Stress urinary incontinence arises where the function of the urethral sphincters and or pelvic floor muscles are compromised; leakage of urine can occur if intra-abdominal pressure is raised, even in the absence of a contraction of the detrusor muscle of the bladder wall. Sphincteric or pelvic floor muscle deficiency is seen in patients who have sustained damage to the sacral segments of the spinal cord, the cauda equina or peripheral nerves within the pelvis. The use of indwelling urethral catheters in patients with neurologenic lower urinary tract dysfunction (NLUTD) is a well-recognised risk factor; tension on a urethral catheter can cause pressure necrosis of the urethral sphincters. Patients who lack urethral sensation or who are cognitively-impaired are at particular risk of sustaining catheter-related urethral damage. Stress incontinence in patients who have neurological disease can, of course, have a non-neurological aetiology such as pelvic floor hypermobility that has resulted from previous pregnancies and childbirth.
Many patients with neurogenic stress incontinence who request treatment of the condition are already using intermittent self-catheterisation to empty their bladders. However, for patients who empty the bladder without using a catheter, for example by using abdominal straining, it is important to recognise that effective treatment of stress incontinence will, almost inevitably, precipitate urinary retention which will necessitate the introduction of intermittent catheterisation or the use of an indwelling suprapubic catheter.
Pelvic floor muscle training is widely used in the neurologically intact population with stress incontinence of urine and is supported by evidence of efficacy in the NICE Female Urinary Incontinence Guideline 125 However, the use of pelvic floor muscle training in patients with NLUTD has received relatively little attention. Patients who retain the ability to voluntarily contract their pelvic floor muscles, despite their neurological deficit, are candidates for pelvic floor muscle training programmes as they have the potential to improve the strength and responsiveness of their pelvic floor musculature. A variety of techniques that might improve the outcomes of pelvic floor muscle training have been described; these include electrical stimulation of the pelvic floor and biofeedback systems. However, there is limited information about the treatment regimes that should be used in patients with NLUTD as well as uncertainty about the effectiveness of the treatment.
A wide range of surgical procedures have been used to treat stress incontinence in patients with NLUTD. The most commonly used approaches involve either providing passive support or compression of the urethra using urethral slings or providing active compression using the artificial urinary sphincter. Slings can be made either from the patient’s own tissues (for example autologous rectus sheath) or from synthetic materials.
Sling surgery is well-established in the management of stress incontinence in the non-neuropathic female population 125 but it is not possible to extrapolate from the neurologically intact population to patients with NLUTD as the pathophysiology of stress incontinence differs between the two groups. In the neuropathic population it is usually the case that there is damage to the function of the muscle of the urethral sphincter (intrinsic sphincter deficiency) while in neurologically intact patients, excessive mobility of the urethra is the commonest cause of stress incontinence. This is an important distinction as it is generally believed that successful treatment of neuropathic stress incontinence is dependent on a sling compressing the urethra rather than simply preventing descent of the urethra from its normal anatomical position.
To date, autologous tissue has been regarded as the standard sling material to use when managing neuropathic stress incontinence. The perceived need for some tensioning of the sling raises the question as to whether synthetic material should be used in view of there being a theoretical increased risk of a synthetic sling or tape eroding through the wall of the urethra. Despite the fact that sling surgery has been widely used to manage neuropathic stress incontinence, there is relatively little published data on the benefits and risks of this therapeutic option.
For thirty years, the artificial urinary sphincter has been used to manage stress incontinence in neuropathic patients both alone and in combination with other reconstructive procedures 126. The AMS 800 (American Medical Solutions) is the only device that has been in widespread use. It works on a hydraulic principle. A pressure-generating reservoir supplies fluid to a cuff-mounted balloon which is fitted around the urethra. As the pressures in the two components equalises, the urethral lumen is occluded and continence aided. The patient uses a pump to empty the cuff balloon and transfer fluid back to the reservoir which then allows voiding to take place or a catheter to be passed. Although high continence rates are reported for patients undergoing artificial urinary sphincter implantation, surgical complication rates and the cost of the device have to be taken into consideration when evaluating the implant’s role.
9.1. Pelvic floor treatments
9.1.1. Does pelvic floor muscle training with or without electrical stimulation or biofeedback compared with treatment as usual, improve outcomes?
184.108.40.206. Clinical Evidence
We searched for RCTs comparing the effectiveness of pelvic floor muscle training with or without electrical stimulation or biofeedback as interventions for improving outcomes for incontinence due to multiple sclerosis or stroke. These conditions were selected as they represented the main diagnostic groups in which pelvic floor muscle training is used in neurogenic LUT dysfunction in clinical practice. We looked for any RCT studies that compared the effectiveness of pelvic floor muscle training with or without electrical stimulation or biofeedback with treatment as usual.
Five RCTs were identified comparing pelvic muscle training for improving incontinence due to neurological disease/injury. All of the RCTs were in the adult population. Table 77 summarises the population, intervention, comparison and length of follow up for each of the studies.
Comparison of pelvic floor muscle training versus sham in patients with multiple sclerosis
Comparison of pelvic floor muscle training plus electrical stimulation versus control in patients with multiple sclerosis
Comparison of pelvic floor muscle training (PFMT) plus electromyography feedback (EMG) plus active neuromuscular electrical stimulation (NMES) versus PFMT plus EMG plus placebo NMES in patients with multiple sclerosis
Table 80. GRADE profile - PFMT plus EMG plus active NMES compared with PFMT plus EMG plus placebo NMES in patients with multiple sclerosis - Clinical study characteristics and clinical summary of findings
Comparison of pelvic floor training and advice (PFTA) versus PFTA plus electromyography (EMG) versus PFTA plus EMG plus neuromuscular electrical stimulation (NMES)
Narrative summary (for outcomes that are not appropriate for GRADE due to incomplete outcome reporting)
Kings Health Questionnaire (KHQ)
Throughout the duration of the study, results for the KHQ were variable both within and between groups, however significant improvements were demonstrated in the Symptom Severity Scale in the PFTA + EMG and PFTA + EMG + NMES groups at all time points (p ≤ 0.034)130
Multiple sclerosis quality of life (MSQoL-54)
Throughout the duration of the study, results for the MSQoL-54 were variable both within and between groups, however significant improvements were demonstrated in the cognitive function sub-scale at all time points in PFTA + EMG + NMES (p ≤ 0.016). In addition, statistically significant improvements were also observed in the emotional well-being sub-scale in PFTA + EMG and PFTA + EMG + NMES (week 24; p ≤ 0.027)130
Attendance at the weekly clinic sessions averaged 78% in all groups. Home use of the EMG unit was 75% recommended. No major effects or problems with usage were indicated 130.
Comparison of pelvic floor muscle training versus control in patients with stroke
220.127.116.11. Economic evidence
No relevant economic evaluations comparing pelvic floor muscle training, with or without electrical stimulation or biofeedback were identified.
In the absence of recent UK cost-effectiveness analysis, relevant unit costs are provided in Table 83 to aid consideration of cost effectiveness.
No evidence could be found that suggested that pelvic floor training is cost-effective in neuropathic patients with urological incontinence. The cost of pelvic floor training, with or without electrical stimulation or biofeedback is unlikely to be high, as shown in the unit costs above. While the costs of these programmes are not insignificant, the GDG felt that if effective this cost may be offset by the reduced costs associated with a reduction in the use of incontinence aids.
18.104.22.168. Evidence Statements
Clinical Evidence Statements
Pelvic floor muscle training (PFMT) versus sham in patients with multiple sclerosis
One study comprising 27 participants found that a statistically significant lower proportion of patients receiving pelvic floor muscle training compared to sham had:
- Frequency (12 weeks) (moderate quality)
- Urgency (12 weeks) (low quality)
One study comprising 27 participants found no significant difference between pelvic floor muscle training and sham for:
- Nocturnal enuresis (12 weeks) (low quality)
- Incomplete emptying (12 weeks) (low quality)
- No. of sessions attended (12 weeks) (low quality)
PFMT plus electrical stimulation compared with control in patients with multiple sclerosis
One study comprising 80 participants found that a statistically significant lower proportion of patients receiving pelvic floor muscle training compared with control had:
- Leakage of urine on minimal effort (3 weeks, 6 months) (very low quality)
- Nocturia (3 weeks, 6 months) (very low quality)
PFMT plus electromyography feedback (EMG) plus neuromuscular electrical stimulation (NMES) compared with pelvic floor muscle training plus EMG in patients with multiple sclerosis
One study comprising 74 participants found no significant difference in:
- Post-void residual urine (9 weeks, 24 weeks) (low quality)
- Withdrawals (9 weeks) (low quality)
Evidence statements could not be produced for the following outcomes of the studies by McClurg 134 as the results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect :
- Urogenital Distress Inventory – irritative subscale (week 24) (very low quality)
- Leakage episodes per 24 hr (week 9) (very low quality)
Comparison of pelvic floor training and advice (PFTA) versus PFTA plus electromyography (EMG) versus PFTA plus EMG plus neuromuscular electrical stimulation (NMES)
One study of 30 participants found no significant difference for each group (PFTA, PFTA + EMG, PFTA + EMG + NMES) when comparing pre vs post treatment values for:
- No. of patients incontinent (week 9, 24) (low quality)
Evidence statements could not be produced for the following outcomes of the study by McClurg 130 as the results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect :
- Incontinence Impact Questionnaire total score
- Leakage episodes per 24 hr
- Post-void residual urine
- Kings Health Questionnaire
- Multiple Sclerosis quality of life
Pelvic floor muscle training plus electrical stimulation compared with control in patients with stroke
One study comprising 18 participants found no significant difference for PFMT compared with control for:
- Withdrawals (12 weeks) (low quality).
Evidence statements could not be produced for the following outcomes of the study by Tibaek 132; 131 as the results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect :
- SF36 total score (6 months) (very low quality).
- Incontinence Impact Questionnaire (6 months) (very low quality).
- No. of incontinence episodes/24 hr (12 weeks) (very low quality)
Economic evidence statements
While the costs of these programmes are not insignificant, the GDG felt that if effective this cost may be offset by the reduced costs associated with a reduction in the use of incontinence aids
9.1.2. Recommendations and links to evidence
9.2. Urethral tape and sling surgery
9.2.1. What is the safety and efficacy of urethral tape and sling surgery compared with a) bladder neck closure b) usual care in neurological disease?
22.214.171.124. Clinical review
We searched for RCTs and observational studies comparing the effectiveness of urethral tape and sling surgery as interventions for improving outcomes in patients with incontinence due to NLUTD. We looked for any RCT or observational studies that compared the effectiveness of urethral tape and sling surgery with bladder neck closure or with treatment as usual.
24 observational studies were identified. Four observational studies were on tape or synthetic sling surgery 135 136 137 138 and 20 were on sling surgery 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158. No studies were identified comparing urethral tape or sling surgery with bladder neck closure. The evidence is presented according to whether the population was adults or adults and children (mixed) or children. The evidence is also presented according to whether augmentation cystoplasty was performed (either before or at the same time as the tape or sling surgery) or not.
Quality of included studies
Overall, the studies were observational studies of very low quality. The vast majority of studies were before and after studies. Most importantly, there were no attempts to eliminate threats to internal validity through the use of a matched comparison group, although in most studies patients had failed to respond to a period of antimuscarinic medication and intermittent self catheterisation, and so it is unlikely that confounding time effects could wholly explain the changes seen from before to after surgery.
Quality of studies
Tapes and slings
Definitions of incontinence were almost always lacking, and so it is unclear what level of severity was used as the threshold measure of “incontinence”. Several studies also failed to clarify the number of patients suffering from incontinence pre-operatively, although in most cases it was implicit that the majority were suffering from incontinence at baseline.
Quality of life outcome
The study reported the ‘no. of patients’ satisfied and did not use a validated measure of quality of life.
126.96.36.199. Economic evidence
No relevant economic evaluations comparing urethral tape and sling surgery with bladder neck closure were identified.
In the absence of recent UK cost-effectiveness analysis, relevant unit costs are provided in Table 94 to aid consideration of cost-effectiveness.
These interventions (bladder neck open procedures) are categorized in the same Healthcare Resource Group (HRG) codes, costing £1,419 for females (LB23Z, NHS reference cost 2009–10), and £4,617 for males (LB21Z, NHS reference cost 2009–10). However, the unit cost of a tape is not included in the cost of the operation, therefore an additional cost of £632 should be added to the operation cost.
These interventions are currently performed in selective patients and as the difference in costs is negligible, the recommendation should be based on clinical grounds. The cost of these interventions is fairly high, however if shown to be effective, the costs could be offset by a reduction in the costs of incontinence aids and an increase in quality of life. The sling operation is marginally cheaper than the cost of synthetic tape surgery, as the sling is made of human tissue and therefore no extra cost is incurred. However, with no useful effectiveness data it is difficult to draw conclusions on the cost effectiveness of either intervention. If the effectiveness of the two operations is considered equivalent, then autologous sling surgery is likely to be cost saving.
188.8.131.52. Evidence Statements
Synthetic Tapes and Slings
Four observational studies (two of the same population) (27.1 mths to 10 yrs) (very low quality) suggested that synthetic tapes and slings were associated with an improvement in continence. Adverse events include reoperation, difficulties caused by catheterisation and urinary tract infections.
Autologous and Biological Slings
Adults with augmentation
Two observational studies (14.25 to 34.3 mths) (very low quality) reported an improvement in continence. Adverse events included Marlex erosions, urinary tract infections and re-operation.
Children with augmentation
Five observational studies (10.3 to 15 mths) (very low quality) reported an improvement in continence. Adverse events included erectile dysfunction and difficulties with catheterisation.
Children without augmentation/unknown
Six observational studies (10 to 36 mths) (very low quality) reported an improvement in continence. Adverse events included vesicovaginal fistula, re-operation and difficulties with catheterisation
Mixed/unknown with augmentation cystoplasty
Seven observational studies reported an improvement with continence (12 mths to 5 yrs) (very low quality). Adverse events included bladder neck occlusion, asymptomatic bacteriuria and difficulties with catheterisation.
Mixed/unknown without augmentation/unknown
One observational study (39 mths) (very low quality) reported an improvement in continence.
Economic evidence statements
- Autologous and biological sling surgery is marginally cheaper than synthetic tape and sling surgery, however as the two operations are already done on highly selected populations, the direct comparison is not necessarily relevant.
- If the effectiveness of the two operations are considered equivalent, then autologous sling surgery is likely to be cost saving compared to synthetic taping.
9.2.2. Recommendations and links to evidence
9.3. Artificial urinary sphincter
9.3.1. What is the safety and efficacy of artificial urinary sphincters compared with other treatments in neurological disease?
184.108.40.206. Clinical evidence
We searched for observational studies that examined the effectiveness of implantation of an artificial urinary sphincter in improving incontinence in people with NLUTD. We looked for any observational studies that compared the effectiveness of implantation of an artificial urinary sphincter with other treatments, but only one study 159 used a comparison group. All studies compared findings before implantation with those after implantation, and these results form the main body of this report.
25 observational studies were identified which evaluated the effects of artificial sphincter implantation on incontinence in neurological disease. The implantation procedures and prostheses used were fairly homogenous across the studies. The only outcomes addressed by these studies were incontinence and adverse events. Table 95 summarises the population characteristics in each study.
Quality of studies
The confidence in the findings reported below is undermined by the lack of attempts to eliminate threats to internal validity through the use of a matched group. Definitions of incontinence were often lacking, and so it is unclear what was the threshold measure. Several studies also failed to clarify the number of patients suffering from incontinence pre-operatively, although in most cases it was implicit that the large majority were suffering from incontinence at baseline.
Pre-test to post-test comparisons
All studies suggested that artificial sphincters would reduce the likelihood of incontinence, through a comparison of incontinence before and after the sphincter implantation, and this did not appear to vary by age group. Table 96 shows results for these studies.
Comparison to other treatments
Sidi 1987159 compared a group of 16 young people of mean age 17.2 years (range 5–44) who had undergone an artificial sphincter implantation to a group of 9 young people of mean age 16.3 years (range 4–27) who had undergone a Young-Dees-Leadbetter bladder neck reconstruction. The mean (sd) postoperative continence score (0–5 range, and a higher score is better) in the artificial sphincter group was 3.62 (0.7) and in the bladder neck reconstruction group was 3.72 (0.9). This difference was not statistically significant (see Appendix B). This was not a randomised trial, and these results are therefore prone to bias from factors such as varying diagnoses.
Adverse events (post implantation)
A variety of adverse effects of the implantation were reported, and the most important ones are documented in the tables below, with the following data concerning patients affected at least once. The most prominent risks are the need for revision (34%), device failure (26%), the need for complete removal (22%), bladder neck erosion or device infection (11%), UTIs (9%) and upper tract complications (8%). There appears to be no risk for mortality. Note that the lack of reporting of an adverse event does not necessarily imply the adverse event was absent, as some events may only be detected if actively sought. It was not possible to make a meaningful comparison of the incidence in adults and children, as only two studies had exclusively adult patients, most data being from studies with mixed-age groups. Table 97 outlines these results, with a summary for each age group and overall. The only study Sidi 1987159 to compare artificial sphincter implantation to another treatment did not report adverse events.
220.127.116.11. Economic evidence
No relevant economic evaluations comparing the use of the artificial urinary sphincter with sling surgery or other treatments were identified.
In the absence of recent UK cost-effectiveness analysis, relevant unit costs are provided in Table 98 to aid consideration of cost-effectiveness.
The comparison relevant for this question is artificial urinary sphincter versus sling surgery. These interventions are categorized in the same HRG code, costing £4,617 for men and £1,419 for women (LB23Z and LB21Z, NHS reference cost 2009–10). There is however the additional cost to consider of the artificial sphincter itself, which costs £4,119 (Personal communication Mid Yorkshire Hospitals NHS Trust). In current practice, the choice of procedure is made on the basis of surgeon preference.
The clinical review has not shown any significant clinical advantage over other interventions with similar aims. Added to which the clinical review has shown that the rate of re-operation and adverse effects associated with artificial sphincter implantation is quite high, adding to the long term costs and reductions in quality of life. It is estimated by the GDG and on the basis of data from several studies that the re-operation rate on artificial sphincters is one re-operation/replacement of the device every ten years183,183. This means that there are additional life-time costs for many patients who have artificial urinary sphincters implanted as a result of the need to manage complications and undertake revisional surgery.
Therefore the cost of alternative surgical options, such as sling surgery, are likely to be considerably lower as a result of there being a requirement for fewer re-operations. This will also be likely to have a quality of life advantage. On this basis, the artificial sphincter is not considered to be cost-effective when compared to sling surgery in circumstances where the procedures would be expected to have similar success rates in treating stress incontinence.
18.104.22.168. Evidence Statements
Clinical evidence statements
22 Observational studies comprising 695 participants suggested that artificial sphincter implantation might improve incontinence (3 – 208 months) (very low quality)
24 Observational studies comprising 835 participants suggested that the main adverse effects of artificial sphincter implantation are device failure (26%), device infection/bladder neck erosion (12%), the need for revision (34%), the need for complete removal (22%), 11UTIs (9%), and upper tract complications (8%). There appears to be no risk for mortality (3–208 months) (very low quality).
Economic evidence statements
The cost of artificial urinary sphincter is considerably high when the cost of the sphincter, surgery and revisional surgery is considered. This does not support the use of artificial sphincters where another option is available that is judged to be of equal efficacy such as autologous sling surgery. However where no other option is available, the gains in quality of life combined with a partially offset cost of incontinence aids may make this intervention cost effective compared to other treatments.
9.3.2. Recommendations and links to evidence
Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.
The rights of National Clinical Guideline Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.
Royal College of Physicians (UK), London
National Clinical Guideline Centre (UK). Urinary Incontinence in Neurological Disease: Management of Lower Urinary Tract Dysfunction in Neurological Disease. London: Royal College of Physicians (UK); 2012 Aug. (NICE Clinical Guidelines, No. 148.) 9, Treatment for stress incontinence.