10.2.1.1. Recommendations

30.

Patients and carers should be educated about and trained in techniques of hand decontamination, insertion of intermittent catheters where applicable, and catheter management before discharge from hospital. [2003]

31.

Community and primary healthcare workers must be trained in catheter insertion, including suprapubic catheter replacement and catheter maintenance. [2003]

32.

Follow-up training and ongoing support of patients and carers should be available for the duration of long-term catheterisation. [2003]

10.3.1.1. Recommendations

33.

Indwelling urinary catheters should be used only after alternative methods of management have been considered. [2003]

34.

The patient’s clinical need for catheterisation should be reviewed regularly and the urinary catheter removed as soon as possible. [2003]

35.

Catheter insertion, changes and care should be documented. [2003]

10.4.1. How to select the right system

Choosing the right system for any given patient will depend on a comprehensive individual patient assessment.

Our search identified one systematic review²³⁹ concerning the approaches to catheterisation. This reported a higher rate of infection associated with indwelling rather than intermittent catheterisation. This finding is reflected in a recent position paper¹⁸⁹ on urinary tract infections in long-term care facilities by the Society for Healthcare Epidemiology of America (SHEA) who recommended that “where clinically appropriate, intermittent catheterisation should be used for urinary drainage rather than a chronic indwelling catheter.”

Two studies were identified in our search which compared catheter options.^125,258 The first focussed on the risk of Meticillin-resistant Staphylococcus aureus (MRSA) colonisation and infection in nursing home patients.²⁵⁸ This study concluded that indwelling catheters posed a greater risk of infection than intermittent catheters. The second studied men with prostatic enlargement and reported a significantly lower rate of infection in those with suprapubic rather than urethral catheters, despite the former being used for two weeks longer.¹²⁵ A non-comparative study of patients with neuropathic bladder demonstrated a low rate of infection (6 percent) associated with the use of long-term suprapubic catheters.²⁴⁰ However, 30% of patients in this study reported other catheter-related complaints. Economic opinion suggests that if staff and resource use are the same, suprapubic catheterisation is more cost effective.^229,240

Eight studies were identified which focussed exclusively on the use of intermittent catheterisation. The study populations encompassed a wide range of patient groups and ages.^{17–19,42,79,174,200,274} One theme emerging from these studies was that the prevalence of bacteriuria is equal between men and women^17,18 though the incidence of clinical UTI appears to be higher in women.^18,19 There is also some evidence that bacteriuria rates are similar between adults and children.⁵⁷

Generally, large studies indicated that the rates of infection associated with intermittent catheterisation were low,^200,274 1 per 87 months,²⁷⁴ and that hydrophilic catheters were associated with a further reduction in infection risk.^19,42

A possible alternative to indwelling and intermittent catheterisation is the penile sheath (condom catheter). Whilst our systematic review did not include a specific question related to the use of penile sheath catheters, there is evidence that this type of device may be preferable in men who are able to empty their bladder and are unlikely to manipulate the system.^57,229 To date there are no controlled studies comparing penile sheaths with indwelling devices.

10.5.1. Review question – intermittent catheters

Long-term urinary catheterisation is considered an important area where updated guidance is required.

The following two questions both address the clinical and cost effectiveness of intermittent self-catheterisation. They were addressed independently for the clinical evidence review, but incorporated into the same economic model.

1. What is the clinical and cost effectiveness of different types of long-term intermittent urinary catheters (non-coated, hydrophilic or gel reservoir) on symptomatic urinary tract infections, bacteraemia, mortality, and patient preference?
2. In patients performing intermittent catheterisation, what is the clinical and cost effectiveness of non-coated catheters reused multiple times compared to single-use on urinary tract infections, bacteraemia, mortality, and patient preference?

10.5.1.3. Cost-effectiveness evidence – original economic model

Methods

A cost-utility analysis was undertaken to evaluate the cost-effectiveness of different types of intermittent catheters. A Markov model was used to estimate the lifetime quality-adjusted life years (QALYs) and costs from a UK NHS and personal social services perspective. Both costs and QALYs were discounted at a rate of 3.5% per annum in line with NICE methodological guidance. The model was built probabilistically to take into account uncertainty surrounding each of the model input parameters.

Population & comparators

The population evaluated in the base case analysis was people with bladder dysfunction caused by spinal cord injury (SCI). This population was chosen for the base case as it most closely matched the population considered by the majority (4/5) of the RCTs included in the clinical review and because this group of patients is one of the largest users of intermittent catheters. The average age of the population entering the model was 40 years and 80% were assumed to be male; this is the average age at injury and gender composition of the UK population of people with SCI.

A similar model exploring the cost-effectiveness of intermittent catheterisation in patients with bladder dysfunction not due to SCI was considered as part of the sensitivity analysis.

The comparators selected for the model were the types of intermittent catheter available to patients living or being cared for in the community:

• Single-use hydrophilic catheters
• Single-use gel reservoir catheters
• Single-use non-coated catheters
• Clean multiple-use non-coated catheters

The GDG indicated that there may be situations in which it would not be practical or advisable for patients to wash and reuse catheters (such as when facilities are not available or patients are unable to wash and dry catheters, or if patients are catheterised by others). Therefore, two models were constructed; they varied only in the inclusion/exclusion of clean multiple-use non-coated catheters as a comparator.

The GDG also noted that in children and young people (≤ 16 years old), symptomatic UTI can cause progressive renal scarring which may lead to renal failure later in life. Renal failure carries a high risk of mortality and morbidity, is associated with very high cost and decreased quality of life. The most recent NICE guideline for Urinary Tract Infection in Children¹⁸¹ concluded that it was not possible to estimate the true risk of renal failure as a result of childhood UTI, did not identify any quality of life values for children with UTI, and did not consider economic modelling a valid option in this population. The current GDG agreed with this decision and noted that none of the studies included in the clinical review which contained symptomatic UTI as an outcome were conducted in children. Given the uncertain risk of harm as a result of symptomatic UTI in childhood, the GDG decided to employ the precautionary principle in their approach to intermittent self-catheterisation (ISC) in children. Therefore, only single-use catheters were considered an option for ISC in children and modelling was not explicitly undertaken in this population.

Approach to modelling

Symptomatic UTI was considered the most important outcome for evaluating the efficacy of different types of intermittent catheters. The GDG also considered the costs and consequences arising from antimicrobial resistant UTIs and catheter-associated bacteraemia to be an important factor to include when assessing the downstream effects of symptomatic UTI. In the absence of any comparative clinical evidence, in the base case analysis it was assumed that urethral complications do not vary between catheter types. This assumption was explored in sensitivity analysis.

The main simplifying assumption of the model was that the probability of antibiotic resistance does not change over time. This assumption was necessary due to a lack of available data about current and historical resistance rates, the complexity of forecasting antibiotic resistance trends over time and within populations, and a lack of examples on which to base methodological approaches.⁴⁹ Different rates of resistance were explored in sensitivity analysis.

Results

This analysis found that clean multiple-use non-coated catheters are the most cost-effective type of intermittent catheter. Although gel reservoir catheters were found to be slightly more effective than clean non-coated catheters, they were associated with a much greater cost. Dividing the incremental cost by the incremental effectiveness results gives a cost-effectiveness ratio of £51,345 per QALY gained. This value far exceeds the £20,000 per QALY threshold set by NICE. By taking into account the standard error of each model input, probabilistic analysis revealed that clean multiple-use non-coated catheters are the most cost-effective option in 99.6% of model iterations.

In patients who are unable to use clean non-coated catheters, gel reservoir catheters were found to be the most cost-effective option, at approximately £3,270 per QALY gained. Compared to hydrophilic catheters, gel reservoir catheters are most cost-effective in 84.2% of model iterations.

In both scenarios, hydrophilic catheters were found to be slightly less effective than gel reservoir catheters. They are also less costly, although their incremental cost is still much greater than the cost of clean non-coated multiple-use catheters. Therefore, hydrophilic catheters are excluded from the further considerations due to extended dominance. Single-use non-coated catheters were found to be slightly less effective and more costly than multiple-use non-coated catheters. They are therefore said to be ‘dominated’ by the more effective, less costly alternatives under consideration.

Figure 2Base case analysis results (probabilistic)

Results for each subgroup are plotted on the incremental cost-effectiveness ratio axis. The non-coated multi-use catheter is the least costly strategy and has been used as the baseline comparator. Therefore, it is plotted at the axis. The slope of the line is the ICER.

Table 52Base case analysis results (probabilistic)

View in own window

Catheter	Total cost	Total QALYs	Incremental cost*	Incremental QALYs*	ICER	Probability CE
In cases where non-coated catheters can be washed and reused
Non-coated used multiple times	£11, 984	11.896	Baseline	Baseline	Baseline	99.6%
Hydrophilic	£38, 883	12.005	£26, 899	0.109	ED	0.00%
Gel reservoir	£40, 346	12.449	£28, 326	0.552	£51, 345	0.4%
Non-coated used once only	£43, 611	11.882	£31, 627	−0.014	D	0.00%
In cases where non-coated catheters cannot be washed and reused
Hydrophilic	£38, 936	12.002	Baseline	Baseline	Baseline	15.1%
Gel reservoir	£40, 391	12.446	£1, 454	0.445	£3, 270	84.2%
Non-coated used once only	£43, 642	11.879	£4, 705	−0.122	D	0.7%

The health gain to individuals using ISC is presented in terms of total and incremental QALYs. Cost is presented as total and incremental cost per catheter strategy. These values are used to calculate the ICER. Because single-use non-coated catheters are less effective and more expensive than non-coated catheters used multiple times, they are said to be dominated and are eliminated from further analysis. Similarly, hydrophilic catheters are excluded by extended dominance. QALYs = quality adjusted life years; ICER = incremental cost-effectiveness ratio; ED = extended dominated; D = dominated; CE = cost-effective at a threshold of £20,000.

*

Incremental costs and QALYs are calculated compared to the option with the lowest cost – non-coated multiple-use catheters and hydrophilic catheters, respectively.

Scenario and sensitivity analyses

Intermittent self-catheterisation (ISC) in patients with bladder dysfunction not due to spinal cord injury

A separate set of probabilities and utilities was collected in order to run a scenario analysis for patients with bladder dysfunction that is not caused by SCI. Assuming that each type of catheter exhibits the same relative efficacy in this population, the conclusion of this scenario analysis is the same as that for patients with SCI: where it is possible to wash and re-use non-coated catheters (in this population gel reservoir catheters are associated with a cost of £149, 559 per QALY gain and so do not represent an efficient use of NHS resources); however, when re-use of non-coated catheters is not an option, gel reservoir catheters represent the most cost-effective option. In both cases, single-use non-coated catheters are excluded from the analysis by dominance and hydrophilic catheters by extended dominance.

Urethral complications

When the relative risk of urethral complications associated with each type of coated catheter is reduced to zero and the cost of complications is doubled (i.e. hydrophilic catheters prevent 100% of urethral complications and those that occur with the use of other catheter types are twice as expensive as assumed in the base case), the conclusion of the analysis is unchanged. This is true regardless of whether or not multiple-use non-coated catheters are considered an option.

Antimicrobial resistance

The conclusions of the model were robust to simultaneously varying the probability of the risk of treatment failure and multidrug resistant UTI to the upper limit of each input’s 95% confidence interval. This shows that given current understanding of the scope of antibiotic resistance, multiple-use non-coated catheters are the most cost cost-effective option for ISC.

This analysis did not take into account the dynamic and extremely complex nature of antimicrobial resistance. Although the GDG sought to use the most current, relevant estimates to inform this analysis, data about the prevalence and mortality associated with antibiotic resistant UTI is limited and it is impossible to predict the future of this phenomenon. If the prevalence, clinical and economic impact of antimicrobial resistance increases beyond the extreme values used in this model, then the cost-effectiveness of clean intermittent catheterisation in this population may have to be re-visited.

Number of non-coated catheters used

The number of clean non-coated catheters used per year was varied between an average of 60 per year (average 5 per month) and 1825 per year (average 5 per day) in a threshold analysis. Clean ISC ceases to be the most cost-effective option when an average of 208 non-coated catheters is used per year; this equivalent to approximately 17.3 catheters per month or 4 per week.

Interpretation and limitations

This analysis combines the best available evidence about the costs and consequences of each type of catheter used for intermittent catheterisation. Based on the results of the model, we can conclude that the small decrease in symptomatic infections associated with single-use gel reservoir and hydrophilic catheters is not enough to justify the large increase in the cost of these catheters compared to multiple-use non-coated catheters. As a result, clean multiple-use non-coated catheters represent the most cost-effective type of catheter for ISC. This conclusion was robust to a wide range of sensitivity analyses, including the increased probability of urethral complications that may be associated with the use of non-coated catheters. However, multiple-use non-coated catheters cease to be the most cost-effective choice when patients use an average of more than two catheters per day. Compliance and behaviour are therefore important factors for healthcare workers to consider when prescribing an ISC regime.

Healthcare workers must also consider other patient-specific situations when deciding which catheter to prescribe. Under the current decision rule, the recommended treatment is identified as that with the highest ICER that falls below the cost-effectiveness threshold. Preferences are incorporated into the cost-utility analysis through the values that are attached to each health state; these values represent the average weight attached to each health state by the general population and are assumed to be independent of factors related to the health care process.

The use of societal values creates the potential for conflict where individual patients hold a strong preference for a particular treatment that is not reflected in the decision made at the societal level.²⁶ It has been suggested that one way to incorporate individual patient preference into cost-effectiveness decisions would be to adopt a two-part decision process which gives the patient the choice of the most cost-effective treatment plus all cheaper options.⁷⁷

Of the five RCTs included in our review of clinical efficacy, three included a measure of patient preference and comfort; none found any difference between catheter types. Nevertheless, it is still possible that patients may find one type of catheter more comfortable or easier to use than another and therefore derive a benefit from the catheter that is not captured in the model.⁷⁶ When deciding between gel reservoir and hydrophilic catheters for patients who cannot use multiple-use non-coated catheters, the GDG did not wish to force the consumption of more costly gel reservoir catheters. If a patient has a strong preference for hydrophilic catheters then the GDG agreed that they should be able to choose this less costly option.

10.5.2. Review question – long-term indwelling catheters

What is the clinical and cost effectiveness of different types of long-term indwelling urinary catheters (non-coated silicone, hydrophilic coated, or silver or antimicrobial coated/impregnated) on urinary tract infections, bacteraemia, frequency of catheter change, encrustations and blockages, mortality, and patient preference?

10.5.3. Is one catheter better than another?

There is some evidence that the balloon material on all silicone Foley catheters has a greater tendency to “cuff” on deflation than latex catheters, particularly when used suprapubically. Cuffing can cause distress and injury to patients when the catheter is removed.¹⁶⁵ Our systematic review showed that smaller gauge catheters (12–14 Ch) with a 10 ml balloon minimise urethral trauma, mucosal irritation and residual urine in the bladder, all factors which predispose to catheter-associated infection.^220,229 A non-systematic review of the literature confirmed this.²⁴⁸ For suprapubic catheterisation, a 16 Ch gauge catheter is usually preferable to avoid blockage.¹⁶² Where there is no difference in the quality of the catheter, the least expensive option should be used.⁷³

One study²⁸⁰ identified by our systematic review compared the use of catheter valves with a standard drainage system and found no significant difference in urinary tract infection but a patient preference for the catheter valve. The Medical Device Agency (now Medicines and Healthcare products Regulatory Agency) suggests patients need to be assessed for their mental acuity, manual dexterity, clothing preferences and use of night drainage bags when considering using catheter valves.¹⁶⁴

10.6.1. Review question

What is the most clinically and cost effective technique (aseptic technique, non-touch, aseptic non touch technique or a clean technique) when handling long-term urinary catheters to reduce colony forming units, urinary tract infections, compliance, MRSA or C. diff reduction and mortality?

10.7.1. Catheterisation is a skilled procedure

Principles of good practice, clinical guidance^270,284 and expert opinion^{74,75,131,141,247} agree that urinary catheters must be inserted using sterile equipment and an aseptic technique. Expert opinion indicates that there is no advantage in using antiseptic preparations for cleansing the urethral meatus prior to catheter insertion.^93,139 Urethral trauma and discomfort will be minimised by using an appropriate sterile, single-use lubricant or anaesthetic gel. The insertion of urinary catheters by healthcare workers who are competent in the procedure will minimise trauma, discomfort and the potential for catheter-associated infection.^{75,93,141,270}

With regard to self-catheterisation, our systematic review found that in a study examining the safety of clean versus sterile intermittent catheterisation in male adults aged 36–96 years, no significant differences were found in infection rates, time to first infection or number of episodes.⁷⁹ A systematic review identified three controlled trials regarding the benefits of sterile or “non-touch techniques” for intermittent catheterisation vs. conventional clean intermittent catheterisation.²³⁹ Data “neither supports nor refutes the need to utilize sterile, as opposed to clean, intermittent catheterisation.” Economic analysis suggests that clean intermittent catheterisation is unlikely to lead to additional infections and the additional cost of sterile catheterisation is unlikely to be justified.^79,274

10.8.1. Leave the closed system alone!

Maintaining a sterile, continuously closed urinary drainage system is central to the prevention of catheter-associated infection.^{75,101,141,261,270,284} The risk of infection reduced from 97% with an open system to 8–15% when a sterile closed system was employed as standard practice.^93,100,139 However, breaches in the closed system such as unnecessary emptying of the urinary drainage bag or taking a urine sample increase the risk of catheter-related infection and should be avoided.^139,208,270 Hands must be decontaminated and healthcare workers should wear clean, non-sterile gloves before manipulation.

Reflux of urine is associated with infection and, consequently, best practice suggests catheters are secured to avoid trauma and drainage bags should be positioned in a way that prevents back-flow of urine.^75,270 Expert opinion also recommends that urinary drainage bags should be supported in such a way that prevents contact with the floor.¹³⁹ For night drainage, a link system should be used to maintain the original closed system, i.e., a bag attached to the end of the day system.²⁴⁹

Drainable urinary drainage bags should be changed in line with the manufacturer’s recommendations, generally every 5–7 days, or sooner if clinically indicated, e.g. malodorous or damaged. Bags that are non-drainable should be used once, e.g., overnight, and emptied before disposal.

10.8.2. Appropriate maintenance minimises infections

10.9.1. Review question

What is the clinical and cost effectiveness of bladder instillations or washouts on reduction of catheter associated symptomatic urinary tract infections and encrustations and blockages?

10.9.2. Changing catheters

There is no definitive evidence as to the optimal interval for changing catheters in patients undergoing long-term urinary drainage via either the urethral or suprapubic route. Our search identified a study which suggested that a higher rate of infection was associated with frequent catheter changes, though evidence is not definitive.²⁷⁷ Expert opinion suggests changing the catheter according to the clinical needs of the patient or as recommended by the catheter manufacturer (usually every 12 weeks).^270,284 Our systematic review identified a study that showed if catheter blockage occurs within a shorter interval, catheters should be changed more frequently to avert a future clinical crisis.⁹⁷ An economic analysis suggested that there may be a cost saving in changing a catheter at six weeks when there is an increased likelihood of blockage (>50%).¹⁸⁵

10.10.1. Review question

In patients with long-term urinary catheters (more than 28 days), what is the clinical and cost effectiveness of prophylactic antibiotics (single dose or short course) during catheter change on reduction of urinary tract infections?

10.10.2. Cost-effectiveness evidence

No cost-effectiveness evidence was identified. No cost effectiveness evidence was identified in the previous 2003 guideline.

From an economic perspective, questions surrounding the use of antibiotic prophylaxis are very complex. A recent Health Technology Assessment performed a literature search in order to develop a conceptual evaluative framework for the economic evaluation of policies against MRSA⁴⁹. Many of the considerations discussed within this review were relevant to the current question and provided a useful background for GDG discussions related to the cost-effectiveness of antibiotic prophylaxis for changing long-term indwelling urethral catheters.

The GDG were also presented with current UK antibiotic and infection-related costs (see economic model in Appendix J).

10.10.3. Evidence statements

Assessing the need for catheterisation

Epidemiological studies of the prevalence and incidence of bacteriuria/clinical urinary tract infection during long-term catheterisation in different populations and different care settings. These should at least encompass the predominant populations; older people and those with neurological deficits in both institutional and domiciliary settings. There needs to be clear definition of the ‘cases’ and the populations from which they are drawn.

Catheter drainage options

Randomised controlled trials of different approaches to urinary drainage. These should compare urethral indwelling catheterisation with and without a drainage bag (i.e., a valve); urethral intermittent catheterisation; suprapubic catheterisation; penile sheath drainage and incontinence pads in appropriate populations. Outcome measures need to include rates of bacteriuria/clinical UTI; tissue damage; patient/carer satisfaction; and cost-benefit.

Randomised controlled trials of the efficacy of antimicrobial impregnated urethral catheters for long-term use.

Catheter maintenance

Randomised controlled trials of strategies to reduce/prevent/manage encrustation and blockage. These need to determine whether catheter maintenance solutions (washouts/installations) are effective in reducing encrustation; blockage; urethral trauma; frequency of catheter replacement; and interventions/visits by healthcare practitioners. The rates of these complications when catheter valves are used in place of drainage bags also needs to be compared.

Cohort studies to determine whether monitoring of urinary pH can be used to predict time to blockage. These need to be undertaken in defined and representative groups.

Randomised controlled trials to establish the optimum time interval between changing equipment. There is a particular need to determine whether the frequency of changing leg bags or catheter valves influences the rates of bacteriuria/clinical UTI.

3. For patients performing intermittent self-catheterisation over the long term, what is the clinical and cost effectiveness of single-use non-coated versus single-use hydrophilic versus single-use gel reservoir versus reusable non-coated catheters with regard to the following outcomes: symptomatic urinary tract infections, urinary tract infection-associated bacteraemia, mortality, patient comfort and preference, quality of life, and clinical symptoms of urethral damage?

4. For patients using a long-term indwelling urinary catheter, what is the clinical and cost effectiveness of impregnated versus hydrophilic versus silicone catheters in reducing symptomatic urinary tract infections, encrustations and/or blockages?

5. When recatheterising patients who have a long-term indwelling urinary catheter, what is the clinical and cost effectiveness of single-dose antibiotic prophylaxis in reducing symptomatic urinary tract infections in patients with a history of urinary tract infections associated with catheter change?