Chapter 29 SELECTING THE BEST SURGICAL OPTION FOR THE TREATMENT OF STRESS URINARY INCONTINENCE

The number of surgical procedures available to treat stress incontinence continues to grow, complicating the process of choosing which procedure should be performed on which patient. The surgeon and patient expect that the procedure chosen will be the one that provides the highest degree of efficacy with the least amount of morbidity. Do certain patients have better outcomes or less morbidity with one surgical treatment or another, and is their evidence to support the decision?

The American Urological Association’s guidelines for the surgical treatment of stress urinary incontinence, published in 1997 by Leach and colleagues,¹ were developed from metaanalyses of the incontinence literature through 1996. The report made recommendations for the surgical treatment of stress urinary incontinence (SUI) in an index patient, defined as a woman with uncomplicated SUI who is undergoing her first incontinence surgery and who has no concomitant morbidities. The guidelines state that for short-term efficacy, any of the four classes of procedures evaluated were acceptable but found that the Burch colposuspension and pubovaginal sling had better long-term outcomes.¹ These guidelines predate the arrival of the mid-urethral sling, the development of alternative allogenic and xenograft sling materials, and the use of laparoscopy in the treatment of incontinence, and they are in the process of being updated.

The guidelines may not be relevant for patients who have any of several complicating factors, including women who have undergone multiple incontinence procedures or who have mixed urinary incontinence symptoms, concomitant prolapse, or a host of other medical comorbidities that may affect the efficacy and safety of a given procedure. To achieve the goal of maximal efficacy with minimal morbidity in nonindex women, we must look beyond the guidelines to other criterion. This chapter looks at a number of nonindex patients and reviews the literature supporting the use of one procedure or another for the treatment of stress incontinence. It discusses the reasons for choosing a procedure for certain patients, even when the evidence is not available. It also presents an algorithm for matching procedures with patients.

CHOOSING AN INCONTINENCE PROCEDURE

Several factors influence the choice of a surgical procedure that is most appropriate for a given patient. These factors include the pathophysiology of the incontinence (e.g., sphincter dysfunction or anatomic incontinence), the presence of mixed urinary incontinence symptoms, a history of incontinence surgery, or the need for concomitant prolapse surgery. Specific medical comorbidities, such as such as the patient’s age, obesity, estrogen status, preexisting voiding dysfunction, history of pelvic irradiation or neurologic disease, may also play a role in the decision. Clinicians have preferences for one procedure over another based on their training and clinical experience. Patients also have certain values or opinions that can factor into which procedure they are willing to undergo. Some patients are risk averse and seek the operation with the longest track record. Others are willing to forgo long-term data for a newer procedure that offers less short-term morbidity.

Evidence based-medicine is founded on the premise that therapeutic decisions for medical and surgical conditions use five levels of scientific evidence.^2,³ Level 1 evidence is derived from meta-analyses of randomized, clinical trials or a single, good-quality randomized, clinical trial. Level 2 evidence is derived from meta-analyses of prospective cohort studies. Level 3 evidence is gained from good-quality case-control studies, level 4 evidence comes from high-quality case series, and level 5 evidence is derived from expert opinion. Although the ideal is to look for the highest level of evidence, this may not exist for some decisions. When Leach and coworkers¹ performed their meta-analysis in 1996, they remarked on the paucity of level 1 evidence in the incontinence literature. Downs and Black’s review⁴ of the incontinence literature also highlighted this problem. Since these reviews, there have been efforts to improve the quality of studies in the incontinence field. A number of randomized, clinical trials have assessed outcomes for various procedures performed for the treatment of SUI.^5,⁶

Unfortunately, it remains difficult to find adequately powered randomized, clinical trials or cohort studies assessing the differences between two procedures in any of the subgroups of incontinence patients who are thought to be at a higher risk for failure or complications. Most data are from analyses of subgroups within larger series. These are usually retrospective analyses or studies that were not designed to look at those subpopulations, and consequently, they are underpowered to identify significant differences. When we try to compare the results from different studies, we are hindered by the lack of standardized definitions and outcomes measures. Recommendations in the literature for how to treat a given high-risk patient population are based on level 4 and 5 evidence consisting of case-control series and expert opinion. The subsequent discussion looks at the evidence in the literature regarding how these factors may influence the choice of procedure performed.

INTRINSIC URETHRAL DYSFUNCTION

Historically, the presence of intrinsic urethral dysfunction (ISD) has been a critical factor in deciding which procedure to perform for the treatment of SUI. The party line has been that patients with ISD should be identified and treated differently from patients without ISD. This was based on a number of studies suggesting that patients with ISD did not do as well with standard bladder neck suspensions as they did with sling procedures. This concept has become at best controversial and at worst irrelevant in the past few years for a number of reasons. First, there is a lack of agreement on the definition of ISD. It has been defined in a variety of ways, including maximal urethral closure pressure (MUCP) less than 20 cm H₂O, Valsalva leak point pressure (VLPP) less than 60 cm H₂O, or an open bladder neck identified on cystourethrography. Attempts to show correlations between these definitions have been unsuccessful, preventing comparison of studies that used different definitions.^7,⁸ Second, the measurement tools used to diagnose ISD have been plagued by poor standardization and poor correlation with physiologic parameters. Third, even when similar definitions are used, there are conflicting results with regard to whether and how ISD relates to outcomes of stress incontinence therapy. Weber^9,¹⁰ critically assessed MUCP and VLPP as diagnostic tests and found that both fall short with regard to standardization, reproducibility, and predictive value for outcomes. Nevertheless, studies reporting on the difference in outcomes between those with or without ISD have had a significant impact on how patients have been treated in over the past 20 to 30 years. It is useful to look at some of this literature to understand how we got from there to where we are today.

Sand and colleagues ¹¹ published a retrospective analysis in 1987 of patients who had undergone Burch colposuspension and found that those who had an MUCP less than 20 cm H₂O had lower success rates than those with higher MUCPs (82% versus 46%).¹¹ At the same time, other investigators were reporting good success rates in women with type III incontinence who had undergone pubovaginal sling procedures.^7,¹² Subsequent reports documented the long-term success rates of the pubovaginal sling in patients regardless of the presence or absence of ISD defined by VLPP. Chaikin and coworkers¹³ reported an overall long-term cure rate of 73% for pubovaginal sling procedures in 251 women with types I, II, or III stress incontinence. Morgan and associates¹⁴ described 247 women who had undergone autologous pubovaginal sling procedures. Continence rates were 91% for type II and 84% for type III incontinence. Hassouna and Ghoniem¹⁵ assessed long-term cure rates with questionnaires in women who had undergone a sling procedure for ISD. They found a 78% cure rate in this population at a mean follow-up of 3 to 4 years. Zaragoza¹⁶ reported 60 consecutive pubovaginal sling procedures in women with type II or type III incontinence and found a 95% continence rate at 2 years. These kind of results, coupled with the belief that there were two distinct types of stress incontinence, led to a widely held view that patients with ISD should undergo pubovaginal sling procedures rather than Burch colposuspension or needle suspension procedures, which should be reserved for women with hypermobility and no ISD.

The data on ISD and its association with outcomes have been less convincing. Culligan and Sand ¹⁷ compared the long-term results for modified Burch and pubovaginal sling procedures in women with a low-pressure urethra (MUCP < 20 cm H₂O). Thirty-six women with urodynamic stress incontinence, a low-pressure urethra, urethral hypermobility, and no significant pelvic organ prolapse were randomly assigned to undergo a modified Burch procedure (n = 19) or sling placement (n = 17). The investigators did not demonstrate any difference between the Burch and pubovaginal sling procedures in subjective cure rates at 3 months (90% versus 100% respectively) or at 6 years of follow-up (93% and 84% respectively). Long-term objective cure rates were also similar for the two groups (84.6% versus 100%). The statistical power of this study was compromised by the unexpectedly high success rates of the Burch colposuspension.¹⁷ Maher and colleagues¹⁸ performed a retrospective review on a subgroup of 45 women (21 Burch and 24 pubovaginal sling procedures) with low preoperative MUCPs from a larger trial comparing outcomes between Burch and pubovaginal sling procedures. Subjective outcomes were based on a visual analogue scale, with patients rating an improvement of 8 or higher on a scale of 1 to 10 considered to be cured and with objective outcome based on the presence of stress incontinence verified by urodynamic study (UDS). The groups were similar in age, body mass index (BMI), parity, presence of hypermobility, and history of previous pelvic surgery. There were no statistically significant differences in outcomes, although there was a trend for higher success rates in the Burch colposuspension group compared with the pubovaginal sling in all categories: subjective (90% versus 71%; P = .12), patient-determined visual analogue score (76% versus 67%; P = .22), and objective success (67% versus 50%; P = .26). Complications were higher in the pubovaginal sling group (5% versus 25%; P = .06).¹⁸ One criticism of this paper is that the success rates for slings are lower than has been reported in most sling series. The investigators have suggested that it may reflect that urge incontinence was included in the definition of failure, whereas other studies did not include it. This underscores the need for similar outcome measures to compare results of studies.

Several studies have looked at the success rates of tension-free vaginal tape (TVT) procedures in patients with a so-called bad urethra or ISD. None of these studies provides more than level 3 or 4 evidence, and none compares TVT procedures with other procedures. Some of the studies suggest women fare worse if they have ISD, whereas other studies report no difference. Rezapour and colleagues ¹⁹ analyzed data from their series of 400 women who had undergone TVT procedures and identified a subgroup of women with MUCPs less than 20 cm H₂O. This group had a lower cure rate (74%) compared with previously reported cure rates for the general population (85% to 90%). When they performed a multivariate analysis of the bad urethra group, they found the worst cure rates in a subpopulation of women who were older, had an immobile urethra, and had low resting MUCPs.¹⁹ Kulseng-Hanssen²⁰ demonstrated more severe postoperative incontinence after TVT procedures in women with MUCPs less than 20 cm H₂O. Paick and coworkers²¹ performed a retrospective analysis of their data from 221 patients who had undergone TVT procedures. They divided the patients into two groups: VLLP more than 60 (n = 160) and VLPP less than 60 (n = 61). At the 6-month follow-up, subjective and objective cure rates were 93% and 82% (P = .013).²¹

Several investigators have suggested that urodynamically defined ISD does not correlate with TVT outcomes. Fritel and associates ²² compared patients with MUCPs less than 20 cm H₂O to those with MUCP more than 20 cm H₂O who had undergone TVT procedures. They reported objective cure rates of 80% and 85%, respectively. Rodriquez and colleagues²³ analyzed their prospective series of distal polypropylene slings and found no differences in outcomes based on urodynamically defined incontinence.

There is no evidence suggesting that there is a difference in outcomes between the transobturator approach to the midurethral sling (TMUS) and the retropubic approach to the mid-urethral sling (RMUS) for patients with ISD. Mellier and associates ²⁴ retrospectively looked at 52 patients with ISD defined by an MUCP less than 30 cm H₂O who had undergone a TMUS procedure (n = 26) or a TVT procedure (n = 26). They did not see any difference in outcomes.²⁴ This study is significantly limited by the nonconcurrence of the procedures and a limited telephone response rate of only 75%.²⁴

Urethral injections of various bulking agents have been used to treat ISD, and the data confirm the effectiveness in this group. Herschorn and coworkers ²⁵ reported long-term follow-up with collagen that did not differ significantly when subanalyzed for urethral function. Not many studies have compared the overall effectiveness directly with that of standard incontinence procedures. Success rates for collagen are generally lower than those obtained by most series of anti-incontinence surgery, but these results may improve if a longer-lasting bulking agent is developed. Given the significant differences in efficacy and safety between collagen and pubovaginal sling procedures, an adequately designed study to compare the two on multiple levels would be a welcome addition to the literature. With the advent of minimally invasive slings, a randomized, clinical trial comparing urethral injection with mid-urethral sling would also be helpful. Until this is done, the bulking agents are an option for women who do not want surgery or are not candidates for a surgical procedure.

In summary, although it makes sense that a poorly functioning urethra causes more severe incontinence and may be more difficult to repair, there is little evidence demonstrating that one procedure is superior to another in treating this population. The tools used to assess urethral function have not been subjected to adequate analysis with regard to their ability to represent changes in urethral function and how that change correlates with clinical symptoms. To answer the question this chapter poses, whether ISD is a variable that should direct our choice of procedures, we need to compare specific procedures in a standardly defined group of patients with ISD in a properly designed randomized trial.

URETHRAL HYPERMOBILITY

The so-called fixed urethra was associated with severe incontinence and poor urethral function even before the urodynamic definitions for ISD were created. Incontinence in the absence of hypermobility was interpreted as an indicator of poor urethral function. There is controversial evidence regarding the relationship between Burch colposuspension outcomes and the presence or absence of hypermobility. However, there is increasing evidence in the TVT literature suggesting that patients with hypermobility have better outcomes after TVT procedures than those with a fixed urethra.^22,²⁶ Fritel found that success of TVT correlated positively with the amount of preoperative hypermobility. Patients were divided into three groups based on hypermobility defined by cystourethrography: more than 60 degrees, 30 to 60 degrees, and less than 30 degrees. Success rates after TVT procedures were 97% for patients with more than 60 degrees of hypermobility, 86% for patients with 30 to 60 degrees of hypermobility, and 70% for those with less than 30 degrees of hypermobility. The lowest success rates were for women with previous surgery and no urethral hypermobility (68%).²² There are two theories about this finding. The first is that the mechanism by which TVT cures SUI is urethral kinking during increases in pelvic pressure that occurs due to a downward movement of the urethra against the fixed mesh of the TVT. The second possible explanation is that a fixed urethra needs external compression to increase the resistance. The tension-free sling, by definition, does not place any compression on the urethra. A comparison of mid-urethral sling approaches is needed to investigate whether this is also true of the TMUS procedures.

Transurethral injections were initially indicated only for patients without evidence of hypermobility. Subsequently, a number of studies have shown them to be equally effective in patients with hypermobility.²⁵

In summary, although there is no level 1 evidence documenting the superiority of pubovaginal sling or Burch procedures over the mid-urethral sling in patients with immobile urethras, there is level 3 and 4 evidence to suggest that the tension-free mid-urethral sling is less effective in patients with an immobile urethra than it is in patients with evidence of hypermobility.

RECURRENT STRESS INCONTINENCE

Patients who have had a previous anti-incontinence procedure have lower success rates and higher morbidity with subsequent procedures. The reasons include a higher incidence of ISD, increased age of the patients and scarring in the retropubic space. Other confounding variables include the type of procedure the patient had previously undergone. A study looking into complications of TVT after Burch colposuspension is not necessarily going to provide information on the safety and efficacy on performing a TVT after a failed TVT. Comparison of results is difficult given the various outcome measures used in different studies.

The only level 1 evidence that compares two procedures for this subgroup of patients is from a prospective, randomized study comparing the Burch colposuspension with the Lyodura sling in the management of women with recurrent urinary stress incontinence and in whom the only previous continence procedure was anterior repair.²⁷ Although the objective success rates for the Burch colposuspension (86%) and the pubovaginal sling (92%) were similar, the incidence of detrusor instability and voiding difficulty was significantly greater after sling procedures (29%) compared with Burch colposuspensions (10%). The metaanalyses by Black and Downs⁴ identified no difference in outcomes between the Burch and sling procedures when performed as secondary procedures. However, they acknowledged that the number of studies was limited and that all had low power to detect significant differences.⁴ Thakar and associates²⁸ reported on a prospective, cross-sectional, observational study of 56 women who had undergone a secondary Burch colposuspension for recurrent stress incontinence. At 4 years, the median subjective cure rate was 71%, and the objective cure rate was 80%. They did not compare these patients with a cohort of women who were undergoing a primary Burch colposuspension.²⁸

Nilsson and Kuuva ²⁹

Only gold members can continue reading. Log In or Register to continue