Background
There are few adequately powered long-term trials comparing midurethral sling and Burch colposuspension. Recent concerns about synthetic mesh with new stringent clinical and research governance support the need for evidence to facilitate shared decision making.
Objective
This study aimed to compare long-term outcomes of open Burch colposuspension with the retropubic midurethral sling.
Study Design
A matched cohort study of 1344 women with urodynamic stress incontinence (without intrinsic sphincter deficiency) who underwent surgery for stress urinary incontinence. Women had either open Burch colposuspension or the retropubic midurethral sling, from January 2000 to June 2018, in a tertiary center. Follow-up was by chart review and one-time phone follow-up until 2019, using a dedicated database. Primary outcomes were the presence or absence of stress urinary incontinence on follow-up, the success of index surgery based on response to validated questionnaires of patient-reported outcomes, and retreatment rates. Secondary outcomes are described below. Matching (1:3) was done at baseline to avoid confounding.
Results
The study included 1344 women who had either Burch colposuspension (336) or retropubic midurethral sling (1008). Mean follow-up was 13.1 years for Burch colposuspension and 10.1 years for retropubic midurethral sling. In the Burch colposuspension group, 83.0% of patients (279 of 336) reported no ongoing stress urinary incontinence at the time of the latest follow-up vs 85.0% (857 of 1008) in the retropubic midurethral sling group ( P =.38). Success in terms of the latest reported International Consultation on Incontinence Questionnaire—Urinary Incontinence Short Form (defined as International Consultation on Incontinence Questionnaire—Urinary Incontinence Short Form score of ≤6) where these data were available were similar within both groups: 76.0% (158 of 208 where this was available) in Burch colposuspension vs 72.1% (437 of 606 where this was available) in retropubic midurethral sling ( P =.32). Where this information was available, success defined by a Patient Global Impression of Improvement of “very much improved” and “much improved” was similar between Burch colposuspension and retropubic midurethral sling groups (84.1% [243 of 289] vs 82.0% [651 of 794]; P =.88). Where data were available, 88.1% of women (178 of 202) in the Burch colposuspension group said they were very likely to recommend the surgery to family or a friend vs 85.0% (580 of 682) in retropubic midurethral sling ( P =.30).Overall, 3.6% needed repeat incontinence procedures (13 in Burch colposuspension group [3.8%] vs 35 in retropubic midurethral sling group [3.5%]; P =.73). The incidence of mesh exposure was 1.0 %. Notably, 1 Burch colposuspension patient had a suture in the bladder during follow-up; 5 patients have reported long-standing pain across the study population. Overall, 51 women reported new-onset overactive bladder symptoms on follow-up: 10 of 336 (3.0%) had Burch colposuspension and 41 of 1008 (4.1%) had retropubic midurethral sling ( P =.41) . The need for future prolapse surgery per index procedure was 3.3% after Burch colposuspension vs 1.1% postretropubic midurethral sling ( P =.01). Moreover, 9 of the 11 patients who needed a prolapse repair after Burch colposuspension required a posterior repair. The incidence of long-term severe voiding difficulty needing self-catheterization was similar in both groups (0.3% in Burch colposuspension and 0.5 % in retropubic midurethral sling group; P =1.00).
Conclusion
This study shows no difference in success, patient satisfaction, or complications between Burch colposuspension and retropubic midurethral sling, although the risk of posterior compartment prolapse operations after Burch colposuspension is increased . Reoperation rates for incontinence were similar in both groups. Chronic pain was a rare outcome.
Introduction
The current lifetime risk of requiring surgery for stress urinary incontinence (SUI) in the United States is approximately 13.5%. The Burch colposuspension (BC) was first described for the surgical management of SUI by John C. Burch in 1961 and is performed through a low transverse incision in the abdomen. It involves the placement of 2 to 3 permanent sutures suspending the vagina to the Cooper’s ligament (iliopectineal ligament) bilaterally to support the urethra and bladder neck. Success ranges from 85% to 90% (1 year postoperatively) to 70% (5 years postoperatively). , Although originally described as an open abdominal procedure, this can also be performed laparoscopically with equally good results. , Up until the 1990s, the BC was considered the gold standard surgery for SUI and accounted for more than 80% of SUI operations.
Why was this study conducted?
Although there have been long-term reports of open Burch colposuspension and midurethral sling outcomes separately, there are few published comparative reports of long-term outcomes of Burch colposuspension and midurethral sling.
Key findings
This long-term comparison of open Burch colposuspension and retropubic midurethral sling indicates both procedures are safe and have the same levels of subjectively reported success.
What does this add to what is known?
Although the midurethral sling is minimally invasive, long-term satisfaction rates were similar for both. Reoperation rates for incontinence were identical in both groups.
Midurethral slings (MUSs) were developed in the 1990s based on the work of Petros and Ulmsten in developing the tension-free vaginal tape (TVT), and the transobturator sling followed this, which was developed in 2001 by Delorme. This innovation revolutionized surgery for SUI owing to the minimally invasive nature and hence could be performed as a day procedure with local anesthesia. , The retropubic MUS (RP-MUS) proved itself to be a robust alternative to BC with subjective cure rates of 85% at 4 years, , with a low incidence of complications including mesh exposure and chronic pain. Although these complications are uncommon, they can cause considerable morbidity and impact on the quality of life. This has led to the MUS being the subject of numerous lawsuits with the subsequent withdrawal of many MUS products from the market after multimillion-dollar payouts for class actions. Regulatory authorities have created new and more robust guidelines for the long-term follow-up of incontinence slings and tighter regulation of innovation related to incontinence surgery. This has caused renewed interest in surgical alternatives to the MUS, including BC, autologous fascial sling (AFS) or pubovaginal sling, and urethral bulking agents. ,
The rapid replacement of BC by MUS as the first choice of surgery for SUI by surgeons and patients has meant that, although there have been long-term reports of BC and MUS outcomes separately, there are few published comparative reports of long-term outcomes of BC and MUS. This study aims to fill this research gap by comparing the long-term results of BC with the RP-MUS.
Materials and Methods
This is a matched cohort study comprising of 1344 women with SUI proven on urodynamic studies who had surgery for SUI by either open abdominal BC or RP-MUS (alone or concomitantly with surgery for prolapse) in the period from January 2000 to June 2018 in our Department of Urogynecology and entered in a dedicated electronic database. The demographic, clinical, and follow-up and questionnaire (as part of follow-up) data were entered retrospectively through regular chart reviews. A one-time follow-up round was conducted in the year 2019 for the whole cohort to try and obtain the most recent information about patient outcomes. After receiving consent from the patient via telephone or returned via mail, the current patient-reported incontinence scores and quality of life were recorded as a one-time evaluation, and all the accruing data were entered contemporaneously on the database.
The institutional ethics committee approved this study. We excluded women with intrinsic sphincter deficiency (ISD) because the BC was not our preferred surgical option being less efficacious in ISD. We also excluded women who had laparoscopic colposuspension to avoid heterogeneity.
The phone and postal follow-up used structured validated questionnaires, including the International Consultation on Incontinence Questionnaire—Urinary Incontinence Short Form (ICIQ-UI SF) and Patient Global Impression of Improvement (PGI-I). Nonvalidated structured questionnaires used routinely in our unit for clinical follow-up of pelvic floor dysfunction and urogynecologic surgery were also used.
All the BCs were performed through a small suprapubic incision with 3 permanent suspension sutures of braided Ethibond (Ethicon, Somerville, NJ), and RP-MUSs were performed as per standard technique as recommended by the manufacturer.
Primary outcomes were the presence or absence of SUI on follow-up, the success of index surgery based on response to validated questionnaires of patient-reported outcomes, and retreatment rates. ICIQ-UI SF success was defined using a cutoff derived from a previous study by Karmakar et al, which showed a high predictive value of success for a postoperative ICIQ-UI SF of ≤6. PGI-I of “very much improved” and “much improved” was also indicative of a successful outcome. , Secondary outcomes were short- and long-term complications such as bladder perforation, urinary retention, specific RP-MUS complications (loosening of the sling, specific mesh complications), chronic pain, voiding difficulty, long-term rates of overactive bladder, and need for future surgery for prolapse.
Owing to disproportionately more subjects in the RP-MUS group than BC, subjects in each group were matched (3:1) with respect to these minimum compulsory variables obtained from current literature to avoid confounding : (1) age, (2) parity and mode of delivery, (3) body mass index at the time of surgery, (4) presence and type or absence of pelvic organ prolapse, and (5) presence of urodynamic features of detrusor overactivity or low bladder compliance.
After the ascertainment of the index surgery (first procedure for SUI), we selected 3 RP-MUSs for each BC case. Randomization was performed with the simple random row selection function of JMP statistical software (version 10.0; SAS Institute, Cary, NC). The same exclusion criteria were applied to both index surgery groups. Primary statistical analysis was performed using IBM SPSS Statistics (version 11-2018, IBM Australia Ltd, St Leonards, New South Wales, Australia) and the jamovi Project (Newcastle, NSW, Australia) 2020; jamovi [Version 1.2] Computer Software, retrieved from https://www.jamovi.org .
Chi-squared and Fisher exact tests were used for categorical variables and Student t test for continuous variables. Statistically significant results were defined as P <.05. For the duration of our follow-up, using 2 adequately powered trials reporting on long-term outcomes, , a sample size per group of 78 is sufficient to yield a significance level (alpha) of 5% and power of 80% to detect a 15% difference in cure rates. This indicates we were able to achieve an adequate sample size for our study.
Results
The study included 1344 women. The index procedure was BC in 336 and RP-MUS in 1008. The demographic and clinical data are presented in Table 1 . The mean follow-up period was 13.1 years for BC and 10.1 years for RP-MUS. For 412 patients (30.6%), we have one-time follow-up data from the round of phone call and mail-in questionnaires performed in 2019 (152 BCs [45.2%] and 260 RP-MUSs [25.8%]). Concomitant procedures were performed in 320 patients (23.8%) ( Table 1 ). For the remaining subjects, the follow-up data were extracted from their last follow-up visit.
| Variables | Index surgery Burch colposuspension (n=336) | Index surgery RP-MUS (n=1008) | P value |
|---|---|---|---|
| Mean age (y±SD) | 55.6±10.1 | 56.1±11.0 | NS |
| Mean BMI (±SD) | 28.3±4.4 | 28.9±4.6 | NS |
| No. of patients with a BMI of >30 kg/m 2 | 128 (38.1%) | 403 (39.9%) | NS |
| No. of patients sexually active at time of surgery | 217 (64.6%) | 662 (65.6%) | NS |
| No. of patients postmenopausal | 269 (80.1%) | 816 (80.1%) | NS |
| Mean parity | 2.5 | 2.6 | NS |
| No. of patients who had had previous prolapse surgery | 104 (30.9%) | 323 (32.0%) | NS |
| No. of patients who had had previous hysterectomy | 122 (36.3%) | 387 (38.3%) | NS |
| Urodynamic diagnosis | NS | ||
| 282 (83.9%) | 851 (84.4%) | |
| 54 (16.1%) | 157 (15.5%) | |
| Stand-alone procedures | 268 (79.7%) | 756 (75.0%) | |
| Concomitant surgery | 68 (20.2%) | 252 (25.0%) | .07 |
| Type of concomitant surgery |
|
|
Primary outcomes results
Patient-reported outcomes were similar in both groups ( Table 2 ). Response to questionnaire item addressing “likelihood of referring to family and friends” was available for 202 of 336 BC patients (60.1%) and 682 of 1008 RP-MUS patients (67.7%). Notably, 178 of 202 patients (88.1%) in the BC group were very likely to recommend the surgery to family or friends vs 580 of 682 (85.0%) in the RP-MUS group ( P =.30). Overall, 48 women (3.6%) needed a repeat incontinence procedure, 13 (3.8%) in BC group vs 35 (3.5%) in RP-MUS group ( P =.73).
