According to widely accepted definitions, stress urinary incontinence is the complaint of involuntary loss of urine on effort or physical exertion (e.g. sporting activities), or during sneezing or coughing.1 It appears to be the most common form of urinary incontinence in women and is often associated with urinary urgency and frequency, giving rise to another subform of involuntary urine loss: mixed incontinence. Prevalence estimates in population-based studies range from 9% to 45% depending on definitions of severity and frequency of incontinence episodes.2 Risk factors include pregnancy and (vaginal) delivery, an increased body mass index and diabetes. Notably, and quite contrary to previously widely held opinions, lower postmenopausal hormone levels do not appear to be a risk factor, but hormone replacement therapy seems to increase the risk of becoming incontinent.
It has been consistently shown in epidemiological studies that all types of urinary (and also faecal) incontinence significantly increase with age.2 While the loss of control of the continence mechanism rarely results in danger for the afflicted, it is often associated with a significant impairment of quality of life and leads to retraction and social isolation, especially in women with mixed incontinence.3
Numerous treatment modalities exist, and surgery has played a major role from the beginning of the twentieth century onward. The time-honoured anterior repair appears as one of the first surgical procedures and is still around today but considered as a prolapse operation. Open abdominal retropubic suspension and, much later, laparoscopic abdominal retropubic suspension (Burch colposuspension and Marshall–Marchetti–Krantz colposuspension) were developed in the 1950s to become the standard procedures until the late 1990s. Suburethral fascial slings, such as the one described by Goebel in 1910, are reserved for difficult abdominal–vaginal operations, mostly recurrent incontinence.4 Parallel to the development of abdominal colposuspensions, Pereyra pioneered the concept of needle suspension, which had the advantage of a primarily vaginal approach while administering a minimally invasive needle technique to anchor the suspension stitches abdominally.5 In selected cases, periurethral injections with bulking agents that help to adapt the urethral mucosa and the implantation of artificial sphincters can be used to treat stress urinary incontinence. In 1996, Ulmsten et al.6 published a case series on a sling operation for patients with stress urinary incontinence with several new aspects: (1) the sling was synthetic and non-absorbable; (2) the method of implantation was minimally invasive and (3) could be performed under local anaesthetic; and (4) the sling was placed under the midurethra (rather than at the bladder neck – the target of most previous incontinence operations) and (5) positioned loosely, causing minimal obstruction postoperatively, a problem faced widely with most other incontinence procedures.6 The ‘tension-free vaginal tape’ (TVT) operation, as it was later named, has subsequently been tested in numerous randomized trials and advanced to become the standard surgical treatment, possibly worldwide, for stress urinary incontinence.
Pelvic floor muscle training
In 1948, gynaecologist Arnold Kegel7 popularized the concept of exercising the pubococcygeus muscle to treat symptoms of uterovaginal prolapse and urinary incontinence.7 He understood the problem of the hidden muscles whose action could not be appreciated without vaginal palpation and, thus, cleverly developed a biofeedback instrument, the Kegel perineometer, to facilitate understanding of the muscular action and training. Since Kegel’s days, pelvic floor muscle training (PFMT) has been refined and studied clinically as, in comparison with surgery and pharmacotherapy, it offers the advantage of being almost completely free of side-effects and complications.
There are three major systematic reviews available via the Cochrane collaboration dealing with the clinical utility of PFMT. Dumoulin and Hay-Smith systematically review the evidence on PFMT versus no treatment or inactive control treatment and clearly establish that, based on the literature, PFMT has a real effect on patient-reported cure and improvement, as well as quality of life.8 Another Cochrane review addresses different approaches to PFMT and poses questions such as how often the training should be carried out or how often there should be contacts with the health provider through the training period; however, to date, it cannot be clearly established which approach would be better than another.9 Patients often receive verbal instructions only when seeing the physiotherapist during PFMT sessions. As a result, both patient and therapist are often left in doubt regarding whether or not the right muscles have been activated in the correct fashion. It has been hypothesized that, in those cases, the result of the treatment may not be optimal, and this is a problem that Herderschee et al.10 focused on in a separate systematic review. The authors found that PFMT under the guidance of vaginal (or rectal) palpation of the pubococcygeus muscle, or by using electronic or pressure-triggered biofeedback devices, provides better results than under verbal instruction only.10
Synthetic midurethral sling surgery
The TVT operation in the late 1990s became a major economic success. The manufacturer, Gynecare Inc., (Somerville, NJ, USA), a subsidiary of the Johnson & Johnson Company (Somerville, NJ, USA), saw a steep rise in sales and became a major player in the field of surgical urology and urogynaecology. Soon other contenders, ‘me-too’ products, cropped up and, in 2001, an alternative approach to the retropubic placement of the sling was developed and the transobturator technique became another possibility for implantation. Suddenly, there were many different options – retropubic bottom-up, retropubic top-down, transobturator outside-in and transobturator inside-out – but the community of incontinence surgeons had learned an important lesson from the landmark systematic review paper by Black and Downs in 1996,11 which had concluded that really there was no evidence to support incontinence operations at all.
It is now with regularity that a new randomized study will do the tedious homework of comparing new surgical methods with the old and one alternative with the other. The outcomes of these various sling techniques were eventually summarized in a Cochrane review.12 We now know that (1) synthetic midurethral slings (SMUS) are as effective as traditional (fascial) slings but with shorter operating times and a lower risk of postoperative voiding difficulties and de novo urge symptoms; (2) SMUS are as effective as open retropubic colposuspension with fewer perioperative complications, less postoperative voiding dysfunction, shorter operating times and hospital stay but significantly more bladder perforations; (3) the results comparing SMUS to laparoscopic colposuspension are less clear but also point to a more favourable outcome for SMUS; (4) the retropubic bottom-up approach is more effective than the top-down technique and is also associated with fewer complications; (5) monofilament tapes have higher objective cure rates and fewer tape erosions than multifilament tapes; and (6) retropubic SMUS have higher objective cure rates than, and equal subjective cure rates as, transobturator SMUS; however, voiding dysfunction, blood loss, bladder perforation and operating time are also reduced with the transobturator approach.12 We can now easily conclude that there is more evidence supporting the use of SMUS surgery than any other operative technique for incontinence management.
Which treatment comes first?
In the absence of any head-to-head comparisons between conservative (PFMT) and surgical (SMUS) treatment, the traditional approach to the management of a patient complaining of stress urinary incontinence has been to provide conservative treatment first and, if that fails, riskier options in terms of side-effects and complications come into play. Will a newly published Dutch randomized trial by Labrie et al.13 change any of this?
In this pragmatic study, patients with stress urinary incontinence or mixed incontinence with predominant stress symptoms were screened and 460 of them randomized to either a strict protocol of PFMT, with biofeedback provided by specially certified physiotherapists, or a SMUS arm (retropubic or transobturator). Participants were allowed to cross over to the other treatment arm if they wished and 49% in the PFMT group did so after a mean of 31 weeks of training, while 11% of patients in the surgical group crossed over to physiotherapy after the operation for various reasons (e.g. voiding difficulties). After 12 months, only 14% and 9% of participants in the PFMT and SMUS arm, respectively, were lost to follow-up. An intention-to-treat analysis (in which all participants were analysed according to their original allocation regardless of crossover) revealed that subjective improvement was seen in 90.2% and 64.4% of patients in the SMUS and PFMT arms, respectively [absolute difference 26.4%, 95% confidence interval (CI) 18.1–34.5%, number needed to treat (NNT) 4]. A subjective cure (as evaluated by a negative answer to the question, ‘do you experience urine leakage related to physical activity, coughing, or sneezing?’) was found in 85.2% of patients in the SMUS arm versus 53.4% of patients in the PFMT arm (absolute difference 31.8%, 95% CI 22.6–40.3%, NNT 3). An objective cure, as substantiated by a negative cough stress test, was found in 76.5% of patients in the SMUS arm compared with 58.8% of patients in the PFMT arm (absolute difference 17.8, 95% CI 7.9–27.3, NNT 6). Of course, it is always tempting to analyse the data ‘per protocol’, i.e. by what kind of treatment the participants actually received, taking into account whether or not they crossed over; however, this would result in bias. Post hoc per-protocol analysis showed a subjective improvement in patients who crossed over to the SMUS arm, in those who were originally allocated to the SMUS arm and in those who received PFMT only, at a level of 87%, 85.2% and 15.9%, respectively (P < 0.001 for PFMT vs. both SMUS arms). There was a similar pattern regarding an objective cure: an improvement was noted in patients who crossed over to the SMUS arm, in those who were originally allocated to the SMUS arm and in those who received PFMT only, at a level of 71.8%, 76.5% and 44%, respectively (P < 0.001 for PFMT vs. both SMUS groups). Unsurprisingly, none of the participants experienced any side-effects or complications during physiotherapy and the surgical patients experienced complications at a moderate level, which is expected with SMUS surgery (there were a total of six bladder and 10 vaginal perforations, seven reoperations, 20 postoperative haematomas and 18 patients with de novo urge symptoms in 314 operated patients).13
Will this study be a game changer? Will the providers concerned with the management of stress-incontinent women now send them to primary surgery with confidence and without wasting time in PFMT classes? The answer is almost certainly no, for various reasons. The first reason seems rather crude and subjective but it is true nevertheless: old habits die hard. It will take some time to digest the results of this trial and it may be better to wait for another study replicating these findings before jumping to conclusions. The next issue is the benignity of the conservative treatment. If a medical intervention has virtually zero risk, the trade-off becomes enormously high, even if the effectiveness is not really that good. Likewise, in comparison with PFMT, even the very effective SMUS operation can easily become a second choice given the risks of complications, however small. Of course, the economic aspects have to be weighed up, but I am quite confident that the Dutch group will eventually come up with some cost-effectiveness data that we can ponder further. But who else will ultimately make the choice between conservative and operative other than the incontinent women? It remains our duty to counsel our patients and facilitate their decision-making based on the current best evidence. This trial certainly adds to our knowledge; therefore, let us learn to integrate it with our own experience and the wishes of our patients.