8Treatment to improve bladder storage

Publication Details

Dysfunction of the urinary bladder during the storage phase of the micturition cycle can take the form of either involuntary contractions of the bladder (neurogenic detrusor overactivity), or a loss of receptive relaxation of the bladder wall leading to a progressive increase in pressure as the bladder fills (reduced bladder compliance).

Both neurogenic detrusor overactivity and impaired bladder compliance can lead to symptoms, such as increased urinary frequency, urinary urgency and incontinence. In both conditions deterioration in renal function may occur due to an inability of the upper urinary tract to expel urine in the face of high pressures within the bladder. Patients may be deemed to be at high risk of renal deterioration either because their neurological condition is known to carry a high risk or as a result of the findings of urodynamic investigations. Conditions that are associated with a high risk of renal deterioration include spinal cord injury and spinal dysraphism while adverse urodynamic features include impaired bladder compliance and neurogenic detrusor overactivity in the face of an uncoordinated urethral sphincter (detrusor sphincter dyssynergia).

Incontinence and urinary frequency in patients with neurological disease also occur in the context of cognitive impairment as a result of difficulties with the interpretation of urinary tract sensations and a loss of the appreciation of the social context of micturition.

There are a number of treatment options available that seek to improve continence through improving the ability of the bladder to store urine. These include behavioural, drug and surgical treatments.

  • Behavioural Treatments to improve bladder storage
    Behavioural treatments encompass a range of approaches that seek to train or re-train the neurological processes that control micturition in a way which promotes urine storage. For example, a patient might be prompted to empty the bladder at regular intervals in order to pre-empt episodes of urinary incontinence. Behavioural approaches in those with neurological disease are used for people with significant cognitive impairments such as dementia, often in the care home or hospital environment and also may be used in the early stages after acute neurological injury or illness as a means of re-establishing continence as the micturition cycle recovers. The treatment does not necessarily aim to alter the neural control of micturition, rather to manage toileting regimes to promote continence.
  • Types of Behavioural Treatments
    Timed voiding – consists of taking the patient to the toilet at set time intervals, for example every 2 hours.
    Prompted voiding – this is used to encourage people to initiate their own toileting. It usually involves positive reinforcement. It involves the use of a carer to take the person with incontinence to the toilet, and so involves education of both the person with incontinence and their carer
    Habit re-training – involves working out an individual’s toileting pattern and then developing a personalised toileting schedule to prevent involuntary voiding.
    Behavioural treatments are not fully standardised, which hampers evaluation of their effectiveness. However, such evaluation is important as these treatments are widely used and can involve considerable use of resources in the form of staff time.
  • Drug Treatments to improve bladder storage
    Acetylcholine is the neurotransmitter which has the primary role in stimulating contraction of the urinary bladder. The detrusor muscle of the bladder wall is rich in muscarinic receptors which, when activated by acetylcholine, trigger bladder contraction. Antimuscarinic drugs are muscarinic receptor antagonists and have the potential to reduce or abolish bladder contractile activity. They have long been established as the first line treatment for detrusor overactivity and symptoms of an overactive bladder. Antimuscarinic drugs may also have effects on bladder sensory mechanisms as muscarinic receptors are also found in the sub-epithelial neural plexus of the bladder 36. The majority of these compounds are administered orally, although some intravesical antimuscarinic preparations have been developed. Early forms of antimuscarinics had a number of troublesome side effects, which newer compounds have sought to ameliorate. Antimuscarinics drugs were formerly known as “anti-cholinergics”.
    Antimuscarinic drugs have been used for many years to treat patients with neurogenic detrusor overactivity although the response of an individual patient to antimuscarinic treatment is variable. There are also important outstanding questions about the ability of antimuscarinic drugs to protect the upper urinary tract in the face of a high pressure, overactive bladder.
    There are seven different types of botulinum toxin (A-G) but it is botulinum toxin type A which has become widely used in clinical practice, although Botulinum toxin type B has also been the subject of clinical trials. Botulinum toxin type A (BTX) acts by blocking the release of acetylcholine and other neurotransmitters from nerve terminals. Injection of the drug into the detrusor muscle using an endoscopic technique was described in Schurch et al in 200037 since then the use of BTX for treating neurogenic detrusor overactivity has become widespread. However, a number of questions have yet to be definitively answered so that the duration and adequacy of the response to the treatment in different patient groups has not been fully elucidated. It is also unclear whether or not the drug is sufficiently effective to prevent the development of hydronephrosis in the patient with high pressure urine storage due to either neurogenic detrusor overactivity or reduced bladder compliance. Finally, the cost of the drug and the requirement for injection via a cystoscope mean that the treatment is associated with significant costs which have to be balanced against clinical benefit; there is a lack of published data looking at economic issues in relation to BTX therapy.
  • Surgical Treatments to improve bladder storage
    In cases where the functional capacity of the bladder is severely compromised and where drug therapies have proved ineffective, augmentation cystoplasty can be considered as a means of increasing bladder capacity and maintaining low storage pressures. Augmentation cystoplasty is a surgical procedure which involves opening the abdomen and exposing the bladder. The bladder is opened widely and a patch, made out of an isolated and de-tubularised length of intestine, is sewn into the defect in the bladder wall thereby increasing the capacity of the organ.
    The principle of auto augmentation involves denuding (but not breaching) the urotheial lining of the bladder, in what is effectively an excision of detrusor muscle. This has sometimes been described with the adjunct of overlaying omentum or of a demucosalised intestinal patch in order to support the exposed bladder mucosa.
    Augmenting a bladder usually impairs its intrinsic ability to empty to completion, and recourse to intermittent catheterisation is usually expected. This can be per urethra or via a continent, catheterisable abdominal conduit. This type of conduit consists of a narrow tube (the appendix is often used as the conduit) one end of which is anastomosed to the bladder while the other end is brought to the skin surface to form a small stoma. The bladder can be drained by passing a catheter through the conduit into the bladder. Urine is prevented from refluxing into the conduit, and leaking onto the skin surface, by creating a flap valve at the site of the anastomosis of the conduit into the bladder. Continent, catheterisable abdominal conduits are often called Mitrofanoff conduits, after the surgeon who helped to establish the principles of the surgical procedure.
    Augmentation cystoplasty has been in routine use for treating selected patients with NLUTD for over two decades 38 but is known to be associated with significant morbidity. It is therefore important that the use of augmentation cystoplasty in patients with NLUTD is accompanied by careful consideration of the risks and benefits of the operation. The evaluation of the cost-effectiveness of augmentation cystoplasty has received little attention to date.

8.1. Behavioural treatments

8.1.1. Do behavioural management programmes (timed voiding, voiding on request, prompted voiding, bladder retraining, habit retraining, urotherapy) compared with a) each other b) usual care, improve outcomes?

8.1.1.1. Clinical evidence

We searched for RCTs and systematic reviews comparing the effectiveness of behavioural management programmes for improving the outcomes of incontinence in patients with neurological disease or injury. We looked for any RCT studies that compared the effectiveness of one or more type of behavioural management programme with another behavioural management programme, or treatment as usual.

No RCTs or systematic reviews were found concerning behavioural therapy for incontinence in neurological disorders. However, two Cochrane systematic reviews and one RCT (which was not included in the Cochrane reviews) which were focussed on behavioural therapy for elderly adults with incontinence were found. It is possible that elderly people might respond differently to behavioural treatment, compared to patients with neurological disorders, because of a different aetiology of incontinence and differing levels of mobility. However, it was felt that in the absence of direct findings, the findings for elderly people might have some relevance, and that the findings could be downgraded for indirectness to account for the differing populations, according to GRADE guidelines. These three studies are summarised in table 13.

Table 13. Characteristics of the included studies.

Table 13

Characteristics of the included studies.

The two identified systematic reviews and single RCT assessed the behavioural interventions of prompted voiding, habit retraining and training mobility and toileting skills (see Table 13 for details of these interventions). The first two behavioural interventions were the only practices contained in the protocol for which we found evidence. Training mobility and toileting skills was also included as a behavioural intervention as the GDG felt it potentially relevant.

The outcomes for prompted voiding which the GDG agreed were closely related to the proposed outcomes listed in 1.2:

  • Numbers with no improvement of wet episodes
  • Proportion of hourly checks that were wet
  • Reduction in the mean proportion of hourly checks
  • Incontinent episodes in 24 hours
  • Self initiated toileting

The outcomes for habit retraining which the GDG agreed were closely related to the proposed outcomes listed in 1.2:

  • Incontinent episodes in 24 hours
  • Voided volume and incontinent volume
  • Prevalence of bacteriuria
  • Prevalence of skin rash*
  • Prevalence of skin breakdown*
  • Impact on caregivers*

The outcomes for Training Mobility and Toileting skills which the GDG agreed were closely related to the proposed outcomes listed in 1.2:

  • Average weight of pads over 24 hours*
  • Micturitions on toilet compared to total micturitions*
  • Change from dependent to independent toileting*
  • Change from independent to dependent toileting*

Most of the outcomes in the 12 RCTs included in the two Cochrane reviews were analysed in GRADE using the data and study quality information provided by the reviews. Separate GRADE profiles were created for the prompted voiding and habit retraining interventions. Those outcomes marked with an asterix (*) were not appropriate for meta-analysis or GRADE, and are described in a narrative account in the appropriate section.

Comparison of prompted voiding to no prompted voiding
Narrative summary (for outcomes that are not appropriate for GRADE due to insufficient information given, such as a lack of variance data, or the presentation of numbers of episodes rather than cases)

Reporting of the outcome of proportion of hourly checks that were wet was not reported adequately to allow meta-analysis in 4 RCTs (Ouslander 2005, Schnelle 1983, Smith 1992, Surdy 1992) 39, as they lacked measures of variance and some used medians. These studies all found that the median or mean number of hourly checks that were wet were numerically greater in the control group, weakly suggesting a beneficial effect of prompted voiding (table 15). No statistical analysis was performed, but it can be seen that the probability of all 4 studies showing this trend by chance alone is only 6.25% (50% raised to the fourth power).

Table 15. Mean or median proportion of hourly checks that were wet.

Table 15

Mean or median proportion of hourly checks that were wet.

Incontinent episodes in 24 hours were reduced by 60% – 80% in the intervention group compared to 20–37% in the control group (Engberg 2002, Smith 1992) 39. Linn (1995) 39 noted that treatment group incontinence reduced from 42% at baseline to 17% after treatment (Table 16). These results were incomplete and so could not be meta-analysed.

Table 16. Incontinent episodes in 24 hours – changes during the course of the study.

Table 16

Incontinent episodes in 24 hours – changes during the course of the study.

Self initiated toileting increased in the intervention group more than the control group in 3 studies (Scnelle 1983, Engberg 2002, Linn 1995) 39 and was greater in the intervention group for the final four weeks in one study (Hu 1989) 39, but these data did not include standard deviations (Table 17).

Table 17. Self initiated toileting – changes during the course of the study.

Table 17

Self initiated toileting – changes during the course of the study.

Comparison of habit retraining plus another treatment to usual care
Narrative summary (for outcomes that are not appropriate for GRADE due to insufficient information given, such as a lack of variance data, or the presentation of numbers of episodes rather than cases)

The following outcomes were not presented in a form that was appropriate for meta-analysis.

Number of incontinent episodes

Colling 199240 showed a significant reduction in the number of episodes of urinary incontinence during the treatment period in the treatment group.

Prevalence of skin rash

Colling 200340 reported a significant decrease in skin rash prevalence from 17.7% at baseline to 9.4% at the end of the intervention period. No data are provided for the usual care group, other than the information that a non-significant increase occurred.

Prevalence of skin breakdown

Colling 200340 reported a significant decrease in skin breakdown prevalence from 11.6% at baseline to 2.3% at the end of the study period in the intervention group. In the control group two patients had skin breakdown at baseline and none at the end of the study period. The prevalence figures for the intervention group appear to be counts of the episodes of skin breakdown rather than counts of participants having at least one episode, as 11.6% of the group size of 32 and 2.3% of the control group size of 24 yield non-whole numbers (3.7 and 0.6 respectively). Thus they cannot be analysed with a meta-analysis.

Impact on caregivers

Colling 200340 reported that caregivers found management of incontinence less stressful at the end of the intervention. A greater number of carers felt more prepared to care for their patient’s incontinence needs than at baseline. No statistically significant changes were reported.

Comparison of training mobility and toileting skills to no treatment in achievement of Independent toileting
Outcome data to which GRADE cannot be applied

No outcomes were appropriate for GRADE.

Narrative summary (for outcome data to which GRADE cannot be applied due to incomplete outcome reporting, for example means and standard deviations, or equivalent, were unavailable).

Average weight of pads over 24 hours

The intervention group had a trend (p=0.07) for an 8% lower weight of pads over 24 hours compared to the comparison group. No further data were given in the paper 41.

Micturitions on toilet compared to total micturitions

The intervention had no significant effect on the number or percentage of micturitions on the toilet. No data were given in the paper 41.

Change from dependent to independent toileting

In the intervention group 6 changed from dependent to independent, compared to 2 in the comparison group (p=0.14). The lack of data on the number who were initially dependent in each group makes this data inappropriate for GRADE 41.

Change from independent to dependent toileting

In the intervention group 4 changed from independent to dependent, compared to 3 in the comparison group (p=0.70). The lack of data on the number who were initially independent in each group makes this data inappropriate for GRADE 41.

8.1.1.2. Economic evidence

No relevant economic evaluations comparing behavioural management programmes with each other or with usual care were identified.

Unit costs

In the absence of recent UK cost-effectiveness analysis, relevant unit costs are provided in table 19 to aid consideration of cost-effectiveness.

Table 19. Unit Costs.

Table 19

Unit Costs.

Economic considerations

No evidence could be found that suggested that behavioural management programmes are cost-effective in neuropathic patients with urological incontinence. The cost of behavioural management advice and programmes is unlikely to be high, as shown in the unit costs above. While the costs of these programmes are not negligible, the GDG felt that, if effective, their cost may be offset by the cost savings associated with a reduction in the use of incontinence aids.

Other NICE guidance, Urinary Incontinence (CG40) 2006, and Lower Urinary Tract Symptoms CG97 2010, recommend behavioural management programmes where cases of incontinence are mild and where conservative management is likely to lead to an improvement in continence.

8.1.1.3. Evidence Statements

Clinical Evidence Statement
Comparison between prompted voiding and no prompted voiding
  • One study comprising 133 participants found that that there was no significant difference between prompted voiding and no prompted voiding for the proportion of people with no improvement in wet episodes (22 weeks)(very low quality).
  • One study comprising 147 participants found that a statistically significant lower proportion of hourly checks that were wet in the prompted voiding group (8 weeks) (very low quality).
  • One study comprising 19 participants found that that there was no significant difference between prompted voiding and no prompted voiding for the reduction in the mean proportion of hourly checks that are wet (8 weeks) (very low quality).
  • Two studies comprising 257 participants found that a statistically significant lower number of incontinent episodes per 24 hours in the prompted voiding group (8–22 weeks)(very low quality).
  • One study comprising 126 participants found that a statistically significant higher amount of self initiated toileting in the prompted voiding group (8 weeks) (very low quality).

Evidence statements could not be produced for the following outcomes of the systematic review 39 as results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect

  • Proportion of hourly checks that were wet
  • Incontinence episodes in 24 hrs
  • Self initiated toileting
Comparison between habit training with one other treatment to usual care
  • Two studies comprising 130 participants found that that there was no significant difference between habit retraining with one other treatment and usual care for the number of incontinent episodes per 24 hours (12 – 26 weeks) (very low quality).
  • One study comprising 56 participants found that that there was no significant difference between habit retraining with one other treatment and usual care for incontinent volume (12 weeks)(very low quality).
  • One study comprising 56 participants found that that there was no significant difference between habit retraining with one other treatment and usual care for prevalence of bacteriuria (12 weeks)(very low quality).
  • Evidence statements could not be produced for the following outcomes of the study by Ostaszkiewicz 40 as results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect
    • Skin rash
    • Skin breakdown
Comparison of training mobility and toileting skills to no treatment in achievement of Independent toileting

Evidence statements could not be produced for the following outcomes of the study by van Houten41 as results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect

  • Weight of pads over 24 hr
  • Percentage of micturations on the toilet
  • Dependent to independent toileting
  • Independent to dependent toileting
Economic evidence statements
  • While the costs of these programmes are not inegligible, if effective their cost may be offset by the cost savings associated with a reduction in the use of incontinence aids (including catheters and pads).

8.1.2. Recommendations and links to evidence

Image

Table

Consider a behavioural management programme (for example, timed voiding, bladder retraining or habit retraining) for people with neurogenic lower urinary tract dysfunction: only after assessment by a healthcare professional trained in the assessment of (more...)

8.2. Antimuscarinics

8.2.1. What is the safety and efficacy of antimuscarinics compared with a) placebo or treatment as usual b) other antimuscarinics for the treatment of incontinence due to neurological disease/overactive bladder due to neurological disease?

Image

Table

Quality of life. Patients and carers’ perception of symptoms.

8.2.1.1. Clinical evidence

We searched for RCTs in adults and RCTs and observational studies in children, comparing the effectiveness of antimuscarinics for improving outcomes for patients with neurogenic detrusor overactivity (formerly called “detrusor hyperreflexia”)or patients with reduced bladder compliance.

This review compares antimuscarinics with either placebo/treatment as usual or with other antimuscarinics. For the adult population RCTs only were included. The within-subject drug comparisons from each RCT are presented separately. For children and young people RCT and observational studies were included. Studies with a sample size of 20 or less were excluded. For the adult population five RCTs were included in the review 43 44 45 46 47. For children and young people, thirteen observational studies were included in the review 48 49 50 51 52 53 54 55 56 57 58 59 60. Table 20 summarises the population, intervention and comparison.

Table 20. Summary of studies included in the clinical evidence review.

Table 20

Summary of studies included in the clinical evidence review.

Trospium vs oxybutynin
Adults – spinal cord injury

Table 25. Trospium vs oxybutynin - Clinical study characteristics and clinical summary of findings

Table 26. Trospium (before vs after treatment) - Clinical study characteristics and clinical summary of findings

No evidence was reported for the following outcomes:

  • Frequency of voiding by day and night, no. of incontinence episodes per week, quality of life, patients and carers’ perception of symptoms, adverse events, treatment adherence, kidney function or bladder compliance

Table 27. Oxybutynin (before vs after treatment) - Clinical study characteristics and clinical summary of findings

8.2.1.2. Economic Evidence

No studies could be found that assessed the cost effectiveness of antimuscarinic agents in the neurogenic population.

In order to aid evaluation of cost effectiveness, unit costs are provided below:

Table 37. Unit Costs of antimuscarinics contained in clinical review.

Table 37

Unit Costs of antimuscarinics contained in clinical review.

The clinical review shows antimuscarinics to be effective in reducing incontinence. The treatments are also low cost. CG40 provides evidence to suggest that antimuscarinics, particularly non-proprietry oxybutynin, are cost-effective in people with non-neurogenic incontinence. While this evidence is lacking in applicability to the neurogenic population, it is suggestive of cost effectiveness. The GDG also suggested that even better results can be achieved in neurogenic populations to non-neurogenic populations. The GDG considered on the basis of these factors combined these treatments are likely to be cost effective in patients with neurogenic lower urinary tract dysfunction.

Due to the fact that there is no high quality evidence to choose between the treatments and it is therefore not possible to recommend one treatment over another, in terms of side effects or effectiveness. All of the treatments are very low cost, with no treatment costing more than 80p per pill, therefore balancing the side effect profile with the cost of the pill is more important than making sure the pill is the lowest cost. Of course, where there is nothing to choose between the two, the lowest cost treatment should be provided.

8.2.1.3. Evidence Statements

Clinical Evidence Statements
Adults
Propiverine vs placebo
Adults - Spinal cord injury

One study of 113 participants found a statistically significant improvement for patients receiving propiverine compared to placebo for

  • Maximum cystometric capacity (14 days) (low quality)

One study of 113 participants found no significant difference for propiverine compared with placebo for

  • Residual urine (14 days) (low quality)
  • Bladder compliance (14 days) (low quality)
  • Drop-outs due to adverse events (low quality)

Evidence statements could not be produced for the following outcomes of the study by Stohrer 46 as results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect

Clinical symptoms
Propiverine vs oxybutynin
Adults - Spinal cord injury

One study comprising 91 participants found no significant difference for propiverine compared with oxybutynin for

  • 24-hr incontinence episodes (21 days) (high quality)
  • 24-hr micturition frequency (21 days) (high quality)
  • maximum cystometric capacity (21 days) (moderate quality)
  • bladder compliance (21 days ) (moderate quality)
  • residual urine (21 days) (moderate quality)
  • adverse events (21 days) (moderate quality)
Properverine (before vs after treatment)

One study comprising 91 participants suggested a difference in favour of propiverine for

  • 24 hr incontinence episodes (21 days) (very low quality)
  • Maximum cystometric capacity (21 days follow up) (very low quality)
  • Bladder compliance (21 days) (very low quality)

One study comprising 91 participants suggested a difference against (increase) propiverine for

  • Residual urine (21 days follow up) (very low quality)
Oxybutynin (before vs after treatment)

One study comprising 91 participants suggested a difference in favour of oxybutynin for

  • 24 hr incontinence episodes (21 days) (very low quality)
  • Maximum cystometric capacity (21 days follow up) (very low quality)
  • Bladder compliance (21 days) (very low quality)

One study comprising 91 participants suggested a difference against (increase) propiverine for

  • Residual urine (21 days) (very low quality)
Trospium vs oxybutynin
Adults – spinal cord injury

One study comprising 95 participants found no significant difference for trospium compared with oxybutynin for

  • maximum cystometric capacity (3 weeks) (low quality)
  • residual urine (3 weeks) (low quality)
  • treatment adherence (3 weeks) (low quality)
  • adverse events (3 weeks) (low quality)
Trospium (before vs after treatment)

One study comprising 95 participants suggested a difference in favour of trospium for

  • Maximum cystometric capacity (3 weeks) (very low quality)

One study comprising 95 participants suggested a difference against (increase) trospium for

  • Residual urine (3 weeks) (very low quality)
Oxybutynin (before vs after treatment)

One study comprising 95 participants suggested a difference in favour of oxybutynin for

  • Maximum cystometric capacity (3 weeks) (very low quality)

One study comprising 95 participants suggested a difference against (increase) trospium for

  • Residual urine (3 weeks) (very low quality)
Oxybutynin vs propantheline
Adults – multiple sclerosis

One study comprising 34 participants found a significant improvement in favour of oxybutynin compared with propantheline for

  • maximum cystometric capacity (6 to 8 weeks) (very low quality)
Oxybutynin (before vs after treatment)

One study comprising 34 participants suggested an improvement in favour of oxybutynin for

  • Maximum cystometric capacity (6 to 8 weeks) (very low quality)
Propantheline (before vs after treatment)

One study comprising 34 participants suggested there was no difference for propantheline (before vs after treatment) for

  • Maximum cystometric capacity (6 to 8 weeks) (very low quality)
Atropine vs oxybutynin
Adults – multiple sclerosis

Evidence statements could not be produced for the following outcome of the study by Fader 43 as results were presented of the intervention effect in a way that meant we could not estimate the size of the intervention effect

  • Incontinence
  • Maximum cystometric capacity
  • Adverse events
Oxybutynin vs placebo
Children and young people

Two studies of 45 participants suggested that, compared to placebo, oxybutynin

  • Improved continence (4 weeks to 21 months) (very low quality)
  • Increased maximum cystometric capacity (4 weeks to 21 months) (very low quality)
  • Increased adverse events (4 weeks to 21 months) (very low quality)
Oxybutynin (before vs after treatment)
Children and young people

Six studies of 267 participants suggested that oxybutynin improved

  • Continence (2 to 60 months)(very low quality)

Five studies of 296 participants suggested that oxbutynin increased

  • Maximum cystometric capacity (2 to 36 months) (very low quality)

Three studies of 155 participants suggested that oxybutynin improved

  • bladder compliance (3 to 36 months) (very low quality)

Four studies of 145 participants suggested that oxybutynin increased

  • adverse events (2 to 60 months) (very low quality)

Two of three studies of 182 participants suggested that oxybutynin increased

  • urinary tract infections (36 to 60 months) (very low quality)

Five studies of 226 participants reported discontinuations ranging from 6% to 65%

Tolterodine (before vs after treatment)
Children and young people

One study of 30 participants suggested that tolterodine

  • Improved continence (12 months) (very low quality)
  • Improved functional bladder capacity (12 months) (very low quality)
  • Increased adverse events (12 months) (very low quality)

The withdrawal rate was 3%

Propiverine (before vs after treatment)
Children and young people

One study of 20 participants suggested that propiverine

  • Improved continence (3 to 6 months) (very low quality)
  • Improved maximum cystometric capacity (3 to 6 months) (very low quality)
  • Improved bladder compliance (3 to 6 months) (very low quality)
  • Increased adverse events (3 to 6 months) (very low quality)
Propiverine vs oxybutynin
Children and young people

One study comprising 255 participants suggested that propiverine and oxybutynin

  • Improved continence (12 months or longer) (very low quality)
  • Improved maximum cystometric capacity (12 months or longer) (very low quality)
  • Increased adverse events (12 mths of longer follow up) (very low quality)

8.2.1.4. Economic Evidence Statements

  • Antimuscarinic agents are likely to be cost-effective for the treatment of patients with urinary incontinence from neurological cause.

8.2.2. Recommendations and links to evidence

Image

Table

Offer antimuscarinic drugs to people with: spinal cord disease (for example, spinal cord injury or multiple sclerosis) and

8.2.3. Research recommendations

Image

Table

What is the safety and efficacy of more recently developed antimuscarinics compared with (a) placebo/usual care and (b) other antimuscarinics in the treatment of neurogenic lower urinary tract dysfunction? Why this is important:

8.3. Botulinum toxin

8.3.1. What is the safety and efficacy of detrusor injections of botulinum toxin type Ai or B compared with a) usual care b) antimuscarinics in neurological disease

8.3.1.1. Clinical evidence

We searched for RCTs comparing the short-term effectiveness of botulinum toxin type A or B compared to usual care, antimuscarinics or augmentation cystoplasty in adults and for observational studies comparing the longer-term effectiveness (two or more injections of botulinum toxin type A or B) in adults. For children we searched for RCTs or observational studies comparing the short-term or long-term effectiveness of botulinum toxin type A or B, usual care, antimuscarinics or augmentation cystoplasty. All of the searches were on interventions for improving incontinence in neurological disease or injury

Adults

No relevant studies were found on botulinum toxin type B. No studies were found comparing botulinum toxin type A with augmentation cystoplasty. The majority of studies comprised patients who were either on antimuscarinics, or antimuscarinics had failed to control their symptoms.

Studies on the shorter-term (one cycle of treatment) efficacy compared botulinum toxin type A with placebo and were on adults with neurogenic detrusor overactivity 62, neurogenic detrusor overactivity secondary to spinal cord injury or multiple sclerosis 63 64 65 66 67.

Eight longer-term (two or more cycles of treatment) observational studies were identified comparing before and after botulinum toxin type A in adults with neurogenic detrusor overactivity 63 68 69 70 71 neurogenic detrusor overactivity due to spinal cord lesions72 73 74 neurogenic lower urinary tract dysfunction 75.

The botulinum toxin type A preparations are individual and not interchangeable so the results are reported by preparation (Botox (Allergan), Dysport (Ipsen), unclear/both preparations).

Tables 33 and 34 summarise the population, intervention, comparison and outcomes for each of the studies. Table 35 summarises the number of injections the adults received in the longer-term efficacy studies.

Children

No studies were identified on botulinum toxin type B. No studies were identified comparing botulinum toxin with augmentation cystoplasty.

The majority of patients were either on antimuscarinics or antimuscarinics had failed to control their symptoms.

One RCT (N=23) in children with neuropathic bladder after repair of myelomeningocele was identified comparing botulinum type A (preparation not specified) plus continued oxybutynin with discontinuation of oxybutynin 76.

The observational studies compared before and after botulinum toxin type A in children with myelomeningocele or spina bifida 77; 78; 79; 80, myelodysplasia 81; spinal cord lesions 82 and neurogenic detrusor overactivity/hyperreflexia 83; 84; 85; 86.

The results are reported by preparation (Botox, Dysport, unclear/both preparations)

Tables 38 to 40 summarise the population, intervention, comparison and outcomes for each of the studies.

Table 38. Summary of studies included in the clinical evidence review - Adults.

Table 38

Summary of studies included in the clinical evidence review - Adults.

Table 39. Summary of studies included in the clinical evidence review - Children.

Table 39

Summary of studies included in the clinical evidence review - Children.

Table 40. Summary of the No. of injections for longer-term observational studies in adults.

Table 40

Summary of the No. of injections for longer-term observational studies in adults.

Adults longer-term follow up data

8.3.1.2. Economic Evidence

Three studies8789 were found but excluded on the basis of potentially serious limitations and partial applicability that (see list of excluded studies).

This area was identified as important for economic evaluation given the uncertainty over the tradeoff between cost and effectiveness. Therefore an original cost-effectiveness analysis was conducted to answer this question.

Novel Cost Effectiveness Analysis

In order to explore the cost effectiveness of botulinum toxin for the treatment of NLUTD, a full cost effectiveness analysis was carried out. The key methodology and results are written up here but the full report can be found in appendix I.

Model Overview
Comparators

The model compares the cost effectiveness of four strategies for the management of incontinence due to neurogenic lower urinary tract dysfunction (NLUTD):

Augmentation Cystoplasty (AC) is a well established major open surgical technique where the bladder is made larger or ‘augmented’ by incorporating a bowel segment into the bladder. Most commonly an ileal segment is used but alternatives include a section of the large intestine. The incorporation of intestine into the bladder prevents effective bladder contractions from occurring and patients usually cannot void completely following the surgery, therefore needing to perform clean intermittent self catheterisation.

The second intervention is the injection of botulinum toxin type A (BTX) into the bladder wall. BTX is currently not licensed for this indication but various trials have shown it to be effective in reducing the frequency of incontinence episodes63,65,66 in patients with incontinence due to NLUTD. The protocol for administration of BTX varies but the method used in this model is 30 endoscopic injections of 300u or 200u into the bladder wall. The operation will take less than 1 hour. Patients with NLUTD will mostly need to use intermittent catheterisation to empty the bladder effectively following treatment.

The third strategy is where BTX is administered for two variable cycles (6–12 months) and then AC is conducted in 100% of those that do not respond to BTX (BTX100AC). BTX continues to be administered in those that do respond.

The final comparator is no treatment or “best supportive care” (No-Rx). This comparator is included as an arm where patients opt to manage their incontinence with a mixture of incontinence appliances: pads, indwelling catheters, sheaths and suprapubic catheters.

Population

The population in this model is made up of patients with NLUTD (Myelomeningocele, Spinal Cord Injury, Multiple Sclerosis etc.) and bladder over-activity who are unresponsive or intolerant to antimuscarinic medication. The patients in the base case are considered to be adults as the paucity of data on children prevents an adequate analysis for the paediatric age group. However the cost effectiveness in children will be tested in a sensitivity analysis. Patients had an average age of 49 with a sex distribution of 53% female and 47% male. Mortality data was adjusted using a standardised mortality ratio from a group of patients with spinal cord injury90. Subgroup analysis was carried out on different groups of patients to determine cost effectiveness in a paediatric population.

However, not all of the comparators are relevant in every situation. For some patients, such as multiple sclerosis patients, the AC comparator is not relevant as they are not suitable for this surgical option. There are therefore two base case comparisons. Base case 1 is all the comparators compared together. The second base case analysis is simply BTX compared with No-Rx.

Model Structure and approach to modelling

A decision tree was constructed in Windows Excel® to model the comparison of cost and effectiveness of the interventions. Upon receiving treatment a patient could end up in one of three possible health states: incontinent, mildly incontinent or continent. Once in any of these health states, they would remain there for the duration of the model. In order to model the long term effects and survival, life tables were then attached to each of the final health states in the tree and a hypothetical cohort of a thousand patients was run through the model. The trials that were used to inform the model used frequency of incontinence episodes as the main outcome. Quality of Life weights were attached to being either incontinent, continent or having mild incontinence on the basis of the frequency of episodes. As adverse events and the presence or absence of urinary tract infections have important quality of life and cost implications, these were also included. The cost components included costs of the treatment itself, the ongoing costs associated with adverse events and any monitoring or follow up treatments. A diagram of the model can be found in Figure 5.

Figure 5. Decision Tree.

Figure 5

Decision Tree.

Outcomes

As outlined previously, with each of the four strategies, an incontinent patient will either become continent, meaning that the treatment was effective; they will have improved continence but will not be fully continent - mild incontinence; or they will remain incontinent. Each of these options is as the main outcome. Due to the inconstant reporting of the effectiveness of treatments between studies, assumptions had to be made about the frequency of incontinence episodes that constituted each outcome. This was done so that costs and effects could be calculated. It was assumed that in the continent group a patient would suffer from one incontinence episode per week; in the mild incontinent group, they would suffer from two episodes per day; and in the incontinent group, they would suffer from five episodes per day.

As well as the main effectiveness estimate, there were also adverse events (AEs) and urinary tract infections (UTIs) to consider. AEs were associated with the strategy used to manage incontinence. The UTIs were associated with the continence status of the patient.

Results
Base case 1 results – All interventions compared

The first base case analysis compared the cost-effectiveness of all the interventions outlined in the methods. The analysis revealed that Augmentation Cystoplasty (AC) is the cost-effective option when compared to botulinum toxin (BTX) and no treatment (No-Rx) for the treatment of incontinence due to NLUTD using a lifetime horizon. The results of the analysis can be seen in Table 65, below. There is a measure of confidence in this result because, at a threshold of £20,000 per QALY, AC is cost-effective with a probability of 78%.

Table 65. Base case results.

Table 65

Base case results.

Figure 6 demonstrates these cost-effectiveness results graphically. We can see that while BTX and AC are similar in cost-effectiveness, AC is more effective but marginally more expensive than BTX alone. The BTX100AC strategy is more effective than the BTX alone strategy but also more expensive; it is more expensive and less effective than AC. No-Rx is the cheapest strategy but it is also the least effective therefore it will only be cost-effective at a very low threshold.

When the costs are broken down into the constituent parts, it is possible to pick out the elements that drive the results. This breakdown can be found in Table 66. The increased effectiveness of AC compared with all other interventions is what makes it the most cost-effective option. It is cheaper than BTX100AC over a lifetime and is more effective; it is not, however, cheaper than BTX alone over a lifetime.

Table 66. Breakdown of costs and outcomes.

Table 66

Breakdown of costs and outcomes.

AC is higher cost than the BTX alone strategy, which is a function of the discount ratek. However, AC is more effective and only marginally more expensive than BTX, meaning it is cost-effective over a lifetime compared with BTX. A time horizon analysis was also carried out on this comparison in Figure 7: this revealed that for the first 5 cycles, about 3 years, BTX alone is cost effective. Between 5 and 16 cycles, about 10 years, BTX with 100% AC after failed BTX is the cost effective strategy. Beyond 16 cycles, AC is cost effective. This shows that for patients with a poor prognosis and for older patients, BTX is a more cost effective option.

Figure 7. Net benefit compared with age.

Figure 7

Net benefit compared with age. Note: CE = Cost Effective

If this is then broken down further into the main comparison, AC-BTX100AC, we can see the key drivers behind AC’s cost effectiveness in Table 67. The BTX100AC strategy is analysed against AC because it is more cost effective and is the most relevant comparison for sub analysis. A patient with AC only will spend more time in the continent group than those in the BTX100AC arm, and their cost of treatment will be lower in spite of higher adverse event rates. The 18 years compared to 11 spent in the continent arm counts towards an increased QALY gain compared with BTX100AC.

Table 67. Cost Breakdown AC-BTX100AC.

Table 67

Cost Breakdown AC-BTX100AC.

Base case 2 results – Botulinum Toxin versus No Treatment

As a second analysis we looked at a comparison of BTX with a no treatment comparator. This was to ensure that we captured the full range of potential patients in the analysis. For some patients, such as multiple sclerosis patients, the AC comparator is not relevant as neurological deterioration is likely to occur and render the management of the augmented bladder problematic. In Table 68 it is possible to see that BTX is cost effective when compared to no treatment with a cost per QALY of under £9,000. This is well below the usual cost effectiveness threshold of £20,000 per QALY gained.

Table 68. BTX – No Treatment base case results.

Table 68

BTX – No Treatment base case results.

Table 69 shows where the cost and outcome differences lie. The cost of no treatment is lower than BTX but it is not zero. This is due to the cost of incontinence appliances such as pads and catheters. BTX is also more effective with increased time spent in the continence and mild incontinence groups. BTX has higher QALYs but also higher costs, so it is cost effective but not dominant.

Table 69. Breakdown of costs and outcomes (BTX – No Rx).

Table 69

Breakdown of costs and outcomes (BTX – No Rx).

As a result of these costs and of the increased effectiveness of BTX, BTX is more expensive but also more effective with a high degree of certainty. This is displayed on the cost effectiveness plane in Figure 8. This shows that using the probabilistic analysis, all of the cost effectiveness ratios for BTX versus no treatment are to the North East of zero meaning that for all 1000 iterations of the model, BTX is more costly and more effective. And the vast majority, 988, of these ratios fall under the £20,000 per QALY threshold.

Figure 8. Cost effectiveness plane.

Figure 8

Cost effectiveness plane.

Conclusions

The results of the model allow four main conclusions to be drawn:

  1. AC is the cost effective intervention over a lifetime horizon in the populations where it is a relevant comparator.
  2. BTX is cost effective compared to AC in patients who are unsuitable for surgery.
  3. A BTX strategy where AC is used (and relevant) in 100% of patients after failed BTX is cost effective compared to a 0% progression to AC strategy but is higher cost.
  4. BTX is cost effective when compared to no treatment.

The results of this model are generally robust to the uncertainty around the assumptions made as shown by the deterministic sensitivity analyses. The probabilistic data shows that at a threshold of £20,000 per QALY gained AC is cost effective with a probability of 78%, again demonstrating the robustness of the model to uncertainty.

The many limitations are almost entirely due to the lack of good quality data to populate the model. Perhaps the most important limitation is the fact that there is no comparative data on AC and BTX. Therefore the comparison between these two interventions is made on the basis of two fairly heterogeneous studies. The BTX vs placebo study was a randomized control trial 63 whereas the study used to provide AC data was based on observational data 91. This disparity means that the outcomes: continence, mild incontinence and incontinence, are not measured in the same way. It was necessary for the GDG to make assumptions about the definition of what constituted these outcomes, which was not ideal but given the available data was the only solution. The result of this is that it makes the comparison of BTX with no treatment more reliable than the comparison of AC with BTX or no treatment. However, the probabilistic analysis allows us to take this uncertainty into account and deal with it explicitly.

The analysis took place in two parts. The first part being the comparison of all interventions in a population where all comparators were relevant, such as a spinal cord-injured population. The second part was a comparison of just BTX with no treatment. This was therefore in a population where AC was not a relevant comparator such as patients with multiple sclerosis. This analysis is therefore generalisable to any patient that suffers from incontinence due to NLUTD in the UK. The model is also of potential relevance to populations outside of the UK as the model is fairly robust to changes in costs and impact of adverse events.

Only one other cost effectiveness study has been done that analyses AC vs BTX. The study by Padmanabhan et al. 201192 showed that BTX would cost about $5,000 less than AC per successful intervention. However this analysis only uses adverse events as outcomes and is a five year study from a US payer perspective. This is in keeping with what our model shows as BTX only is shown to be cost effective when compared with AC for the first six years of the model. However as the Padmanabhan study is from a US payer perspective and does not consider outcomes beyond adverse events, its relevance to the UK perspective is limited.

8.3.1.3. Evidence Statements

Clinical Evidence Statements
Shorter-term safety and efficacy
Adults, Botox 200 U

Evidence statements could not be produced for the following outcomes of the study by Cruz 63 and Schurch 67 as results were presented in a way that meant we could not estimate the size of the intervention effect :

  • I-QoL (mean change score (no SD) (6 weeks) (high quality).
  • I-QoL (final median score)(6 weeks, 24 weeks) (moderate quality).

One study of 184 participants found a statistically significant reduction for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/week (mean change score) (6 weeks) (moderate quality).

One study of 40 participants found no statistically significant difference for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/day (mean change score) (6 weeks) (low quality).

One study of 40 participants found a statistically significant reduction for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/day (mean change score) (24 weeks) (low quality).

One study of 184 participants found a statistically significant improvement for participants receiving botulinum toxin type A compared to placebo for:

  • Maximum bladder capacity (mean change score) (6 weeks) (high quality).

Evidence statements could not be produced for the following outcomes of the study by Schurch 66 as results were presented in a way that meant we could not estimate the size of the intervention effect :

  • Maximum bladder capacity (mean change score (no SD)) (6 weeks, 24 weeks) (moderate quality).

One study of 181 participants found a statistically significant increase for participants receiving botulinum toxin type A compared to placebo for:

  • All adverse events (end of scheduled follow-up) (moderate quality)

One study of 181 participants found no statistically significant difference for participants receiving botulinum toxin type A compared to placebo for:

  • Muscle weakness (end of scheduled follow-up) (moderate quality).

One study of 181 participants found a statistically significant increase for participants receiving botulinum toxin type A compared to placebo for:

  • Urinary tract infections (end of scheduled follow-up) (moderate quality).
Adults, Botox 300 U

One study of 57 participants found a statistically significant improvement for participants receiving botulinum toxin type A compared to placebo for:

  • I-Qol (mean change scores) (6 weeks, 24 weeks) (moderate to high quality).

Evidence statements could not be produced for the following outcomes of the study by Cruz 63 and Schurch 67 as results were presented in a way that meant we could not estimate the size of the intervention effect:

  • I-QoL (mean change score (no SD)) (6 weeks) (high quality).
  • I-Qol (final median score) (6 weeks, 24 weeks) (moderate quality).

One study of 183 participants found no statistically significant difference for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/week (mean change score) (6 weeks) (moderate quality).

One study of 57 participants found a statistically significant reduction for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/day (mean final score) (6 weeks, 24 weeks) (high quality).

One study of 40 participants found a statistically significant reduction for participants receiving botulinum toxin type A compared to placebo for:

  • Incontinence episodes/day (mean change score) (6 weeks, 24 weeks) (low quality).

One study of 183 participants found a statistically significant improvement for participants receiving botulinum toxin type A compared to placebo for:

  • Maximum bladder capacity (6 weeks) (high quality).

Evidence statements could not be produced for the following outcomes of the study by Herschorn 65 and Schurch 66 as results were presented in a way that meant we could not of the intervention effect in a way that meant we could not estimate the size of the intervention effect :

  • Maximum bladder capacity (final median score) (6 weeks, 24 weeks) (high quality).
  • Maximum bladder capacity (mean change score (no SD)) (6 weeks, 24 weeks) (moderate quality).

One study of 179 participants found a statistically significant increase for participants receiving botulinum toxin type A compared to placebo for:

  • All adverse events (end of scheduled follow-up) (moderate quality).

Two studies of 145 participants found no statistically significant difference for participants receiving botulinum toxin type A compared to placebo for:

  • Muscle weakness (end of scheduled follow-up) (moderate quality).

Three studies of 285 participants found a statistically significant increase for participants receiving botulinum toxin type A compared to placebo for:

  • Urinary tract infection (end of schedules follow up) (moderate quality).
Adults, Dysport

Evidence statements could not be produced for the following outcomes of the study by Ehren 62 as results were presented in a way that meant we could not estimate the size of the intervention effect:

  • Quality of life
  • Continence
  • Maximum cystometric capacity
Both/unclear preparations

No studies were identified.

Longer-term follow-up data
Adults, Botox

4 studies of 261 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in quality of life

4 studies of 246 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in continence

5 studies of 161 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in maximum cystometric capacity

1 study of 17 participants suggested that botulinum toxin type A was associated with a:

  • Decrease in urinary tract infections
Adults, Dysport

1 study of 199 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in maximum cystometric capacity
  • Improvement in patient satisfaction
Adults, both/unclear

1 study of 66 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in continence

1 study of 27 participants reported

  • Muscle weakness
Children
Botulinum toxin type A plus oxybutynin compared with botulinum toxin type A oxybutynin discontinued
(both/unclear preparation)

Evidence statements could not be produced for the following outcomes of the study by Neel 76 as results were presented in a way that meant we could not of the intervention effect in a way that meant we could not estimate the size of the intervention effect

  • Continence
  • Maximum cystometric capacity
  • Side effects
Botulinum toxin A pre vs post treatment
Children, Botox

4 studies of 77 participants suggested that botulinum toxin type A was associated with an:

  • Improvement in continence

6 studies of 108 participants suggested botulinum toxin type A was associated with an:

  • Increase in maximum cystometric capacity

1 study of 27 participants suggested botulinum toxin type A was associated with an:

  • Improvement in kidney function

4 studies of 74 participants suggested botulinum toxin type A was associated with an:

  • Increase in urinary tract infections
Children, Dysport

1 study of 19 participants suggested botulinum toxin type A was associated with an:

  • Increase in maximum cystometric capacity
Children, both/unclear preparation

1 study of 15 participants suggested botulinum toxin type A was associated with an:

  • Improvement in continence

2 studies of 35 participants suggested botulinum toxin type A was associated with an:

  • Increase in maximum cystometric capacity
Economic Evidence Statements
  • Augmentation cystoplasty is cost effective compared to botulinum toxin type A in patients where it is suitable.
  • Botulinum toxin type A is cost effective compared to augmentation cystoplasty in patients who are unsuitable for surgery.
  • A Botulinum toxin type A strategy where augmentation cystoplasty is used (and relevant) in 100% of patients after failed Botulinum toxin type A is cost effective compared to a 0% progression to augmentation cystoplasty strategy but is higher cost.
  • Botulinum toxin type A is cost effective when compared to no treatment.

8.3.2. Recommendations and links to evidence

Image

Table

Offer bladder wall injection with botulinum toxin type A to adults: with spinal cord disease (for example, spinal cord injury or multiple sclerosis) and

8.3.3. Research recommendations

Image

Table

What is the safety and efficacy of botulinum toxin compared with (a) usual care, (b) antimuscarinics and (c) augmentation cystoplasty in people with neurogenic lower urinary tract dysfunction? Why this is important

Image

Table

What is the safety and efficacy of botulinum toxin compared with (a) usual care, (b) antimuscarinics and (c) augmentation cystoplasty in people with primary cerebral conditions with lower urinary tract dysfunction? Why this is important

8.4. Augmentation cystoplasty

8.4.1. What is the safety and efficacy of augmentation cystoplasty compared with a) botulinum toxin b) usual care in neurological disease c) urinary diversion?

Image

Table

Augmentation cystoplasty Botulinum toxin

8.4.1.1. Clinical evidence Review

We searched for observational studies comparing the effectiveness of augmentation cystoplasty as an intervention for improving incontinence in people with neurogenic lower urinary tract dysfunction (NLUTD). We searched for any observational studies that compared the effectiveness of augmentation cystoplasty with one or more of botulinum toxin, urinary diversion and usual care; however no studies made these comparisons, and all compared findings before surgery with those after surgery.

33 observational studies were identified, evaluating the effects of augmentation cystoplasty on incontinence associated with NLUTD 93 94 95 96 97 98 99 100 101 102 103 104 25 105 106 107 108 109 110 111 112 113 114 115 116 91 117 118 119 120 121 122 123. The augmentation procedures were fairly homogenous across 27 of the studies, varying only by the section of intestine used for the augmentation. However, 4 studies reported auto-augmentation 106 107 120 121, and one used dural tissue 93, and findings from these potentially distinct studies will be highlighted in the following report. There were 11 studies in children (<19 years), 9 in adults (≥19 years) and 13 in mixed age-group samples. The results are reported by outcome. Table 70 summarises the population, age range, follow-up periods and type of surgical material for each of the studies.

Table 70. Summary of studies included in the clinical evidence review.

Table 70

Summary of studies included in the clinical evidence review.

Quality of studies

The quality of all evidence was classified as very low. The studies were retrospective observational studies and therefore graded low by default (see Chapter 4). The further downgrade was due to a lack of attempts to eliminate threats to internal validity through the use of a matched comparison group. However it should also be noted that 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. 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.

Incontinence outcome

All studies suggested that augmentation cystoplasty would reduce the likelihood of incontinence. Auto-augmentation appeared to show less benefit than intestinal augmentation, but this impression was based on only one study. Tables 71–73 show results for children, adults and mixed-age studies respectively.

Table 71. Effects of augmentation on incontinence in children.

Table 71

Effects of augmentation on incontinence in children.

Table 72. Effects of augmentation on incontinence in adults.

Table 72

Effects of augmentation on incontinence in adults.

Table 73. Effects of augmentation on incontinence in mixed age groups.

Table 73

Effects of augmentation on incontinence in mixed age groups.

Need for intermittent catheterisation outcome

This outcome was weakly reported, with many studies failing to clearly specify the number of patients using intermittent catheterisation pre-operatively or post operatively. Overall, the effects of augmentation cystoplasty on the need for intermittent catheterisation are unclear, but there is a possibility that the need for intermittent catheterisation may increase. [Note: Expert evidence advises that intermittent classification is usually required following augmentation cystoplasty in patients with a neuropathic bladder]. Table 74 summarises the results.

Table 74. Effects of augmentation on the need for intermittent catheterisation.

Table 74

Effects of augmentation on the need for intermittent catheterisation.

Quality of life/Patient satisfaction (post surgery) outcomes

Five studies collected data on patient satisfaction or quality of life, and all suggested that the procedure led to patient satisfaction and had a positive impact on quality of life. However, non-validated questionnaires were used for all studies except Mitsui 2008111, and there were no reports of methods to reduce bias during collection of these data. The quality of these data is therefore very low.

Children

Mitsui 2008 (n=15) 111: 14/15 were ‘satisfied with surgery’ post operatively.

Adults

Herschorn 1998 (n=59) 100: 41/59 patients delighted, 12/59 pleased, 6/59 mostly satisfied. On a scale of 0–2, with 0 representing the highest satisfaction, mean response was 0.42. All but one would go through the surgery again.

Zachoval 2003 (n=9) 123: Quality of life score was 0.7/6 post-operatively (with 6=unbearable quality of life), compared to 5/6 pre-operatively.

Mixed age-group

Quek 2003 (n=26) 115: ‘Nearly all patients expressed extreme satisfaction’ and all but one would recommend the procedure to others. Mean satisfaction score out of 10 was 8.7.

Khastgir 2003 (n=32) 102: 26/27 reported excellent quality of life post surgery, and improvement in the management of the urinary tract. 27/27 would recommend the surgery to others. 0/27 reported deterioration in sexual function, and 5/7 reported a reduction in UTIs.

Adverse events (post surgery)

A variety of adverse effects of the surgery were reported, and the most important ones are documented in the tables below, with the data below concerning patients affected at least once. The most commonly reported adverse events were symptomatic UTIs [children aggregate: 13/117; adult aggregate: 34/61; mixed age aggregate: 8/90; all groups 55/268], bladder stones [children aggregate: 7/101; adult aggregate: 13/40; mixed age aggregate: 82/605; all groups 102/746], and bowel obstruction [children aggregate: 7/101; mixed age aggregate: 29/752; all groups: 36/853]. Auto augmentation appeared to show a greater numerical frequency of renal adverse effects than intestinal augmentation in children, but this evidence was from one study only. 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 (e.g. vesicoureteral reflux [VUR]). Table 75 summarises these results.

Table 75. Adverse effects.

Table 75

Adverse effects.

Bladder capacity and detrusor pressure outcome

All studies showed some evidence of benefit. Many studies failed to report useful measures of variance, instead reporting ranges or no measure at all. In general, auto augmentation led to more modest effects than intestinal augmentation. Table 76 summarises these results.

Table 76. Effects of augmentation on bladder capacity and detrusor pressures.

Table 76

Effects of augmentation on bladder capacity and detrusor pressures.

8.4.1.2. Economic evidence

This area was identified as important for economic evaluation given the uncertainty over the tradeoff between cost and effectiveness. Therefore an original cost-effectiveness analysis was conducted to answer this question.

Please see cost- effectiveness analysis in Appendix I for the full model write-up including methods, results and discussion.

8.4.1.3. Evidence Statements

Clinical Evidence Statements

23 Observational studies comprising 680 participants suggested that augmentation might improve incontinence (2 – 210 months)(very low quality).

15 Observational studies comprising 459 participants suggested that augmentation might increase the need for intermittent catheterisation (2 – 175 months)(very low quality).

5 Observational studies comprising 141 participants suggested that augmentation might improve patient satisfaction and quality of life (2 – 175 months)(very low quality).

23 Observational studies comprising 1155 participants suggested that the main adverse effects of augmentation are UTIs, bladder stones and bowel obstruction (2 – 210 months)(very low quality).

23 Observational studies comprising 451 participants suggested that augmentation might improve bladder capacity and reduce detrusor pressures (2 – 210 months)(very low quality).

Health economics evidence statements
  • Augmentation cystoplasty is cost effective compared to botulinum toxin type A in patients where it is suitable.
  • Botulinum toxin type A is cost effective compared to augmentation cystoplasty in patients who are unsuitable for surgery.
  • A Botulinum toxin type A strategy where augmentation cystoplasty is used (and relevant) in 100% of patients after failed Botulinum toxin type A is cost effective compared to a 0% progression to augmentation cystoplasty strategy but is higher cost.
  • Botulinum toxin type A is cost effective when compared to no treatment.

8.4.2. Recommendations and links to evidence

Image

Table

Consider augmentation cystoplasty using an intestinal segment for people: with non-progressive neurological disorders and

Footnotes

i

At the time of publication (August 2012), botulinum toxin type A did not have UK marketing authorisation for this indication. The prescriber should follow relevant professional guidance, taking full responsibility for the decision. Informed consent should be obtained and documented. See the GMC’s ‘Good practice in prescribing medicines – guidance for doctors’ for further information.

k

The discount rate is applied to all costs and outcomes. The discount rate is applied to future costs and outcomes to establish their present value. The rate of 3.5% reduction in value per year is based on the interest rate. If we invested now for a future expenditure, how much it would cost in present value.