Heinz Koelbl led the peer-review process as the Associate Editor responsible for the paper.
Randomized Controlled Trial Comparing Two Procedures for Anterior Vaginal Wall Prolapse
Article first published online: 18 MAR 2013
© 2013 Wiley Periodicals, Inc.
Neurourology and Urodynamics
Volume 33, Issue 1, pages 72–77, January 2014
How to Cite
Minassian, V. A., Parekh, M., Poplawsky, D., Gorman, J. and Litzy, L. (2014), Randomized Controlled Trial Comparing Two Procedures for Anterior Vaginal Wall Prolapse. Neurourol. Urodyn., 33: 72–77. doi: 10.1002/nau.22396
Conflict of interest: none.
Clinical Trials ID: NCT00271102.
- Issue published online: 11 DEC 2013
- Article first published online: 18 MAR 2013
- Manuscript Accepted: 14 FEB 2013
- Manuscript Received: 12 NOV 2012
- anterior colporrhaphy;
- paravaginal defect repair
To compare success rates between anterior colporrhaphy and abdominal paravaginal defect repair for treatment of anterior vaginal wall prolapse.
This was a prospective randomized controlled trial comparing anterior colporrhaphy plus polyglactin 910 mesh (vaginal) to paravaginal defect repair (abdominal) in women with symptomatic anterior vaginal wall prolapse. Pelvic organ prolapse quantification staging (POP-Q), pelvic floor distress inventory, pelvic floor impact questionnaire, and pelvic organ prolapse/urinary incontinence sexual questionnaires were administered pre and post-operatively. Women were followed up to 2 years. The primary outcome was anterior POP-Q stage, with failure defined as ≥stage II.
We enrolled 70 patients, 35 in each group. Demographic and most peri-operative characteristics were similar between the groups. Mean anterior vaginal wall prolapse repair time (39 min) was shorter for vaginal versus abdominal repair (60 min; P < 0.001), with more concurrent hysterectomies in the vaginal (71%) versus abdominal group (42%), P = 0.01. At 2 years, objective failure rates for the vaginal and abdominal groups were 32% and 40%, respectively, P = 0.56. Subjective failure rates were lower and similar for both groups. Patient satisfaction rates were 88% for the vaginal and 73% for the abdominal group, P = 0.11. Quality of life questionnaires showed significant improvement from baseline but no difference between the groups (P = 0.12).
At 2 years follow-up, anterior colporrhaphy with polyglactin 910 mesh and abdominal paravaginal defect repair have similar success rates, with most objective failures being asymptomatic. Neurourol. Urodynam. 33:72–77, 2014. © 2013 Wiley Periodicals, Inc.
Anterior vaginal wall prolapse, occurs secondary to loss of pelvic support holding the bladder in place. Different theories have been proposed to explain the etiology of an anterior vaginal wall prolapse. One early theory suggests the presence of attenuation and weakening of the underlying vaginal layers supporting the bladder. This results in the formation of a central defect. Another theory suggests that anterior wall defects arise due to a break in the connective tissue/fascial attachment of the pelvic floor musculature to the fibrous thickening of the levator ani fascia or the white line. This results in weakness of the support system of the bladder, and subsequently leads to the formation of a site-specific or lateral wall defects. More recently, it has been proposed that most anterior wall defects result from impairment of the pubovisceral portion of the levators and the apical vaginal suspension complex (namely the cardinal and uterosacral ligaments).
Different surgical procedures have been developed that address the mechanism of failure in an anterior vaginal wall prolapse. Two commonly performed procedures are the anterior colporrhaphy and the paravaginal defect repair.[1, 5] The concept behind the anterior colporrhaphy is to repair a central defect by plicating the bladder under a layer of pubovaginal muscularis/adventitia and trimming the redundant weakened vaginal epithelium. On the other hand, the technique of the paravaginal defect repair is to identify and repair the breakage points of the pelvic floor supporting the bladder to the pelvic side wall.
Although anterior vaginal wall prolapse can be readily diagnosed on pelvic examination, the challenge is to accurately distinguish between the different types of anterior wall defects.[6-8] When different experienced gynecologic surgeons examine patients with anterior vaginal wall prolapse, there is little agreement as to the nature and extent of the defect.[6, 7] Moreover, others have questioned the clinical relevance of whether anterior wall prolapse is due to a central or lateral defect.[8, 9] This is especially true when one considers the absence of any randomized trials in support of a better surgical outcome between an anterior colporrhaphy versus a paravaginal (or lateral) defect repair.[9, 10]
We therefore hypothesized that there exists no difference in the success rate between an anterior colporrhaphy and an abdominal paravaginal defect repair. The aim of our study was to compare the post-operative objective and subjective outcomes between the two procedures up to 2 years after surgery.
MATERIALS AND METHODS
We performed a prospective randomized controlled trial comparing anterior vaginal colporrhaphy with polyglactin 910 mesh to abdominal paravaginal defect repair in patients with primary or recurrent symptomatic anterior vaginal wall prolapse. Inclusion criteria were women over the age of 18 with a symptomatic anterior vaginal wall prolapse scheduled for reconstructive surgery. Patients were excluded if they were pregnant or planning to have a future pregnancy, had more than one previous failed anterior vaginal wall repair, or were unwilling to give informed consent. The study received IRB approval and enrolment of all patients occurred after obtaining an informed consent. Randomization was performed using a computer generated random number table with type of surgery indicators placed in sealed opaque envelopes. Envelopes were opened at random by the research assistant on the pre-operative exam (consent day), typically a week before surgery. Although blinding of surgeons and patients was not possible due to the differences of the surgical approaches, all pre- and post-operative data collection was performed by the same investigator.
Pelvic examination was performed with the patient in the dorsal lithotomy position using the Peterson speculum. The extent of the maximum point of prolapse was measured with Valsalva or cough. We used the pelvic organ prolapse staging (POP-Q) system as recommended by the International Continence Society. All patients were asked to fill out the validated long forms of the pelvic floor distress inventory (PFDI), and the pelvic floor impact questionnaire (PFIQ), as well as the short form of the pelvic organ prolapse/urinary incontinence sexual questionnaire (PISQ-12).
The surgical steps of each procedure were standardized. Additional prolapse and incontinence procedures were allowed and performed in a standardized fashion in both groups. All patients received general anesthesia as well as preoperative antibiotic prophylaxis. An abdominal hysterectomy with sacrocolpopexy was the first procedure performed in women undergoing concomitant uterine prolapse surgery. This was followed by either the paravaginal defect repair or the anterior colporrhaphy. A mid-urethral retropubic TVT sling was then performed in women with stress urinary incontinence using the technique described by Ulmsten. Finally, a traditional posterior colporrhaphy was performed using delayed absorbable sutures in women with rectoceles.
The anterior vaginal wall colporrhaphy was performed in the traditional way. A dilute solution of vasopressin was used to infiltrate the vaginal epithelium. Dissection of the vaginal muscularis and adventitia was carried up to the level of the pubic rami. Plication of the bladder in the midline with 0-polydioxanone (Ethicon, Somerville, NJ) interrupted mattress sutures was performed extending from the undersurface of the pubic rami to the base of the bladder. Small bunched up (2 cm × 2 cm) pieces of polyglactin 910 (vicryl) mesh were secured within the imbricated fold of vaginal muscularis and adventitia as the surgeon tied the sutures encircling the mesh. The addition of vicryl mesh was shown to be superior to no mesh while performing traditional anterior colporrhaphy. After completion of the repair, and if necessary, the redundant vaginal epithelium was trimmed and closed with a 2-0 polyglactin 910 (vicryl) running locked suture.
For the paravaginal defect repair, we followed the technique described by Shull et al. Access to the pelvic sidewall was achieved retropubically. With the non-dominant finger of the surgeon in the vagina tenting it up, the assistant gently retracted the bladder medially. Sutures of 0-polydioxanone were placed through the full thickness of the vagina avoiding the epithelium. The needle was then passed through the fascia of the obturator internus or pubococcygeus muscles on the pelvic sidewall. Three to five stitches were placed at one cm intervals bilaterally extending from the bladder neck to the vaginal apex to re-approximate the vagina to the arcus tendineus fascia pelvis.
Patients were seen and examined post-operatively at 2 weeks, 6 weeks, 3 months, 12 months, and 24 months. All data collection, including POP-Q staging and quality of life questionnaires were performed by the research coordinator (LL) who had no involvement with the surgery. Demographic and preoperative characteristics measured were age, weight, height, parity, menopausal status, use of estrogen, smoking history, previous hysterectomy, previous bladder repair or other reconstructive surgery. Intraoperative and post-operative characteristics measured included: blood loss, operative time, other concomitant procedures, complications, hospital stay, number of days of catheter use, urinary tract infections. The PISQ-12, PFDI, and PFIQ were administered during clinic visits pre-operatively, and at 12 months post-operatively. Subjective outcomes were based on the PFDI, PFIQ, and a general patient satisfaction question. The latter was assessed by asking the following question: “How satisfied or dissatisfied are you with the result of surgery to lift your bladder up?” Possible responses were very satisfied, satisfied, neutral, dissatisfied, very dissatisfied. In the final analysis, very satisfied and satisfied were grouped together as being satisfied, and neutral, dissatisfied, and very dissatisfied were grouped together as being dissatisfied.
The primary outcome was the most dependent point of the anterior vaginal wall, with success defined as POP-Q stages 0 or I, and failure defined as stage II or more (most dependent part of the anterior vaginal wall protruding to 1 cm above the hymen or beyond). Secondary outcomes were POP-Q stage at other compartments, QOL questionnaires, and peri-operative complications. Power calculation was based on the following considerations. The recurrence rate of anterior vaginal wall prolapse has been reported to be 5% after an abdominal paravaginal defect repair[15, 17] and 3–70% after an anterior colporrhaphy repair,[14, 16, 18] and Stanton et al. We proposed that for a difference in success rates between the two procedures to be clinically significant, there should be at least 25% difference (i.e., a success rate of 70% for the anterior vaginal colporrhaphy). Based on these estimates, we needed 32 patients in each group with an alpha value of 0.05 and power of 80%. Accounting for a 10% drop-out rate, we decided to recruit a total of 70 patients, or 35 in each group.
Chi-square or Fisher's exact tests were used for categorical data, and student's t-test or Wilcoxon's rank sum test for continuous data. To measure the change in stage of prolapse and quality of life questionnaires before and after surgery the change from baseline values was calculated and the average change between the two different procedures was tested using the student's t-test. Logistic regression analysis was used to predict failure rate controlling for potential baseline and demographic confounders including the most recent follow-up time from surgery. Kaplan–Meier survival analysis was used to estimate the proportion of failure rate by the most recent follow-up time between the two surgeries.
The study was conducted from January 2006 to February 2010. We enrolled 70 patients with 35 in each group (Fig. 1). Complete baseline and demographic data were available for all 70 patients. There were three patients who were censored from the final analysis: one patient died before surgery of unrelated causes; two patients withdrew before the surgery. Furthermore, one patient randomized to anterior colporrhaphy with polyglactin 910 mesh decided she would rather have a paravaginal repair instead. Since this was an intent-to-treat analysis, she was analyzed with the anterior colporrhaphy group. Excluding her altogether, or including her with the paravaginal defect group did not make a difference in the primary or secondary outcomes (data not shown). Sixty-seven (96%) had one or more post-operative visit up to 2 years from surgery, 34 in the anterior colporrhaphy group and 33 in the paravaginal defect repair group (Fig. 1).
The demographic and other health characteristics were similar between the anterior colporrhaphy with polyglactin 910 mesh and paravaginal defect repair groups (Table I). Moreover, most intra and post-operative parameters including blood loss, catheter use, and hospital stay were similar. Complications associated with the index surgery were also similar: four for anterior colporrhaphy and seven for paravaginal defect repair. These included: urinary tract infection (UTI), ureteral kinking, and cystotomy in the anterior repair group; UTI, blood transfusion, small bowel obstruction, and one patient who had several complications including a cystotomy, post-operative wound breakdown and infection. It is likely that many of these complications were due to concomitant procedures and not directly related to the type of anterior vaginal wall repair. There were 11 more hysterectomies in the anterior colporrhaphy group compared to the paravaginal repair group (P = 0.01). A similar number of women in both groups had mid-urethral slings for stress urinary incontinence, rectocele and apical vault repairs (Table II).
|Anterior colporrhaphy (n = 35)||Paravaginal repair (n = 35)||P-valuea|
|Age (mean, SD)||54.3 (10.6)||53.4 (12.7)||0.74|
|BMI in kg/m2 (mean, SD)||27.8 (4.3)||29.2 (4.7)||0.19|
|Parity (median, IQR)||2 (2, 3)||2 (2, 3)||0.13|
|Pre-op stage anterior (median, IQR)||3 (2, 3)||3 (2, 4)||0.78|
|Pre-op point Ba in cm (mean, SD)||+1.4 (1.5)||+1.8 (1.9)||0.29|
|Pre-op point C in cm (mean, SD)||−3.4 (1.4)||−3.3 (1.3)||0.84|
|Menopause (n, %)||21 (60%)||20 (57%)||0.92|
|HRT (n, %)||3 (9%)||4 (11%)||0.71|
|Smoking (n, %)||5 (14%)||6 (17%)||0.70|
|Past hysterectomy (n, %)||9 (26%)||15 (43%)||0.11|
|Any urinary incontinence (n, %)||31 (89%)||29 (83%)||0.78|
|Sexually active (n, %)||22 (63%)||25 (71%)||0.82|
|Previous bladder repair (n, %)||5 (14%)||2 (6%)||0.43|
|Anterior colporrhaphy (n = 34)||Paravaginal repair (n = 33)||P-valuea|
|EBL in cc (median, IQR)||225 (150,350)||200 (100,300)||0.34|
|Total operation time in minutes (mean, SD)||283 (84)||267 (85)||0.45|
|Complications (n, %)||4 (12%)||7 (21%)||0.13|
|Hospital stay in days (median, IQR)||3 (2, 3)||2 (2, 3)||0.36|
|Catheter use in days (median, IQR)||6 (2, 9)||5 (2, 8)||0.76|
|Number of other procedures (median, IQR)||4 (3, 4)||3 (2, 4)||0.17|
|Hysterectomy (n, %)||25 (74%)||14 (42%)||0.01|
|Open sacrocolpopexy (n, %)||25 (74%)||24 (73%)||0.94|
|Midurethral slings (n, %)||26 (76%)||25 (76%)||0.95|
|Rectocele (n, %)||19 (56%)||16 (48%)||0.55|
Pre-op anterior wall POP-Q stage was similar between the two groups with a median stage of III. Of those patients who had a 2-year follow-up, objective failure rates for the anterior colporrhaphy with polyglactin 910 mesh and paravaginal defect repair groups were 8/25 (32%) and 10/25 (40%), respectively, P = 0.56. There were no significant differences in objective failure rates as defined by POP-Q stage between the two groups at all follow-up time intervals. The overall objective failure rate at last follow-up was 12/34 (35%) for anterior colporrhaphy with polyglactin 910 mesh and 11/33 (33%) for paravaginal defect repair (P = 0.61). There were no failures in the central compartment in either group. There was one recurrence (stage II) in the posterior compartment in each group.
This study was designed during the period of 2004–2005 when stage II prolapse or more was regarded as an acceptable outcome measure. More recently, a more acceptable objective outcome for surgical failure has become the degree of prolapse beyond the hymen. Here again, there was no statistically significant difference in failure rates by post-operative point Ba or change in point Ba between the two surgery types by time of follow-up (Table III). Furthermore, failure was independent of pre-operative stage. Ratio of success to failure was 20:11 and 23:13 in women with stage II versus those with stage III/IV, respectively (P = 0.87).
|Anterior colporrhaphy (n = 34)||Paravaginal repair (n = 33)||P-valueb|
|Pre-op anterior stage (median, IQR)||3 (2, 3)||3 (2, 4)||0.78|
|Post-op (stage II or more)|
|6 weeks: n (%, N)||2 (6%, 34)||1 (3%, 30)||0.99|
|3 months: n (%, N)||4 (13%, 30)||2 (7%, 30)||0.67|
|1 year: n (%, N)||6 (23%, 26)||7 (28%, 25)||0.69|
|2 years: n (%, N)||8 (32%, 25)||10 (40%, 25)||0.56|
|Last follow-up: n (%, N)||12 (35%, 34)||11 (33%, 33)||0.87|
|Pre-op point Ba in cm (mean, SD)||+1.4 (1.5)||+1.8 (1.9)||0.29|
|Post-op change in point Ba in cm|
|6 weeks (mean, SD)||−4.02 (1.65)||−4.4 (2.13)||0.42|
|3 months (mean, SD)||−3.83 (1.89)||−4.1 (1.87)||0.63|
|1 year (mean, SD)||−3.30 (1.75)||−3.5 (2.16)||0.74|
|2 years (mean, SD)||−2.84 (1.8)||−3.7 (1.99)||0.13|
|Satisfaction (satisfied/very satisfied) at last follow-up: n (%)||30 (88%)||24 (73%)||0.11|
Subjective failure rates were lower than the objective failures for both groups. Patients reported satisfaction (satisfied or very satisfied) rates of 30/34 (88%) for the anterior colporrhaphy with polyglactin 910 mesh and 24/33 (73%) for the paravaginal defect repair group, P = 0.11 (Table III). Comparing QOL questionnaires including the PFDI, PFIQ, and the PISQ-12 (for those who were sexually active) before and after surgery showed a significant improvement within each surgical group (P < 0.05); however, there was no significant difference in the change in score of any of the questionnaires between the two surgical groups (P = 0.12; Table IV). Of the 13 women who were sexually active at year one post-operatively, only two (15%) had worsening of their PISQ-12 scores in the anterior colporrhaphy group versus 3 out 20 (15%) in the paravaginal defect repair group. More specifically, 1 year after surgery, only three women (two in the anterior and one in the paravaginal defect repair group) had a response of “usually” or “always” to question number 5 of the PISQ-12: “Do you feel pain during sexual intercourse.”
|Anterior colporrhaphy (n = 34)||Paravaginal repair (n = 33)||P-valuea|
|Pre-op PFDI (UDI) mean (SD)||125 (55)||106 (48)||0.15|
|Post-op PFDI (UDI) change mean (SD)||−105 (66)||−83 (47)||0.16|
|Pre-op PFDI (POPDI) mean (SD)||127 (64)||116 (59)||0.47|
|Post-op PFDI (POPDI) change mean (SD)||−98 (72)||−83 (57)||0.37|
|Pre-op PFDI (CRADI) mean (SD)||95 (65)||102 (69)||0.69|
|Post-op PFDI (CRADI) change mean (SD)||−66 (73)||−65 (52)||0.99|
|Pre-op PFIQ (UIQ) mean (SD)||208 (62)||206 (95)||0.93|
|Post-op PFIQ (UIQ) change mean (SD)||−91 (71)||−82 (84)||0.64|
|Pre-op PFIQ (POPIQ) mean (SD)||148 (61)||152 (81)||0.84|
|Post-op PFIQ (POPIQ) change mean (SD)||−39 (60)||−51 (88)||0.55|
|Pre-op PFIQ (CRAIQ) mean (SD)||145 (58)||148 (77)||0.87|
|Post-op PFIQ (CRAIQ) change mean (SD)||−38 (62)||−29 (71)||0.60|
|Pre-op PISQ-12 mean (SD)||17 (8)||14 (6)||0.12|
|Post-op PISQ-12 change mean (SD)||−6 (9)||−4 (6)||0.36|
Finally, Figure 2 shows survival analysis from time of surgery until last follow-up time by whether a failure occurred (if not then censored) at that most recent follow-up time. There was no difference between the two groups using the log-rank test.
At a follow-up of up to 2 years, this randomized trial comparing anterior colporrhaphy with polyglactin 910 mesh to abdominal paravaginal defect repair for anterior vaginal wall prolapse failed to show any significant difference in objective or subjective success rates between the two procedures. Both procedures had a similar objective failure rate of about 33–35%, with most being asymptomatic. Our objective and subjective results are similar to that reported in the literature,[5, 14] and Chmielewski et al. Complication rates associated with the repair were also similar with the majority of complications being either minor (i.e., UTI) or unrelated to the actual bladder repair.
Isolated anterior vaginal vault prolapse was not that common; many women had a concomitant procedure including mid-urethral sling, hysterectomy, posterior vaginal wall repair and sacrocolpopexy. Sacrocolpopexy was chosen as the procedure of choice for central wall defects because, in our experience, women with recurrent cystoceles and concurrent central vault prolapse have better anatomic outcomes when compared to vaginal vault suspension procedures. There were a total of 18 women (nine in each group) who did not have a sacrocolpopexy. The mean (standard deviation) of pre-operative point C in this group was −6 cm (±4). These women had isolated anterior compartment defects with or without stress UI, and no central wall defects.
It could be argued that the final subjective and objective outcomes of no difference between the two groups were a result of the combination of prolapse and incontinence procedures performed rather than the actual anterior vaginal wall prolapse repair. This argument may be true. In many circumstances, obtaining a good vaginal vault support with a sacrocolpopexy may obviate the need for additional anterior vaginal wall repair. We acknowledge this as a weakness of our study; however, most other published trials on anterior colporrhaphies also include a similar percentage of patients with concomitant vault suspension procedures.[8, 16, 22-24] It is well recognized that isolated compartment defects are rare, and that ideally all defects should be corrected at the same time.[5, 25] Large multicenter trials will be needed to compare failure rates of isolated anterior colporrhaphy to paravaginal defect repair in the absence of other pelvic floor repair.
The pathophysiology of an anterior vaginal wall prolapse remains controversial. Recent evidence suggests that the majority of anterior vaginal wall prolapse may be neither due to a localized central nor lateral wall defect. Using dynamic MRI, Hsu et al. have shown that 77% of anterior wall prolapse is explained by apical descent and midsagittal anterior vaginal wall length. There are no clear diagnostic criteria to distinguish one type of defect from the other in the clinical setting. Moreover, there is radiological (MRI) evidence that repair of a central wall defect (anterior colporrhaphy) corrects 50% of lateral wall defects. Hence, and based on the findings from our study, the clinical relevance of anterior vaginal wall prolapse nomenclature is brought into question.
In one randomized trial, use of an absorbable mesh at the time of anterior colporrhaphy had a higher cure rate than anterior colporrhaphy where no mesh was used. Others did not find any difference when bovine pericardium was used to reinforce the anterior colporrhaphy. Furthermore, others have shown that anterior repair with a polypropylene mesh overlay or polypropylene mesh kit procedures[24, 27] had superior objective (but not necessarily subjective) cure rates when compared to the traditional anterior colporrhaphy. However, use of synthetic non-absorbable mesh kits is associated with complications such as mesh erosion rates of as high as 25%. Their use has also become controversial more recently with the FDA safety communication: “Update on Serious Complications Associated with Transvaginal Placement of Surgical Mesh for Pelvic Organ Prolapse”. On the other hand, although the paravaginal defect repair appears to be an appropriate surgical option, there exists no evidence to suggest that it is superior to the anterior colporrhaphy. To avoid abdominal access for isolated anterior vaginal wall prolapse, one can alternatively perform vaginal paravaginal defect repairs; however, here again, there exist no randomized control trials comparing the latter to the anterior colporrhaphy.
Because of the surgical nature of the procedures, this study could not be blinded. However, an attempt was made to make data collection free of bias by having the research coordinator, not involved in surgery, perform all the data collection pre and post-operatively. Most of our patients (96%) had at least one post-operative visit. However, it was not possible to examine all patients at 2 years post-operatively despite all our efforts to encourage follow-up. However, women who failed to follow-up at 2 years were equally distributed between the two groups. Therefore, even considering them as failures would have still resulted in similar non-significant differences between the two groups.
In conclusion, we found no difference in objective and subjective success rates in women undergoing anterior colporrhaphy with polyglactin 910 mesh versus abdominal paravaginal defect repair for up to 2 years from surgery.
- 20Reanalysis of a randomized trial of 3 techniques of anterior colporrhaphy using clinically relevant definitions of success. Am J Obstet Gynecol 2011; 205:e1–8., , , et al.
- 23Outcomes after anterior vaginal wall repair with mesh: A randomized, controlled trial with a 3 year follow-up. Am J Obstet Gynecol 2010; 203:e1–8., , , et al.
- 28http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm 262435.htm (accessed August 11, 2011).