regression model, the following independent potential predictor variables were used: type of surgical intervention (0 = no colposuspension; 1 = colposuspension), maximum urethral closure pressure (MUCP; 35 = 0; >35 = 1), and body mass index (BMI). Goodness of fit of logistic

3 790 european urology 51 (2007) Table 1 Preoperative data Sacropexy (SP) 6 15 Hysterectomy + SP Hysterocolposacropexy 11 5 Age, yr Menopause, n Previous surgery 8 12 Hysterectomy 6 10 Prolapse repair 2 6 SUI surgery 1 6 BMI, kg/m Parity, median 2 2 IIQ score UDI score SUI = stress urinary incontinence; IIQ = Incontinence Impact Questionnaire; UDI = Urogenital Distress Inventory. model was tested by using the Hosmer-Lemeshow test. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated both for contingency tables and logistic regression. The level of statistic significance was set at p < All calculations were carried out with SPSS release 13.0, 2004 (Statistical Packages for Social Sciences, SPSS, Chicago, IL). 3. Results The mean age of the 66 patients was 62 9 yr (range: yr), BMI was 24 2 kg/m 2, and median parity was 2 (range: 1 5). Sixteen patients had already undergone hysterectomy, 8 had prolapse repair, and 7 had SUI surgery. Fifty-six patients were postmenopausal. The two groups did not differ significantly in any of the above variables, in the severity of genital prolapse (Table 1), and in their urodynamic data (Table 2). Thirty-four women were randomised to group A (Burch procedure) and 32 to group B. Intraoperative and early postoperative complications are shown in Table 3. Three women in group A had postoperative urinary retention that resolved spontaneously after 4 5 d in one patient and after self-catheterisation for 10 d and 3 mo, respectively, in the other two. The median follow-up was mo (range: mo) in group A and (range: mo) in group B. The mean IIQ and UDI scores are shown in Table 1. Incontinence results at the primary outcome time point of 6 mo after surgery were stable. There was no increase or decrease in continence in the groups over time. Table 2 Preoperative urodynamic data Free uroflowmetry, n Volume, ml Q max, ml/s PVR, n 6 11 Detrusor overactivity, n 7 5 MCC, ml Reduced compliance, n 4 4 Pop det, cm H 2 O Pdet max,cmh 2 O Pdet qmax,cmh 2 O P/F volume Q max, ml/s PVR, n MUPC, cm H 2 O Q max = maximum flow rate; PVR = post-void residue; MCC = maximum cystometric capacity; Pop det = opening detrusor pressure; Pdet max = maximum detrusor pressure; Pdet qmax = detrusor pressure at maximum flow; P/F = pressure/flow; MUPC = maximal urethral closure pressure.

4 european urology 51 (2007) Table 3 Complications Blood transfusions 3 3 Fever 1 1 Augmented hepatic enzymes 1 Severe constipation 2 1 Wound hematoma 1 Voiding disturbances 3 Table 4 Development of ex novo stress incontinence Preoperative Postoperative Preoperative Postoperative Voiding symptoms (1 improved) Storage symptoms Incontinence G1 4G2 1G3 2G1 1G3 1U 2US 1S 1S 1US 6S 2S 1US U = urge incontinence; S = stress incontinence; US = mixed incontinence. Postoperative results showed ex novo incontinence in 12 of 34 patients in group A and in 3 of 32 patients in group B with a significant difference in frequency (OR = 5.3; 95%CI, ; p < 0.05). In particular, ex novo urge or mixed incontinence occurred in 3 of 34 patients in group A and in 2 of 32 in group B; ex novo stress incontinence developed in 9 of 34 in group A and 1 of 32 in group B(Table 4). At follow-up, three patients complained of voiding difficulties; all other patients are cured or improved. Storage symptoms improved in both groups (Fig. 1). When the patients who had undergone previous incontinence surgery were removed from data analysis, no difference between the groups emerged. Removing only the patient who had undergone previous incontinence surgery in group A, the result remained 12 of 33 (36.4%) postoperative incontinent patients because this patient was continent postoperatively. When the six patients of group B were removed from the data analysis, the postoperative incontinence rate was 3.8% (1 of 26) because two of the six patients were affected by ex novo incontinence. The difference between the two groups remains significant ( p < ). At preoperative urodynamics 19 patients showed an MUCP 35. In this group of patients 12 (63.15%) were continent after surgery and 7 were incontinent (3 G1, 2 G2, 2 G3). In the 44 patients with MUCP >35, 34 (77%) were continent postoperatively and 7 were incontinent (5 G1 and 2 G2). Results are better in group B patients (Table 5). The multiple logistic regression analysis showed that, among potential predictors, only the type of surgical intervention was significantly and independently related to postoperative incontinence with worse results in patients who underwent colposuspension (OR = 7.6; 95%CI, ; p < 0.05). Fig. 1 Postoperative storage symptoms. VS = voiding symptoms; SS = storage symptoms.

5 792 european urology 51 (2007) Table 5 Urodynamic results based on MUCP Colposuspension No colposuspension Preoperative Postoperative Preoperative Postoperative MUCP 35 cm H 2 O 9 4 continent 10 8 continent 5 incontinent 2 incontinent 2 G1; 2 G2; 1 G3 1 G1, 1 G3 MUCP >35 cm H 2 O continent continent 6 incontinent 1 incontinent 4 G1, 2 G2 1 G1 Data are missing for two patients in group a and one in group B. MUCP = maximal urethral closure pressure. No significant relationships with the dependent variables were found for MUCP and BMI. 4. Discussion The incidence of postoperative SUI after prolapse surgery is not well established and it is reported as ranging from 8% to 60% [2]. Several studies have indicated various incontinence rates after different prophylactic anti-incontinence procedures, but data on long-term follow-up in these series are sparse [9,10]. In attempts to predict postoperative SUI in currently continent women, some surgeons recommend reduction testing with replacement of the prolapse into the intended postoperative position and many different methods have been described, including the use of pessary, cotton swab, Sims speculum, vaginal packing, ring forceps, and manual reduction. Unfortunately, sensitivity, specificity, and predictive values for all reduction tests are not known [11,12]. Reduction manoeuvres may artificially obstruct the urethra or overly straighten the urethrovesical junction, thus creating a falsepositive or false-negative impression of the patient s sphincteric mechanism [13]. Anecdotal clinical experience suggests that when prolapse is replaced more deeply in the pelvis, SUI is more likely to occur because the anterior vagina and urethrovesical angle is flattened. Different surgical procedures place the vaginal apex in different pelvic locations, and it is difficult to reproduce similar effects by prolapse manipulation. Moreover, it is not known whether reduction testing is equally effective in predicting postoperative incontinence after different types of prolapse surgery [11]. Thus, in the continent patient, the reconstructive surgeon is left with one of three options: a routine anti-incontinence operation, with the risk that some patients will be overtreated; a postoperative wait and watch approach, reserving a second operation for women whose SUI symptoms warrant it; and finally, use of unreliable reduction testing in an attempt to predict which patients will have troublesome postoperative SUI [11]. To our knowledge only three studies address the incidence of postoperative incontinence without a concomitant anti-incontinence procedure. Stanton et al. [14] reported 11% postoperative SUI within 3 mo of anterior colporrhaphy with or without vaginal hysterectomy; in two observational studies of continent women with various grades of prolapse who underwent POP surgery (without urethropexy), Borsta and Rud reported 22% and 28% rates of postoperative SUI [15,16]. In recent years some studies suggested that patients with occult incontinence undergoing pubovaginal sling [17,18], urethropexy, or needle procedure [2,19] have a 0 50% chance of postoperative SUI (depending on the definition of incontinence). In a randomised study on patients with occult incontinence and urethral hypermobility treated with either a Muzsnay needle suspension or bladder-neck endopelvic fascial plication, Bump and Hurt [20] reported 14% and 7% postoperative SUI at 6 mo, respectively, concluding that occult incontinence does not predict the need for a urethropexy. When the results of posterior pubourethral ligament plication and Pereyra needle suspension were compared in 73 women with genital prolapse and potential SUI, Colombo et al. [21] showed that 50% and 76% of them were objectively continent after surgery. The Burch operation along with other reconstruction surgery has generally been recommended [22,23] for treating coexisting SUI and POP. The Colpopexy and Urinary Reduction Efforts (CARE) randomised controlled trial in patients with either a positive or negative preoperative pessary test who were followed up after abdominal sacrocolpopexy [24] reported opposite results. The discrepancies with outcomes in our patients are

6 european urology 51 (2007) probably due to different factors. (1) The sacropexy technique. Our surgical technique is an integral pelvic organ prolapse reconstruction. We perform a wide preparation of the anterior and posterior vaginal wall [6,7]. Positioning the anterior mesh at such depth, at the urethral level, could in itself correct any potential incontinence. This is only one possible explanation because different surgical techniques could lead to different results. (2) Patient enrolment. We included only patients without subjective or objective incontinence. Including patients with occult incontinence could lead to different results. (3) Length of follow-up. The Brubacker study had a 3-mo follow-up, whereas our study had a mean follow-up of 38.5 mo. In the last paragraph of the results Brubacker et al. reported outcomes for stress and urge incontinence after 1 yr. At 3 mo, 23.8% in the Burch group versus 44.1% met one or more criteria for SUI ( p < 0.001) and 32.7% versus 38.4% ( p = 0.048, not significant) for urge incontinence. At 1 yr, 20.9% in the Burch group as compared with 39.7% in the control group met one or more criteria for stress incontinence ( p = 0.02), whereas 27.6% of those in the Burch group and 35% of those in the control group met one or more criteria for urge outcome ( p = 0.37). These data might lead one to think that a longer follow-up might be associated with slightly different results. Cosson et al. reported that only 34% of the patients with prolapse and preoperative SUI achieved a complete correction of the dysfunction with a Burch procedure during sacrocolpopexy; 85% of cases occurred during the first 3 yr [25]. When the Burch procedure is combined with sacrocolpopexy, results do not appear as good as those of an isolated Burch procedure, which shows a long-term cure rate of 69% after yr [26]. Cosson et al. maintain that failure might be due to excessive traction on the anterior mesh and suggest performing colposuspension before colpopexy [25]. Our results confirm Cosson s observations. Frequency of postoperative incontinence was significantly greater in patients who underwent colposuspension, indicating that colposuspension is not useful in preventing SUI in patients undergoing colposacropexy. Abdominal colpopexy could be more at risk for postoperative SUI because of the change in the vaginal axis. Even though we have been performing sacropexy for years without excessive traction on the vaginal walls, in some predisposed patients (i.e., those with occult intrinsic sphincter deficiency) the direction of traction alone might be enough to cause SUI. Applying an adjunctive factor, such as the Burch procedure, with an anterior traction on the bladder neck, may increase the risk of SUI. Other factors such as surgical damage to urethral sphincter innervation and to the periurethral vascular plexus might also come into play. In our study, multivariate analysis showed that colposuspension was directly related to postoperative incontinence, increasing its risk, whereas MUCP and BMI were not predictive factors for outcome. One final point is that one can debate the justification of any additional anti-incontinence surgery because of the risk of increased perioperative morbidity, especially voiding dysfunction and urinary retention. A significant rate of such problems could outweigh the potential benefits of preventing incontinence. This must be considered even though, in our experience, the Burch procedure does not create significant obstructive problems because of the lack of excessive traction. 5. Conclusions Although some investigators advise concomitant SUI operations in all patients with severe POP regardless of whether demonstrable incontinence is present or not, our data support the prudent casespecific policy of concomitant incontinence surgery only when clearly indicated. In centres where this policy is enforced, postoperative absence of SUI ranges from 96% to 100% in short- to medium-range follow-up periods [14,19,27]. Our preliminary data cast doubts on whether colposuspension should be performed during sacropexy as prophylaxis for postoperative incontinence because it seems to emerge as overtreatment. Incontinence developed ex novo in 35% of continent patients. Larger and more rigorous studies are needed to determine whether the risk of complications justifies the benefit of a prophylactic procedure, especially in continent patients. Acknowledgement The authors would like to thank Dr Geraldine Anne Boyd for her help in editing the paper in English. References [1] Haessler A, Lin LL, Ho MH, Betson LH, Bateia NN. Reevaluating occult incontinence. Curr Opin Obstet Gynecol 2005;17:

7 794 european urology 51 (2007) [2] Bergman A, Koonings PP, Ballard CA. Predicting postoperative urinary incontinence development in women undergoing operation for genitourinary prolapse. Am J Obstet Gynecol 1988;158: [3] Baden WF, Walker T. Surgical repair of vaginal defects. Philadelphia: Lippincott; [4] 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: [5] Bump RC, Fantl JA, Hurt WG. The mechanism of urinary continence in women with severe uterovaginal prolapse: results of barrier studies. Obstet Gynecol 1988;72: [6] Costantini E, Lombi R, Micheli C, Parziani S, Porena M. Colposacropexy with Goretex mesh in marked vaginal and utero-vaginal prolapse. Eur Urol 1998;34: [7] Costantini E, Mearini L, Bini V, Zucchi A, Mearini E, Porena M. Uterus preservation in surgical correction of urogenital prolapse. Eur Urol 2005;48: [8] Lilliefors HW. On the Kolmogorov-Smirnov tests for normality with mean and variance unknown. J Am Stat Assoc 1967;62: [9] Cross CA, Cespedes DR, McGuire EJ. Treatment results using pubovaginal slings in patients with large cystoceles and stress incontinence. J Urol 1997;158: [10] Gordon D, Groutz A, Wolman I, Lessing JB, David MP. Development of postoperative urinary stress incontinence in clinically continent patients undergoing prophylactic Kelly plication during genitourinary prolapse repair. Neurourol Urodyn 1999;18: [11] Nygaard IE, McCreery R, Brubacker L, et al., for the Pelvic Floor Disorders Network. Abdominal sacrocolpopexy: a comprehensive review. Obstet Gynecol 2004;104: [12] Bhatia NN, Bergman A, Gunning JE. Urodynamic effects of a vaginal pessary in women with stress urinary incontinence. Am J Obstet Gynecol 1983;147: [13] Karram MM. What is the optimal anti-incontinence procedure in women with advanced prolapse and potential stress incontinence? Int Urogynecol J Pelvic Floor Dysfunct 1999;10:1 2. [14] Stanton SL, Hilton P, Norton C, Cardozo L. Clinical and urodynamic effects of anterior colporrhaphy and vaginal hysterectomy for prolapse with and without incontinence. Br J Obstet Gynecol 1982;89: [15] Borstad E, Rud T. The risk of developing urinary stressincontinence after vaginal repair in continent women. A clinical and urodynamic follow-up study. Acta Obstet Gynecol Scand 1989;68: [16] Borstad E, Skrede M, Rud T. Failure to predict and attempts to explain urinary stress incontinence following vaginal repair in continent women by using a modified lateral urethrocystography. Acta Obstet Gynecol Scand 1991;70: [17] Barnes NM, Dmochowski RR, Park R, Nitti WW. Pubovaginal sling and pelvic prolapse repair in women with occult stress urinary incontinence: effects on postoperative emptying and voiding symptoms. Urology 2002; 59: [18] Chaikin DC, Groutz A, Blaivas JG. Predicting the need for anti-incontinence surgery in continent women undergoing repair of severe urogenital prolapse. J Urol 2000; 163: [19] Groutz A, Gordon D, Wolman I, et al. The use of prophylactic Stamey bladder neck suspension to prevent post-operative stress urinary incontinence in clinically continent women undergoing genitourinary prolapse repair. Neurourol Urodyn 2000;19: [20] Bump RC, Hurt WG, Theofrastous JP. Randomized prospective comparison of needle colposuspension versus endopelvic fascia plication for potential stress incontinence prophylaxis in women undergoing vaginal reconstruction for stage III or IV pelvic organ prolapse. The Continence Program for Women Research Group. Am J Obstet Gynecol 1996;175: [21] Colombo M, Maggioni A, Scalambrino S. Surgery for genitourinary prolapse and stress incontinence: a randomized trial of posterior pubourethral ligament plication and Pereyra suspension. Am J Obstet Gynecol 1997;176: [22] Colombo M, Vitobello D, Proietti F, Dilani R. Randomized comparison of Burch colposuspension versus anterior colporrhaphy in women with stress incontinence and anterior vaginal wall prolapse. Br J Obstet Gynecol 1992;167: [23] Bergman A, Koonings PP, Ballard CA. Primary stress urinary incontinence and pelvic relaxation: prospective randomised comparison of three different operations. Am J Obstet Gynecol 1989;161: [24] Brubaker L, Cundiff GW, Fine P, et al., for the Pelvic Floor Disorders Network. Abdominal sacrocolpopexy with Burch colposuspension to reduce urinary stress incontinence. N Engl J Med 2006;354: [25] Cosson M, Boukerrou M, Narducci F, Occelli B, Querleu D, Crèpin G. Long-term results of the Burch procedure combined with abdominal sacrocolpopexy for treatment of vault prolapse. Int Urogynecol J 2003;14: [26] Alcalay M, Monga A, Stanton S. Burch colposuspension: a year follow-up. Br J Obstet Gynaecol 1995;102: [27] Klutke JJ, Ramos S. Urodynamic outcome after surgery for severe prolapse and potential stress incontinence. Am J Obstet Gynecol 2000;182: