The Burch colposuspension for recurrent urinary stress incontinence following retropubic continence surgery

Authors

  • Christopher Maher,

    Urogynaecology Fellow , Corresponding author
    1. Urogynaecology Department, The Royal Women S Hospital and Mercy Hospital for Women, Melbourne, Australia
      Correspondence: Dr C. Maher, Royal Women's Hospital, Grattan Street, Carlton, 3053 Victoria, Australia.
    Search for more papers by this author
  • Peter Dwyer,

    Director (Urogynaecology)
    1. Urogynaecology Department, The Royal Women S Hospital and Mercy Hospital for Women, Melbourne, Australia
    Search for more papers by this author
  • Marcus Carey,

    Urogynaecologist
    1. Urogynaecology Department, The Royal Women S Hospital and Mercy Hospital for Women, Melbourne, Australia
    Search for more papers by this author
  • Donna Gilmour

    Urogynaecology Fellow
    1. Urogynaecology Department, The Royal Women S Hospital and Mercy Hospital for Women, Melbourne, Australia
    Search for more papers by this author

Correspondence: Dr C. Maher, Royal Women's Hospital, Grattan Street, Carlton, 3053 Victoria, Australia.

Abstract

Objective To evaluate the Burch colposuspension with Cherney incision in women with recurrent urinary stress incontinence after retropubic continence surgery.

Design A retrospective review.

Participants All 53 women had recurrent urinary stress incontinence after retropubic continence surgery with an average of 2.1 (range 1–5) previous failed continence procedures per woman.

Main outcome measures Subjective and objective success rates and complications including detrusor instability, voiding difficulties and genital prolapse.

Results Forty-seven women (89%) had no or occasional (< one episode per week) stress or urge incontinence. One woman had persistent stress incontinence and five urge incontinence. Forty-two women (80%) rated their surgery as being highly successful and 38 women (72%) had no urinary leakage due to genuine stress incontinence or detrusor instability on repeat urodynamic evaluation. Three women (6%) developed de novo detrusor instability post-operatively. Two women (4%) had voiding difficulties post-operatively that necessitated the use of intermittent self-catheterisation for at least four months. The median length of follow up was nine months (4–72). In 39 women (73%), marked retropubic fibrosis was found at the time of surgery.

Conclusions Marked retropubic fibrosis should be expected in women with recurrent stress inconti- nence after retropubic continence surgery. The Burch colposuspension with the assistance of the Cherney incision and sharp dissection of retropubic fibrosis is an effective and safe procedure for women with this condition.

INTRODUCTION

All surgical continence procedures are associated with unsuccessful outcomes, and the management of women with recurrent urinary stress incontinence remains a complex and challenging problem. Whilst most authors believe that previously failed continence surgery is a significant risk factor for subsequent continence surgery1,2, not all agree3,4.

Almost all studies of recurrent urinary stress incontinence reviewed women who had a variety of continence procedures including anterior vaginal repairs, needle suspensions, retropubic suspension procedures or cess rate of surgery in women with recurrent urinary incontinence varies with the type of continence surgery previously undertaken. In 1974 Lee suggested that “a retropubic operation, whether in the form of a suspension, suspension plus plication, or a fascia1 sling, appears to offer the patient with recurrent incontinence the best chance of permanent support of the bladder neck and urethra and good urinary control18. The retropubic approach to recurrent genuine stress incontinence after retropubic surgery is a considerably greater challenge than that experienced following a failed anterior vaginal repair, due to fibrosis and scarring of the Cave of Retzius.

With the increasing popularity of retropubic continence surgery over vaginal repair in the management of primary stress incontinence, it is inevitable that we will see more women with recurrent urinary stress incontinence following retropubic continence surgery. The only report of the findings and efficacy of repeat retropubic continence surgery involved women whose primary surgery was performed between 25 and 40 years ago18. During this time there has been a significant change in the diagnostic and surgical aspects of urinary stress incontinence management. The aim of this study is to report the objective and subjective success rates and complications of the modified Burch colposuspension in women with recurrent urinary stress inconti- nence following failed retropubic surgery.

METHODS

Between April 1993 and December 1997, 53 women with recurrent genuine stress incontinence following at least one failed retropubic continence procedure underwent a modified Burch colposuspension. Previous retropubic continence surgery included open and laparoscopic Burch colposuspension, needle suspensions, slings and Marshall-Marchetti-Krantz procedures. Due to the tertiary referral nature of our department, it is impossible to differentiate which type of needle suspension or sling was previously performed. Three women had had the primary retropubic continence surgery per- formed within our department: one open Burch colposuspension, one pubovaginal sling and one Stamey needle suspension. All women had failed to respond to pelvic floor exercises.

All women were assessed pre- and post-operatively using a standardised history, examination and urodynamic evaluation. Vaginal prolapse was assessed in the left lateral position and graded using a modified Bayden Walker clasification19. Using a Sims speculum (McFarlone Medical, Sydney, Australia) with the patient in the left lateral position, bladder neck mobility was assessed visually before and during straining and coughing. The bladder neck was graded as well- or poorly-supported. Selection criteria for the Burch colposuspension included women with genuine stress incontinence and adequate mobility of the anterior vaginal wall to allow colposuspension.

Urodynamic assessment was performed using subtracted dual-channel cystometry at a filling rate of 100 mumin with simultaneous pressure flow studies. Urethral pressure measurements were made with the patient in the sitting position at rest and during coughing, with the bladder empty and again at cystometric capacity. Bladder and urethral pressures were measured with a dual sensor 7 French microtransducer catheter (Gaeltec, Isle of Skye, Scotland) as described by Asmussen and Ulmsten20. Abdominal pressure measurement was recorded with a single-sensor 7 French catheter placed vaginally. The diagnosis of voiding dysfunction was made if the maximal urinary flow rate was c > 15 mWs on two occasions with a voided volume > 150 mL and the residual urine was > 150 mL.

Surgical outcome was assessed using patient symptoms, a patient determined satisfaction score and post-operative urodynamic evaluation. At review, women were asked to describe their satisfaction with the procedure on a visual analogue scale ranging from zero to ten (Fig. 1.) Women who scored their surgery as eight or greater were classified as a patient determined success. Subjective success was defined as no or occasional (c once a week) stress or urge incontinence. Objective success was defined as no urinary leakage secondary to stress incontinence or detrusor instability on repeat urodynamic assessment.

Figure 1.

Patient satisfaction score.

Statistical analyses using Pearson's > χ2 and Fisher's exact tests were used where appropriate to look for differences in variables. The Wilcoxon rank sum test was used to assess differences in paired variables and the McNemar test for paired binary data. All terminology conforms to the recommendations of the International Continence Society unless otherwise stated21.

Surgical technique

The Burch colposuspension was performed through a low pfannenstiel incision. A Cherney incision was used to detach the rectus abdominus muscles from their insertion on the pubic bone when retropubic fibrosis was present or when the bladder neck was well-supported. This incision facilitated improved exposure to the Cave of Retzius, enabling retropubic fibrosis and scarring to be safely freed. Sharp dissection, using scissors or diathermy, was usually required to free the morbidly adherent bladder and urethra from the posterior surface of the pubis. This dissection should be performed directly behind the pubis to minimise inadvertent damage to the bladder or urethra and is essential to allow exposure to the paravaginal tissue surrounding the blad- der neck. It is important to be mindful of the course of the obturator neurovascular bundle during the dissection of retropubic fibrosis as bleeding from these vessels is difficult to control and may be associated with significant post-operative pain if the obturator nerve is entrapped in ligatures. Once the retropubic dissection is completed and the bladder is free from adhesions, the bladder is reflected medially over the paravaginal fascia and three number 1 ethibond sutures were placed bilat- erally, through the vaginal wall at the level of the bladder neck. These sutures were approximately 1 cm apart and were used to suspend the vagina within 3–4 cm of the ipsilateral ilio-pectineal ligament.

The rectus abdominus muscle is re-attached to the insertion points on the pubic rami with a delayed absorbable number 1 PDS suture. After routine abdominal closure a cystoscopy was performed to visualise the insertion of the suprapubic catheter, to check ureteric patency and to exclude intravesical sutures. All surgery was performed by or under the supervision of the two senior authors. The suprapubic catheter was clamped on day three and removed once consecutive residual urine volumes were < 100 mL with a voided volume of > 150 mL.

RESULTS

Patient demographics and previous continence surgery is shown in Table 1. All women had genuine stress incontinence and had undergone at least one unsuccessful retropubic continence procedure. They had previously undergone 112 continence procedures with an average of 2.1 (range 1–5) previous failed continence procedures per woman. Five women had undergone two previous retropubic continence procedures. Thirteen women (24%) had no previous vaginal continence surgery. Six women (11 %) had a maximum urethral closure pressure of 20 cmH, O or less at cystometric capacity. Three women had co-existing detrusor instability and three had voiding dysfunction pre-operatively. Thirty-three women (62%) had a well-supported blad- der neck pre-operatively.

Table 1.  Patient details. Values are given as median [range] or n (%). BMI =body mass index; MMK = Marshall-Marchetti-Krantz procedure.
  1. *Five women had two previous retropubic continence procedures.

Age (years)51 [32–85]
BMI (kg/m2)29 [19–41]
Parity3 [0–10]
Previous continence surgery 
Vaginal repairs 
  013 (24)
  129 (55)
  29 (17)
  ≥32 (4)
Previous retropubic continence surgery* 
Burch colposuspensions 
  Open26 (49)
  Laparoscopic8 (15)
Needle suspensions12 (19)
MMK8 (15)
Sling4 (8)

The median operating time was 68 minutes (40–124), inpatient days six (3–12) and catheter days five (3–27). The Cherney incision was utilised in 48 women (90%). At surgery, 39 of 53 women (73%) were found to have severe retropubic fibrosis. The median length of follow up was nine months (4–72). Pre and post-operative urinary symptoms and urodynamic findings are shown in Tables 2 and 3. Forty-seven of the 53 women (89%) had no or occasional (< one episode per week) stress or urge incontinence. One woman had persistent stress incontinence, and five urge incontinence. Forty-two of the 53 women (80%) rated their surgery as being highly suc- cessful, and 38 (72%) had no urinary leakage due to genuine stress incontinence or detrusor instability on repeat urodynamic evaluation.

Table 2.  Pre- and post-operative urinary symptoms for 53 women. Values are given as n.
SymptomsPre-operativePost-operative
  1. *< once per week.

  2. P < 0.001.

Stress incontinence  
  None or occasional*052
  Frequent531
Urge incontinence  
None or occasional*4148
  Frequent125
Urgency  
  None or occasional*4345
  Frequent108
Complete emptying  
  Yes5150
  No23
Urinary stream  
  Satisfactory5251
  Poor12
Table 3.  Pre- and post-operative urodynamic data. Values are given as median (range) unless otherwise indicated. MUCP = maximum urethral closure pressure.
VariablePre-operativePost-operativeP
Flow rate (mUs)27 (6–89)20 (7–41)0.042
Residual volume (mL)5 (0–122)7 (0–225)0.4372
Capacity (mL)496 (295–500)498 (332–500)0.782
Urethral length (mm)21 (7–32)21 (10–36)0.816
MUCP40 (6–166)46 (8–110)0.294
Transmission (%)60 (30–100)96 (37–100)< 0.001
Stress incontinence (n)5310< 0.001
Detrusor instability (n)350.486

There was a significant reduction in urine flow rate post-operatively following the Burch colposuspension (P=0.04), and also a significant increase in the trans- mission of abdominal pressure to the urethra following the Burch colposuspension (P < 0.001). There was no significant difference in residual urine volume, bladder capacity, urethral length or maximum urethral closure pressure post-operatively (Table 3). Three of the 20 women (1 5%) with a poorly-supported bladder neck pre-operatively had persistent genuine stress incontinence post-operatively. Seven of the 33 women (2 1 %) with a well-supported bladder neck pre-operatively had persisting genuine stress incontinence post-operatively. Although there was a trend towards a poor surgical out- come in women with poor bladder neck mobility, this trend was not significant (P= 0.73).

Three of five women (60%) with pre-operative detrusor instability had stable bladders post-operatively. Three women (6%) developed de novo detrusor instability post-operatively. One of the three women with pre-operative voiding difficulties had persisting voiding difficulties post-operatively. Two women (4%) developed voiding difficulties post-operatively which necessitated the use of intermittent self-catheterisation for at least four months post-operatively.

Ten women underwent concomitant prolapse surgery, including seven posterior vaginal repairs, two abdominal sacral colpopexys and one abdominal paravaginal repair. Two women (4%) developed grade two or greater posterior vaginal wall prolapse post-operatively. The objective success rate of the modified Burch colposuspension in women undergoing concomitant prolapse surgery was 70%. At surgery, one woman had an inadvertent cystotomy, and in one the obturator vein was lacerated during dissection of retropubic fibrosis. No woman had an intra-operative hemorrhage of > 500 mL and no woman required blood transfusion. Two women (4%) developed incisional hernias post-operatively. Two women complained of persistent wound pain on final review. Four women had urinary tract infections post-operatively; one developed recurrent urinary tract infection secondary to obstructed voiding and required intermittent self-catheterisation and prophylactic antibi- otics for four months.

DISCUSSION

Recurrent urinary stress incontinence is a complex and challenging problem. A large variety of procedures including the Burch colposuspension15, needle suspension4,11,13 Marshall-Marchetti-Krantz operation15, the Zacharin procedure2 and sling5–10,12,14–16 have been described in the management of this condition. The sling procedure is the most commonly described surgery for recurrent stress incontinence and a variety of organic materials including rectus sheath7,9, fascia lata8,15, vagi- nal wall111,12, lyophilised dura materlS and round ligament5 have been utilised. A polypropylene (non- absorbable synthetic material) sling6,14 and a bone anchoring sub-urethral sling17 have also been described in the management of recurrent genuine stress incontinence. The success rate of surgery for the management of recurrent stress incontinence range from 66%-100%11–12.

While most authors believe previous continence surgery is a risk factor for the failure of subsequent continence procedures1,2, others have reported outcomes that are not significantly different from primary continence procedure3,4,15. Most reports on recurrent stress incontinence include women who have previously undergone a combination of vaginal and/or retropubic continence surgery without differentiating between the two groups. This makes it difficult to assess whether women who have recurrent stress incontinence following retropubic continence surgery are at greater risk of failure of subsequent continence surgery than women who have undergone only vaginal continence surgery. Enzelsberger et al.15, in a recent prospective randomised trial, compared the Burch colposuspension with the Lyodura sling in the management of women with recurrent stress incontinence. The only continence surgery previously undertaken was the anterior vaginal repair. While the objective success rate of the Burch colposuspension (86%) and the pubovaginal sling (92%) were similar, the incidence of detrusor instability and voiding difficulty was significantly greater following the sling (29%) than the Burch colposuspension (10%). Women who had undergone a previous anterior vaginal repair were not at increased risk of an adverse outcome at sub sequent Burch colposuspension or pubovaginal sling.

Although there is debate on the effect of previous continence surgery on outcome, there is no doubt that repeat retropubic surgery is technically difficult. Retropubic adhesions between the bladder, urethra, paravaginal tissues, and retropubic periostium will increase complications and lower the success rate of repeat retropubic continence surgery. Lee and Symmonds18 are the only authors to report their experience of managing recurrent stress incontinence in women who had all failed to respond to previous retropubic continence surgery. In 1975, they described performing a Mar- shall Marchetti-Krantz operation in women who had all undergone a previous Marshall-Marchetti-Krantz procedure and found “in most cases little or no evidence of adhesion or scarring in the periurethral and vesical neck region”. They suggested most women, at the first procedure, had an ‘alleged’ Marshall-Marchetti-Krantz and not a true suspension of the urethra and bladder neck. In their series, 66% of women had a hypermobile bladder neck pre-operatively, and 83% were subjectively continent post-operatively.

Since this report in 1975, there are no further reports that include only women with recurrent stress incontinence after failed retropubic surgery. In our experience of women undergoing repeat retropubic surgery for recurrent stress incontinence, 73% of the women had dense retropubic fibrosis andor the presence of permanent sutures in the bladder neck region. This dense retropubic fibrosis explains why 62% of the women had a well-supported bladder neck preoperative1 y, in contrast to the 33% reported by Lee and Symmonds. The marked difference in retropubic fibrosis found in our series may be explained by the improved surgical technique, as the retropubic approach to stress incontinence has gained in popularity. Also, the move from absorbable22 to non-absorbable23 sutures in retropubic continence procedures may be an important factor.

We were unable to demonstrate a significant adverse effect of pre-operative limited bladder neck mobility on surgical outcome in this group. “his may be due to the surgical technique allowing safe division of retropubic fibrosis around the bladder neck and resulting in increased mobility of the bladder neck region prior to suture placement. Alternatively, the trend towards poor surgical outcome with decreasing pre-operative bladder neck mobility may have been significant if the sample size was larger. Ramon et al.4. reported subjective cure rates of 80% using needle suspensions in the management of recurrent stress incontinence. Women who had undergone vaginal or retropubic procedures were not differentiated, but 80% of these women had a hypermo- bile bladder neck pre-operatively. They concluded that while a needle suspension was appropriate in women with a hypermobile bladder neck, in women with a well- supported bladder neck and recurrent stress incontinence a sling procedure was indicated.

Breen et al.16 has recently described the fascia lata suburethral sling in the treatment of recurrent incontinence in which the majority of women had previously undergone retropubic continence surgery. Ninety percent of women had no urinary leakage, or less than three episodes per week. Ten of the 60 women (17%) had the sling removed, eight due to voiding difficulties and two due to retropubic abscesses. Three of the 60 women (5%) developed detrusor instability post-operatively, Neither pre-operative bladder neck mobility nor the operative findings in the retropubic space were described.

Morgan has written extensively on the management of recurrent stress incontinence, reporting excellent results with a low complication rate using the polypropylene ling6,14. He reports that “the key to surgi- cal success is the complete release of the base of the bladder and the urethra from a bed of scar, followed by repositioning of the bladder neck in an intra-abdominal retropubic position”. He felt this was safely performed with two surgical teams operating together to completely release the bladder and urethra. We agree that the urethra and bladder needs to be completely dissected free from the posterior surface of the pubic symphysis. The Cherney incision allows greater exposure to the retropubic space to deal with retropubic fibrosis. While the Cherney incision offered excellent retropubic exposure, it may also be responsible for the 4% incidence of incisional hernia following this operation.

The efficacy of the modified Burch colposuspension in women undergoing repeat retropubic continence surgery is encouraging, although long term follow up will be important. How this surgery achieves continence is easily understood in women with poor support to the bladder neck pre-operatively, but is less obvious in those with a well-supported bladder neck. Frequently, once the retropubic fibrosis and adhesions were released, the paravaginal tissue in the region of the bladder neck was often mobile and without any indication of prior retropubic continence surgery. This may indicate inappropriate placement of sutures or slings at prior retropubic continence surgery. Retropubic fibrosis in women with recurrent stress urinary incontinence and a well-supported bladder neck after retropubic continence surgery may act to pull the bladder neck open rather than support and stabilise the area. It is our policy to offer these women with a well-supported bladder neck and adequate anterior vaginal wall mobility a Burch colposuspension. Women with a rigid anterior vaginal wall would be offered a sling or a transurethral bulking agent.

Increasingly we will be asked to manage recurrent urinary stress incontinence after retropubic continence surgery. Marked retropubic fibrosis should be expected. The Burch colposuspension with the assistance of the Cherney incision and sharp dissection of retropubic fibrosis is an effective and safe treatment for women with this condition.

Ancillary