Pressure redistributing devices

National Clinical Guideline Centre (UK)

6Pressure redistributing devices

6.1. Introduction

Pressure relieving and redistributing devices are widely accepted methods of both preventing and treating pressure ulcers. A vast range of devices, including different types of mattresses, overlays, cushions and seating, are available which vary considerably in both cost and mechanism. Generally, these devices work by reducing pressure, friction or shearing forces and may be unpowered and considered ‘low-tech’ or ‘static’, or powered devices which are ‘high-tech’.

The selection of device by the healthcare professional is likely to depend upon the person's mobility, the result of skin assessment, the severity and site of the pressure ulcer, weight, staff availability and skill. The choice of a pressure redistributing device by a healthcare professional should also account for a person's wishes and tolerability of the device.

It is generally accepted that these devices should be used in conjunction with other treatment strategies such as repositioning and management of the pressure ulcer (for example, by use of an appropriate dressing).

The GDG were interested in what the most clinically and cost effective pressure redistributing devices are for the management of pressure ulcers.

Recommendations on the use of pressure redistributing devices for the prevention of pressure ulcers can be found in part 1 of the guideline.

6.2. Review question: What are the most clinically and cost effective pressure redistributing devices for the management of pressure ulcers?

For full details see review protocol in Appendix D.

6.2.1. Clinical evidence (adults)

A Cochrane Review looking at support surfaces for treating pressure ulcers was retrieved from the search and this was used as the basis for our review. It included 17 randomised trials⁶^,²⁹^,⁴⁴^,⁵¹^,⁵⁵^-⁵⁷^,⁵⁹^,⁷³^,⁸⁹^,¹²³^,¹²⁶^,¹³⁶^,¹⁴⁵^,¹⁶¹^,¹⁶²^,¹⁸⁰ Two additional randomised trials were found via systematic searches and were used to update the Cochrane review.³⁹^,¹⁰⁸ Evidence from the RCTs are summarised in the clinical GRADE evidence profile below. See also the study selection flow chart in Appendix D, forest plots in Appendix I, study evidence tables in Appendix G and exclusion list in Appendix J.

Various types of devices were used to redistribute pressure, and for the purposes of the Cochrane review, these were categorised as ‘low-tech’ (non-powered), constant low pressure (CLP), high-tech and other devices.

The ‘low-tech’ CLP devices included:

Standard foam mattresses
Alternative foam mattresses or overlays
Gel-filled mattresses or overlays
Fibre-filled mattresses or overlays
Air-filled mattresses or overlays
Water-filled mattresses or overlays
Bead-filled mattresses or overlays
Sheepskins

The high-tech support surfaces included:

Alternating-pressure mattresses or overlays
Air-fluidised beds
Low-air-loss beds

The other support surfaces included:

Turning beds or frames
Operating table overlays
Wheelchair cushions

The Cochrane Review notes that this classification has since been updated by the National Pressure Ulcer Advisory Panel (EPUAP & NPUAP 2009) and will be considered in future updates of their review.

Table

Summary of included studies.

6.2.1.1. Low tech constant pressure devices

Table 42

Clinical evidence profile: water mattress overlay versus low-tech mattress.

Table 43

Clinical evidence profile: 3 dimensional macroporous overlay versus gel overlay.

6.2.1.2. High-tech pressure devices

Table 44

Clinical evidence profile: low-air-loss bed versus low-tech foam mattress overlay.

Table 45

Clinical evidence profile: low-air-loss bed versus low-air-loss overlay.

Table 46

Clinical evidence profile: air-fluidised therapy (AFT) versus standard or conventional therapies.

Table 47

Clinical evidence profile: alternating-pressure mattress versus alternating-pressure mattress.

Table 48

Clinical evidence profile: alternating-pressure mattress overlay versus alternating-pressure mattress.

Table 49

Clinical evidence profile: air-filled devices versus alternating pressure mattress.

Table 50

Clinical evidence profile: alternating-pressure cushion versus dry flotation cushion.

Table 51

Clinical evidence profile: profiling bed versus foam mattress.

Table 52

Clinical evidence profile: constant force mattress versus low-air-loss mattress.

Table 53

Clinical evidence profile: wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushion.

6.2.1. Economic evidence (adults)

Published literature

One study was included with a relevant comparison.⁶¹ This is summarised in the economic evidence profile below (Table 54). See also the study selection flow chart in Appendix D and study evidence tables in Appendix H.

Table 54

Economic evidence profile: alternating pressure overlays verses alternating pressure mattress replacements verses high specification foam mattresses.

One study that met the inclusion criteria was selectively excluded¹⁹⁶ – this is summarised in Appendix H, with reasons for exclusion given.

One additional study was found which included devices for the management of pressure ulcers as part of a more complex management strategy.¹⁹³ This study was not included as the cost-effectiveness of the strategy as a whole is evaluated, and does not provide information on the cost-effectiveness of the device alone.

Unit costs

Unit costs were presented to aid consideration of cost effectiveness (see Table 55).

Table 55

Unit costs.

Note - these prices have been obtained directly from manufacturers, and represent the list price for the NHS. It is acknowledged that prices vary locally; therefore these prices are illustrative only. The devices included in the table are those identified by GDG members as being commonly used, and should not be interpreted as recommended devices.

6.2.2. Clinical evidence (neonates, infants, children and young people)

No RCTs or cohort studies were identified. Recommendations were developed using a modified Delphi consensus technique. Further details can be found in Appendix N.

6.2.3. Economic evidence (neonates, infants, children and young people)

Published literature

No relevant economic evaluations were identified.

Economic considerations

In the absence of economic evidence, the GDG considered relevant UK NHS unit costs of various mattresses and overlays (see Table 56) These were considered alongside clinical evidence obtained from the Delphi consensus panel to inform qualitative judgement about cost-effectiveness.

Table 56

Unit costs.

Note: the costs above are included for illustrative purposes only and should not be interpreted as recommendations in favour of these particular devices. These are list prices only and local prices may vary.

6.2.4. Evidence statements

6.2.4.1. Clinical (adults)

6.2.4.1.1. Water mattress overlay versus low-tech mattress

One study (n=120) showed there may be no clinical difference between a water mattress overlay and a low-tech mattress for the proportion of people with pressure ulcers completely healed within trial period, but the direction of the estimate of effect favoured the low-tech mattress (very low quality).
One study (n=120) showed there may be a clinical benefit for a water mattress overlay compared to a low tech mattress for reducing reduction in pain. No estimate of precision could be derived (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of ulcers
- Reduction in size and/or volume of pressure ulcer
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.1.2. 3-D macroporous overlay versus gel overlay

One study (n=72) showed there may be no clinical difference between a 3-D macroporous overlay and a gel overlay for the proportion of people with pressure ulcers completely healed. The direction of the estimate of effect favoured the gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay may be more clinically effective for reducing mortality (all-cause) when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay is potentially more clinically effective for reducing suspension due to worsening of pressure ulcers when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay may be more clinically effective for reducing suspension due to intolerance when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay is potentially more clinically effective for reducing unchanged/worsened pressure ulcers when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay is potentially more clinically effective for improving pressure ulcers when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay may be more clinically effective for increased comfort when compared to a gel overlay (very low quality).
One study (n=72) showed a 3-D macroporous overlay is potentially more clinically effective for reducing discomfort when compared to a gel overlay (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing (time to event data)
- Rate of healing (continuous data)
- Rate of change in size of ulcer (absolute and relative) (continuous data)
- Reduction in size of ulcer and volume of ulcer
- Pain (pressure ulcer related)
- Time in hospital or NHS care (continuous data)
- Side effects
- Health-related quality of life (continuous data)

6.2.4.1.3. Low-air loss-bed versus low-tech foam mattress overlay

One study (n=84) showed a low-air-loss bed is potentially more clinically effective for the proportion of people with pressure ulcers completely healed within the trial period when compared to a low-tech foam mattress overlay (very low quality).
One study (n=49) showed there may be no clinical harm between a low-air-loss bed and a low tech foam mattress overlay for the proportion of people with pressure ulcers completely healed within the trial period, but the direction of the estimate of effect could favour the low-tech foam mattress overlay (very low quality).
Two studies (n=133) showed a low-air-loss bed is potentially more clinically effective for the proportion of people with pressure ulcers completely healed within the trial period at when compared to a low-tech foam mattress overlay (very low quality).
One study (n=49) showed there may be no clinical difference between a low-air-loss bed and a low-tech foam mattress overlay for pressure ulcers reduced by 1 grade or more, including completely healed (very low quality).
One study (n=84) reported there is potentially a clinical benefit for a low-air-loss bed compared to a low-tech foam mattress overlay for rate of ulcer healing. No estimate of precision could be derived (very low quality).
One study (n=48) reported a mean difference for change in pressure ulcer size for stage 2 pressure ulcers was 2 higher (0.73 to 3.27) in a low-air-loss bed compared with a low-tech foam mattress overlay. The clinical importance is unknown (very low quality).
One study (n=29) reported a mean difference for change in pressure ulcer size for stage 3 and stage 4 pressure ulcers was 24.7 higher (20.37 to 29.03)in a low-air-loss bed compared with a low-tech foam mattress overlay. The clinical importance is unknown (very low quality).
One study (n=39) showed there is no clinical benefit of a low-air-loss bed for patient acceptability (mean comfort score) when compared with a low-tech foam mattress overlay (low quality).
One study (n=84) showed there may be a clinical harm for a low-air-loss bed compared to a low-tech foam mattress overlay for all-cause mortality (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Pain (prssure ulcer related)
- Time in hospital or NHS care
- Side effects
- Health-related quality of life

6.2.4.1.4. Low-air-loss bed versus low-air-loss overlay

One study (n=93) showed there may be a clinical benefit for a low-air-loss bed compared to a low-air-loss overlay for reduction in size of pressure ulcer (median change in pressure ulcer area). No estimate of precision could be derived (very low quality).
One study (n=93) showed there may be a clinical benefit for a low-air-loss bed compared to a low-air-loss overlay for reduction in size of pressure ulcer (mean change in pressure ulcer surface area). No estimate of precision could be derived (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of ulcers
- Proportion of completely healed within trial period
- Pain (pressure ulcer-related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.1.5. Air-fluidised therapy (AFT) versus standard/conventional therapies

One study (n=65) showed there may be no clinical difference between an air-fluidised therapy and a standard/conventional therapy for proportion of people with 50% reduction in total surface area (very low quality).
One study (n=35) showed air-fluidised therapy is potentially more clinically effective for proportion with improvement in pressure ulcers when compared to standard/conventional therapy (very low quality).
One study (n=65) showed air-fluidised therapy is potentially more clinically effective for proportion with improvement in pressure ulcers when compared to standard/conventional therapy (low quality).
Two studies (n=100) showed air-fluidised therapy is potentially more clinically effective for proportion with improvement in pressure ulcers when compared to standard/conventional therapy (low quality).
One study (n=40) showed there may be a clinical benefit for air-fluidised therapy compared to standard/conventional therapy for change in mean ulcer area (stage 2 or 3 ulcers). No estimate of precision could be derived (very low quality).
One study (n=65) showed there may be a clinical benefit for air-fluidised therapy compared to standard/conventional therapy for change in total pressure ulcer surface area (very low quality).
One study (n=18) showed air-fluidised therapy is more clinically effective for patient acceptability (patient satisfaction) when compared to standard/conventional therapy (low quality).
One study (n=27) showed air-fluidised therapy is potentially more clinically effective for patient acceptability (more people experiencing an increase in comfort) when compared to standard/conventional therapy (very low quality).
One study (n=27) showed there may be a clinical benefit for air-fluidised therapy compared to standard/conventional therapy for patient acceptability (fewer people experiencing a reduction in comfort) (very low quality).
One study (n=97) showed there may be a clinical benefit for air-fluidised therapy compared to standard/conventional therapy for time in hospital (very low quality).
One study (n=27) reported there may be no clinical difference between air-fluidised therapy and standard/conventional therapy length of stay in hospital after randomisation, but the direction of the estimate of effect could favour standard care. No estimate of precision could be derived (very low quality).
One study (n=27) showed air-fluidised therapy is potentially more clinically effective for pain (more people experiencing a reduction in pain) when compared to standard or conventional therapy (very low quality).
One study (n=27) showed there may be a clinical benefit for air-fluidised therapy compared to standard/conventional therapy for pain (fewer people experiencing an increase in pain) (very low quality).
One study (n=112) showed air-fluidised therapy is potentially more clinically effective at reducing all-cause mortality when compared to standard/conventional therapy (very low quality).
Two studies (n=177) showed there may be no clinical difference between an air-fluidised therapy and standard/conventional therapy at reducing all-cause mortality, the direction of the estimate of effect favoured air-fluidised therapy (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of pressure ulcers
- Proportion of completely healed within trial period
- Side effects
- Health-related quality of life

6.2.4.1.6. Alternating-pressure mattress (Nimbus 1) versus. alternating-pressure mattress (Pegasus Airwave)

One study (n=30) showed an alternating-pressure mattress (Nimbus 1) is potentially more clinically effective for proportion of people with pressure ulcers completely healed (grade 2 and above) when compared to an alternating-pressure mattress (Pegasus Airwave) (very low quality).
One study (n=30) showed there may be a clinical harm for a alternating-pressure mattress (Nimbus 1) compared to an alternating-pressure mattress (Pegasus Airwave) for proportion of people with decrease in pressure ulcer size (grade 2 and above) (very low quality).
One study (n=30) showed an alternating-pressure mattress (Nimbus 1) may be more clinically effective than an alternating-pressure mattress (Pegasus Airwave) for proportion of people with increase in pressure ulcer size (grade 2 and above) (very low quality).
One study (n=41) reported medians an alternating-pressure mattress (Nimbus 1) and an alternating-pressure mattress (Pegasus Airwave) for patient acceptability (comfort). The median for both interventions was 8/10. No estimate of effect or precision could be derived (very low quality).
One study (n=41) showed there may be no clinical difference between an alternating-pressure mattress (Nimbus 1)and an alternating-pressure mattress (Pegasus Airwave) for mortality, but the direction of the estimate of effect could favour either intervention (very low quality).
One study (n=41) reported there may be no clinical difference between an alternating-pressure mattress (Nimbus 1) and an alternating-pressure mattress (Pegasus Airwave) for rate of reduction in pressure ulcer surface area (grade 2 and above), but the direction of the estimate of effect could favour either intervention (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Side effects
- Health-related quality of life

6.2.4.1.7. Alternating-pressure mattress (Nimbus 3) with cushion (Aura) and 4-hourly turning versus an alternating-pressure mattress (Pegasus Cairwave Therapy System) with seating cushion (Proactive 2) and 8-hourly turning

One study (n=141) showed no clinical difference between an alternating-pressure mattress (Nimbus 3) with cushion (Aura) and 4-hourly turning for proportion of people with pressure ulcers completely healed (grade 2 and above) when compared with an alternating-pressure mattress (Pegasus Cairwave Therapy System) with seating cushion (Proactive 2) and 8-hourly turning, the direction of the estimate of effect favours the 4-hourly turning (moderate quality).
One study (n=112) reported the mean difference for time in hospital for an alternating-pressure mattress (Nimbus 3) with cushion (Aura) and 4-hourly turning, and an alternating-pressure mattress (Pegasus Cairwave Therapy System) with seating cushion (Proactive 2) and 8-hourly turning. The mean for the alternating-pressure mattress (Nimbus 3) was 21.6 days and 21.7 days for the alternating-pressure mattress (Pegasus Cairwave Therapy System). No estimate of effect or precision could be derived (very low quality).
One study (n=141) showed an alternating-pressure mattress (Nimbus 3) with cushion (Aura) and 4-hourly turning is potentially more clinically harmful for mortality when compared to an alternating-pressure mattress (Pegasus Cairwave Therapy System) with seating (Proactive 2)cushion and 8-hourly turning (low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of ulcers
- Reduction in size and/or volume of ulcer
- Pain (pressure ulcer related)
- Patient acceptability
- Side effects
- Health-related quality of life

6.2.4.1.8. Alternating-pressure mattress (Nimbus 3) versus alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)

One study (n=32) reported medians for patient acceptability (comfort) for people in hospital for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). The median for the alternating-pressure mattress (Nimbus 3) was 5 (very comfortable) and 4 (comfortable) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of effect or precision could be derived (very low quality).
One study (n=32) reported medians for patient acceptability (comfort)for elderly people in a hospital or nursing home for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). The median for the alternating-pressure mattress (Nimbus 3) was 5 (very comfortable) and 4 (comfortable) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of effect or precision could be derived (very low quality).
One study (n=32) showed there may be a clinical harm for an alternating-pressure mattress (Nimbus 3) compared to a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro) for mortality (very low quality).
One study (n=32) reported medians for absolute reduction in wound surface area per day (grade 2 and above) for people in hospital for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). The median for the Nimbus 3 alternating-pressure mattress was 0.12cm² (range 0 to 0.21cm2) and 0.08cm²(range 0.04 to 0.33cm²) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of precision could be derived (very low quality).
One study (n=32) reported medians for relative reduction in wound surface area (grade 2 and above) for people in hospital for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro),. The median for the Nimbus 3 alternating-pressure mattress was 2.44% (range 0 to 7.14%) and 1.34% (range 1.11 to 2.88%) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of precision could be derived (very low quality).
One study (n=32) reported medians for absolute reduction in wound surface area per day (grade 2 and above) for elderly people in hospital or a nursing home for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). The median for the Nimbus 3 alternating-pressure mattress was 0.11cm² (range 0.04 to 0.41cm²) and 0.05cm²(range 0 to 0.48cm²) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of precision could be derived (very low quality).
One study (n=32) reported medians for relative reduction in wound surface area (grade 2 and above)for elderly people in hospital or a nursing home for an alternating-pressure mattress (Nimbus 3) and a different alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). The median for the alternating-pressure mattress (Nimbus 3) was 1.57% (range 0.45 to 5%) and 0.99% (range 0 to 2.54%) for alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro). No estimate of precision could be derived (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Proportion of completely healed within trial period
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Side effects
- Health-related quality of life

6.2.4.1.9. Alternating-pressure mattress overlay versus alternating-pressure mattress

One study (n=113) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for proportion of people with pressure ulcers completely healed (grade 2 and above), but the direction of the estimate of effect could favour the alternating-pressure mattress (very low quality).
One study (n=158) showed there is no clinical benefit of an alternating-pressure mattress overlay for pressure ulcer improvement (grade 1 or 2) when compared with an alternating-pressure mattress (low quality).
One study (n=158) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for worsening of pressure ulcers (grade 1 or 2), but the direction of the estimate of effect could favour either intervention (very low quality).
One study (n=1971) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (number requesting changes for comfort or other device-related reason) (low quality).
One study (n=158) reported the mean for time in hospital for an alternating-pressure mattress overlay and an alternating-pressure mattress. The mean for an alternating-pressure mattress was 22.17 days and 20.02 days for an alternating-pressure mattress. No estimate of precision could be derived (very low quality).
One study (n=113) showed an alternating-pressure mattress overlay is potentially more clinically harmful for mortality when compared to an alternating-pressure mattress (very low quality).
One study (n=113) reported medians for an alternating-pressure mattress and an alternating-pressure mattress overlay for time to ulcer heading. The median for both interventions was 20 days. No estimate of precision could be derived (very low quality).
One study (n=69) showed there is potentially no clinical difference between an alternating-pressure mattress and an alternating-pressure mattress overlay for absolute change in pressure ulcer surface area (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people with negative comments on mattress motion). The direction of the estimate of effect favoured the air-filled devices (very low quality).
One study (n=1820) showed there is no clinical difference between an alternating-pressure mattress overlay for patient acceptability (proportion of people with positive comments on mattress motion) when compared with an alternating-pressure mattress the direction of the estimate of effect favoured the alternating-pressure mattress (low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people with negative comments on getting into/out of bed). The direction of the estimate of effect favoured the alternating-pressure mattress (low quality).
One study (n=1820) showed there is no clinical difference between an alternating-pressure mattress overlay for patient acceptability (proportion of people with negative comments on movement in bed) when compared with an alternating-pressure mattress. The direction of the estimate of effect favoured the alternating-pressure mattress (low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people with positive comments on movement in bed). The direction of the estimate of effect favoured the alternating-pressure mattress (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people commenting on temperature as hot orwarm). The direction of the estimate of effect favoured the alternating-pressure mattress (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people commenting on temperature as sweaty or sticky). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people commenting on temperature as cold/cool), but the direction of the estimate of effect favoured either intervention (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people reporting the mattress not working or not working properly). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people commenting that it was hard to tuck the sheet under, sheets come off or gather or the mattress cover slips). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people commenting the mattress/bed was too high). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people reporting the mattress as slippery). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people reporting the mattress as too softor edges soft or sloping). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for patient acceptability (proportion of people reporting they were not able to use backrest). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there is potentially no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for side effects (mattress related fall). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for side effects (mattress-related suspected contact dermatitis). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for side effects (mattress-related climbed over/fell through cot sides). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
One study (n=1820) showed there may be no clinical difference between an alternating-pressure mattress overlay and an alternating-pressure mattress for side effects (mattress deflation during transfer). The direction of the estimate of effect favoured the alternating-pressure mattress overlay (very low quality).
No evidence was found for the following outcomes:
- Rate of reduction in size of pressure ulcers
- Pain (pressure ulcer related)
- Side effects
- Health-related quality of life

6.2.4.1.10. Air-filled devices versus alternating-pressure mattress

One study (n=60) showed there may be a clinical benefit for an air filled device compared to an alternating-pressure mattress for the proportion of people with pressure ulcers completely healed (grade 2 and above) (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of pressure ulcers
- Reduction in size and/or volume of ulcer
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.1.11. Alternating-pressure cushion versus dry flotation cushion

One study (n=25) showed there may be a clinical benefit for a dry flotation cushion for proportion of people with pressure ulcers completely healed (grade 2 and above) compared to an alternating-pressure cushion (very low quality).
One study (n=25) showed there may be a clinical benefit for a dry flotation cushion for mortality when compared to an alternating-pressure cushion (very low quality).
One study (n=25) showed there may be a clinical benefit for a dry flotation cushion for rate of ulcer healing (area of ulcer) when compared to an alternating-pressure cushion (very low quality).
One study (n=25) showed there may be a clinical benefit for an alternating-pressure cushion compared to a dry flotation cushion for rate of ulcer healing (volume of ulcer) (very low quality).
One study (n=25) showed an alternating-pressure cushion is potentially more clinically effective for percentage change in pressure ulcer area per day when compared to a dry flotation cushion (very low quality).
One study (n=25) showed there may be a clinical benefit for an alternating-pressure cushion compared to a dry flotation cushion for percentage change in volume of pressure ulcer per day (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Reduction in size and/or volume of pressure ulcer
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Health-related quality of life

6.2.4.1.12. Profiling bed versus foam mattress

One study (n=14) showed there is potentially a clinical benefit for a profiling bed compared to a foam mattress for the proportion of people with completely healed pressure ulcers (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Rate of reduction in size of pressure ulcers
- Reduction in size and/or volume of pressure ulcer
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.1.13. Constant force mattress versus low-air-loss mattress

One study (n=18) showed a constant force mattress is potentially more clinically effective for the percentage rate of pressure ulcer closure per week when compared to a low-air-loss mattress (very low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Reduction in size and/or volume of pressure ulcer
- Proportion of completely healed within trial period
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.1.14. Wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushion

One study (n=44) showed a wheelchair cushion equipped with individualised cyclic pressure-relief protocol is more clinically effective for pressure ulcer closure rate (area) when compared to a standard wheelchair cushion (low quality).
One study (n=44) showed a wheelchair cushion equipped with individualised cyclic pressure-relief protocol is more clinically effective for PUSH score improvement when compared to a standard wheelchair cushion (low quality).
One study (n=44) showed a wheelchair cushion equipped with individualised cyclic pressure-relief protocol is more clinically effective for percentage surface area reduction improvement when compared to a standard wheelchair cushion (low quality).
One study (n=44) showed a wheelchair cushion equipped with individualised cyclic pressure-relief protocol is more clinically effective for PUSH score improvement when compared to a standard wheelchair cushion (low quality).
One study (n=44) showed a wheelchair cushion equipped with individualised cyclic pressure-relief protocol is more clinically effective for percentage PUSH score improvement when compared to a standard wheelchair cushion (low quality).
No evidence was found for the following outcomes:
- Time to complete healing
- Proportion of completely healed within trial period
- Pain (pressure ulcer related)
- Time in hospital or NHS care
- Patient acceptability
- Side effects
- Mortality (all cause)
- Health-related quality of life

6.2.4.2. Economic (adults)

One cost-utility analysis found that alternating pressure mattress replacements dominate alternating pressure overlays and standard care (high specification foam mattresses) in the treatment of pressure ulcers. This study was assessed to be directly applicable with potentially serious limitations.

6.2.4.3. Clinical (neonates, infants, children and young people)

No evidence was identified.

6.2.4.4. Economic (neonates, infants, children and young people)

No evidence was identified.

6.3. Recommendations and link to evidence

6.3.1. Adults

Table

Use high-specification foam mattresses for adults with a pressure ulcer. If this is not sufficient to redistribute pressure, consider the use of a dynamic support surface. Do not use standard-specification foam mattresses for adults with a pressure ulcer. (more...)

6.3.2. Neonates, infants, children and young people

Table

Use a high-specification cot or bed mattress or overlay for all neonates, infants, children and young people with a pressure ulcer.

Table

If pressure on the affected area cannot be adequately relieved by other means (such as repositioning), consider a dynamic support surface, appropriate to the size and weight of the child or young person with a pressure ulcer, if this can be tolerated. (more...)

Table

Consider using specialist support surfaces (including dynamic support surfaces where appropriate) for neonates, infants, children and young people with pressure ulcers, taking into account their current pressure ulcer risk and mobility. Tailor the support (more...)

Publication Details

Copyright

Publisher

National Institute for Health and Care Excellence (NICE), London

NLM Citation

National Clinical Guideline Centre (UK). The Prevention and Management of Pressure Ulcers in Primary and Secondary Care. London: National Institute for Health and Care Excellence (NICE); 2014 Apr. (NICE Clinical Guidelines, No. 179.) 6, Pressure redistributing devices.

Summary of included studies

Study	Intervention/comparison	Population	Outcomes	Length of study
Allman 1987⁶	Air-fluidised therapy (CLINITRON) repositioned every 4 hours versus conventional treatment (including 2-hourly turns, heel and elbow protectors, alternating-pressure mattresses).	People undergoing surgery aged 18 or over with pressure ulcers of all stages. Graded using the Shea staging system.	Median change in total surface area of ulcers; improvement in condition of pressure ulcer; pain response.	Mean 13 days follow-up (range 4-77 days).
Branom 2001²⁹	PressureGuard CFT (constant force therapy) (non-powered mattress) versus LAL mattress.	People in hospital from long term and sub-acute care centre specialising in ventilator-dependent adults and those with extensive wound care needs. Bedridden adults with a pressure ulcer at grade 3 or 4 on trunk or pelvis (classification system not reported).	Meeting the goals of pressure ulcer treatment as determined by medical team (including wound closure, maintenance of condition and preparation for flap). The rate of pressure ulcer healing over 8 weeks.	8-week follow-up.
Caley 1994³⁸	LAL bed (Monarch, Mediscus) versus LAL overlay.	People in acute care with existing pressure ulcers.	Median change in pressure ulcer area	Average 24-day follow-up.
Cassino 2013³⁹	Three-dimensional overlay (AIARTEX), made of 3-D macro-porous material vs dry viscoelastic polyurethane polymer overlay (AKTON)	Long-term care patients	Complete healing; improved; unchanged/worsened; suspension due to worsening; suspension due to intolerance; mortality; comfort	12 weeks
Clark 1998⁴⁴	ProActive 2 cushion (Pegasus) (cushion for day chairs and wheelchairs, seating automatically adjusts to an individual's weight) versus ROHO cushion (dry flotation system). All individuals had a Pegasus Airwave System in bed.	Elderly adults in 2 acute care hospitals and 2 nursing homes. Grade 2 pressure ulcers or above, classification system not reported.	Number of pressure ulcers healed completely; rate of healing (cm²/day); rate of healing (cm³/day).	Average 58.6 days (ProActive and 43.73 days (ROHO).
Day 1993⁵¹	Air suspension bed (Therapulse, Kinetic concepts); foam mattress overlay (Geomatt, SpanAmerica). Wound care standardised for 2 groups.	Hospitalised adults with existing grade 2-4 pressure ulcers, graded using the NPUAP classification system.	Mean pressure ulcer size (initial minus end) divided into grade 2 and grade 3/4 ulcers; mean comfort scores.	7-day follow-up.
Devine 1995⁵⁵	Alternating-pressure mattress (Nimbus I) (modular, with rows of figure-of-8 shaped cells; 2 sets of cells are inflated and deflated over 10 minute cycle) versus alternating-pressure mattress (Pegasus Airwave) (double layer mattress with a 3-cell alternating cycle lasting 7.5 minutes). All participants were subject to the standard hospital protocol for wound dressing; details of this were not provided.	Elderly adults in hospital with pressure ulcers of grade 2 or above. The classification system used was not specified.	Complete healing at 4 weeks; comfort; median rate of reduction in area (cm²/day); withdrawal rates by group and reasons for withdrawal.	4-week follow-up.
Evans 2000⁵⁶	Alternating-pressure mattress replacement system (APMRS) (Nimbus 3) versus alternating-pressure mattress replacement system (APMRS) for adults in hospital (P.Biwave, P.Airwave.P.Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro) for people in a nursing home. Turning and wound care was standardised for 2 groups.	People in a hospital or nursing home, aged over 65 years with either grade 2 or 3 pressure ulcer (classification system not reported), or grade 2 pressure ulcer and difficulty to reposition in bed, unable to tolerate 30 degree tilt, unable to move in bed, in bed for over 20 hours or 24 hours, over 108kg and bed-bound or who have undergone spinal anaesthetic.	Absolute and relative reduction in pressure ulcer surface area; comfort.	2-week follow-up period.
Ferrell 1993⁵⁹	LAL bed (KINAIR) versus 10cm convoluted foam overlay on top of standard foam mattress .Both groups had similar co-interventions as per standard care that is mobilisation as much as possible; 2-hourly turning during waking hours; avoidance of head-of-bed elevation; avoidance of dragging people on sheets; nutritional support; infection control.	Older adults in an elderly nursing home with multiple medical problems, and with trunk or trochanter pressure ulcers. Grade 2 pressure ulcers or above (Shea grading system).	Rate of healing; pressure ulcer surface area was traced twice/week on plastic film and area measured using planimetry; ulcers completely healed (covered with epithelium).	Median follow-up of 33 days (LAL group) and 40 days (foam mattress).
Groen 1999⁷³	Foam replacement mattress (3 layers of polyurethane foam designated as comfort, load-distributing and support layers) versus Secutex water mattress (placed on top of standard hospital mattress, 3 PVC sections holding 26 litres water each, with heating element).Standard turning protocol (every 2-3 hour) for both groups.	People in a nursing home, aged over 59 years with a pressure ulcer on trunk, grade 3 (superficial cutaneous or subcutaneous necrotic) or grade 4 (deep subcutaneous necrotic). The classification system used was not reported.	Proportion with healed pressure ulcers at 4 weeks; mean pressure ulcer severity score at 4 weeks.	4-week follow-up.
Keogh 2001⁸⁹	Profiling bed with a pressure reducing foam mattress or cushion versus flat-based bed with a pressure relieving or redistributing mattress or cushion.	People from 2 surgical and 2 medical wards; over 18 years old; Waterlow score of 15-25; tissue damage no greater than grade 1 (EPUAP grading system).	Proportion with healed grade 1 pressure ulcers.	5-10 days follow-up.
Makhsous 2009¹⁰⁸	Wheelchair cushion equipped with an individualised cyclic pressure-relief protocol versus regular wheelchair cushions. Treatment was specific to the individual and a variety of wound care modalities applied when required (topical wound dressings eg wound gel, hydrocolloid, alginate, foam and moisture barrier) also silver antimicrobial dressings and negative pressure wound therapy.	Wheelchair users with spinal cord injury (paraplegia or tetraplegia) with existing stage 2 or 3 pressure ulcers (classification system not specified) in the sacral and/or ischial area.	Healing of pressure ulcers; healing rate of pressure ulcers; PUSH score improvement; % surface area healing; % PUSH score improvement.	30 days follow-up.
Mulder 1994¹²³	Air suspension bed (Therapulse, Kinetic concepts) (a pulsating air suspension therapy – cushions alternatively inflate and deflate but classed as LAL rather than AP) versus convoluted foam mattress overlay. Wound care and repositioning standardised for both groups.	People in a nursing home with grade 3-4 pressure ulcers (International Association of Enterostomal Therapists staging system).	Wound closure; pressure ulcer improvement (pressure ulcer reduced by 1 grade or more, including healed completely).	Maximum 12-weeks' follow-up or until ulcers healed, whichever came first.
Munro 1989¹²⁶	Air-fluidised bed (Clinitron) versus standard care. The bed/mattress in the standard care group was not described. Sheepskins or gel pads were placed beneath ulcer areas. Standard care involved positioning and massage.	Males with grade 2 or 3 pressure ulcers (classification system not specified), expected to remain in hospital for at least 15 days.	Change in mean pressure ulcer area (mm²); individuals' perception of pain; patient satisfaction.	15-day follow-up.
Nixon 2006¹³⁶	Alternating-pressure overlay within 24 hours of admission versus alternating-pressure mattress within 24 hours of admission.	Adults over 55 years of age, from vascular, orthopaedic, medical or care of the elderly wards with an expected length of stay at least 7 days and Braden score of 1 or 2, or an existing grade 2 pressure ulcer (grading system not specified).	Proportion of people developing a new pressure ulcer of grade 2 or above; time to development of new pressure ulcers; proportion of participants developing a new pressure ulcer within 30 days; healing of existing pressure ulcers; patient acceptability; adverse events.	30-day follow-up.
Osterbrink 2005¹⁴⁵	Repose device versus small cell versus large cell.	Participants recruited from aged care facility, acute care hospitals and home care setting, over 18 years old, with at least 1 grade 2 pressure ulcer at any bony prominence (EPUAP classification). If recruited from hospital, must have been nursed on care of the elderly, neurological or surgical units.	Pressure ulcer healing success; weekly changes in pressure ulcer (ulcer size, grade, wound bed, edge appearance and local treatment).	Follow-up time as long as clinical circumstances allowed. Maximum duration 42 days.
Russell 2000¹⁶⁰	Two types of alternating cell mattress systems with pressure-relieving cushions: Huntleigh Nimbus 3 with Aura cushion and 4-hourly turning versus Pegasus Cairwave Therapy System with Proactive 2 seating cushion and 8-hourly turning.	Individuals from care of the elderly units with pressure ulcers of grade 2 and above (Torrance classification system). Average age 83.9 and 84.6 years in the 2 groups.	Pressure ulcer healing: all types, and divided into heel and sacral pressure ulcers at 12 and 18 months.	18-month follow-up.
Russell 2003¹⁶²	Alternating-pressure, multicell mattress with 10 minute cycle time (Nimbus 3) versus fluid overlay mattress (RIK static). All adults had standard 4-hourly repositioning, but could have additional turning at the individual's request.	Adults with grade 1 or 2 pressure ulcers (EPUAP classification) admitted to hospital. Mean age 80 years. Baseline Waterlow scores 21.8 and 21.3 in groups 1 and 2 respectively and baseline Burton scores 14.6 and 14.2.	Improved pressure ulcer response; length of hospital stay.	Length of follow-up unclear, but presumably until discharge from enrolment hospital.
Strauss 1991¹⁸⁰	Home air-fluidised therapy (CLINITRON) when grade 3 or 4 prsesure ulcers present, plus the consultative and technical services of a visiting nurse specialist versus conventional or standard therapy, individual specific and prescribed (n=50), but included alternating –pressure pads, air-filled mattresses, water-filled mattresses, high density foam pads.	People: with at least 1 grade 3 or 4 pressure ulcer (Shea classification); who would probably require future hospitalisation for the pressure ulcer; with severely limited mobility; for who home air-fluidised therapy was a practical option, likely to comply, live at least 1 year; aged 16 years or over.	Pressure ulcers classified by blinded observers as improved; unchanged; worse; or not accessible; pressure ulcer-related hospitalisations and costs per patient; pressure ulcer-related hospital days per person.	36-week follow-up

Table 42Clinical evidence profile: water mattress overlay versus low-tech mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Water mattress overlay	Low-tech mattress	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed at 4 weeks – grade 3 ulcers (no classification system specified), people in a nursing home, 4-week follow-up)⁷³
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	27/60 (45%)	29/60 (48.3%)	RR 0.93 (0.63 to 1.37)	34 fewer per 1000 (from 179 fewer to 179 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	48.3%	RR 0.93 (0.63 to 1.37)	34 fewer per 1000 (from 179 fewer to 179 more)	Very low	Critical
Percentage reduction in pain (change values)– grade 3 ulcers (no classification system specified), people in a nursing home, 4-week follow-up⁷³
1	Randomised trial	Very serious^a^,^d	No serious inconsistency	No serious indirectness	Very serious^c	None	35.9%	16.2%	MD 19.7% higher	-	Very low	Important
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of reduction in size of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size or volume of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Groen (1999) did not provide details of the randomisation method used. There was unclear allocation concealment and no blinding of outcome assessors was reported. There was insufficient reporting of incomplete outcome data and no details provided of type of analysis used. The authors report selectively and the details of the grading system used are not specified.

b: The confidence interval crossed both MID points.

c: There was not enough data to analyse in Revman.

d: There were baseline differences in pain at start of trial (40% in water mattress overlay group and 20% for low-tech mattress group).–

Table 43Clinical evidence profile: 3 dimensional macroporous overlay versus gel overlay

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	3-D microporous overlay	Gel overlay	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed – people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	3/35 (8.6%)	5/37 (13.5%)	RR 0.63 (0.16 to 2.46)	50 fewer per 1000 (from 114 fewer to 197 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	13.5%	RR 0.63 (0.16 to 2.46)	50 fewer per 1000 (from 113 fewer to 197 more)	Very low	Critical
Mortality (all-cause) – people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	3/35 (8.6%)	7/37 (18.9%)	RR 0.45 (0.13 to 1.62)	104 fewer per 1000 (from 165 fewer to 117 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	18.9%	RR 0.45 (0.13 to 1.62)	104 fewer per 1000 (from 164 fewer to 117 more)	Very low	Important
Suspension due to worsening of pressure ulcers– people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	9/35 (25.7%)	17/37 (45.9%)	RR 0.56 (0.29 to 1.09)	202 fewer per 1000 (from 326 fewer to 41 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	46%	RR 0.56 (0.29 to 1.09)	202 fewer per 1000 (from 327 fewer to 41 more)	Very low	Important
Suspension due to intolerance– people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	5/35 (14.3%)	2/37 (5.4%)	RR 2.64 (0.55 to 12.75)	89 more per 1000 (from 24 fewer to 635 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	5.4%	RR 2.64 (0.55 to 12.75)	89 more per 1000 (from 24 fewer to 635 more)	Very low	Important
Unchanged/worsened– people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	16/35 (45.7%)	22/37 (59.5%)	RR 0.77 (0.49 to 1.2)	137 fewer per 1000 (from 303 fewer to 119 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	59.5%	RR 0.77 (0.49 to 1.2)	137 fewer per 1000 (from 303 fewer to 119 more)	Very low	Critical
Improved– people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	16/35 (45.7%)	9/37 (24.3%)	RR 1.88 (0.96 to 3.68)	214 more per 1000 (from 10 fewer to 652 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	24.3%	RR 1.88 (0.96 to 3.68)	214 more per 1000 (from 10 fewer to 651 more)	Very low	Critical
Patient comfort (fair to excellent) – people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	27/35 (77.1%)	19/37 (51.4%)	RR 1.5 (1.05 to 2.16)	257 more per 1000 (from 26 more to 596 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None		51.4%	RR 1.5 (1.05 to 2.16)	257 more per 1000 (from 26 more to 596 more)	Very low	Important
Patient comfort (poor) – people in long-term care - all grade of ulcers (EPUAP-NPUAP)³⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	8/35 (22.9%)	18/37 (48.6%)	RR 0.47 (0.23 to 0.94)	258 fewer per 1000 (from 29 fewer to 375 fewer)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	48.7%	RR 0.47 (0.23 to 0.94)	258 fewer per 1000 (from 29 fewer to 375 fewer)	Very low	Important
Time to complete healing (time to event data)
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of healing (continuous data)
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of change in size of ulcer (absolute and relative) (continuous data)
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size of ulcer and volume of ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care (continuous data)
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life (continuous data)
-	-	-	-	-	-	-	-	-	-	-	-	-

a: The study was unblinded. There were baseline differences for grade of pressure ulcers, but the higher grade were in the intervention group.

b: The confidence interval crossed both MID points.

c: The confidence interval crossed 1 MID point.

Table 44Clinical evidence profile: low-air-loss bed versus low-tech foam mattress overlay

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Low-air-loss bed	Foam mattress overlay	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed - Shea grade 2 ulcers or above, people in an elderly nursing home, mean 36 days follow-up^g⁵⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	26/43 (60.5%)	19/41 (46.3%)	RR 1.3 (0.87 to 1.96)	139 more per 1000 (from 60 fewer to 445 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	-	46.3%	RR 1.3 (0.87 to 1.96)	139 more per 1000 (from 60 fewer to 444 more)	Very low	Critical
Proportion of people with pressure ulcers completely healed - International Association of Enterostomal Therapists (IAET) staging system stage 3 and 4 ulcers, people in a nursing home, 12 weeks follow-up^g¹²³
1	Randomised trial	Very serious^c	No serious inconsistency	No serious indirectness	Very serious^d	None	5/31 (16.1%)	3/18 (16.7%)	RR 0.97 (0.26 to 3.58)	5 fewer per 1000 (from 123 fewer to 430 more)	Very low	Critical
1	Randomised trial	Very serious^c	No serious inconsistency	No serious indirectness	Very serious^d	None	-	16.7%	RR 0.97 (0.26 to 3.58)	5 fewer per 1000 (from 124 fewer to 431 more)	Very low	Critical
Proportion of people with pressure ulcers completely healed (meta-analysed) – Shea grade 2 ulcers or above and International Association of Enterostomal Therapists staging system stage 3 and 4 ulcers – people in an elderly nursing home¹²³^;⁵⁹
2	Randomised trials	Very serious^a^,^c	No serious inconsistency	No serious indirectness	Serious^b	None	31/74 (41.9%)	22/59 (37.3%)	RR 1.25 (0.84 to 1.86)	93 more per 1000 (from 60 fewer to 321 more)	Very low	Critical
2	Randomised trials	Very serious^a^,^c	No serious inconsistency	No serious indirectness	Serious^b	None	-	31.5%	RR 1.25 (0.84 to 1.86)	79 more per 1000 (from 50 fewer to 271 more)	Very low	Critical
Pressure ulcers reduced by 1 grade or more including healed completely - International Association of Enterostomal Therapists staging system stage 3 and 4 ulcers, people in a nursing home, 12-weeks follow up¹²³
1	Randomised trial	Very serious^c	No serious inconsistency	No serious indirectness	Very serious^d	None	10/31 (32.3%)	5/18 (27.8%)	RR 1.16 (0.47 to 2.86)	44 more per 1000 (from 147 fewer to 517 more)	Very low	Critical
1	Randomised trial	Very serious^c	No serious inconsistency	No serious indirectness	Very serious^d	None	-	27.8%	RR 1.16 (0.47 to 2.86)	44 more per 1000 (from 147 fewer to 517 more)	Very low	Critical
Rate of healing (mm²/day) median (25th, 75th percentiles) - Shea grade 2 ulcers or above, people in a nursing home, mean 36 days follow-up⁵⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^f	Seriousⁱ	9.0 (4.0, 19.8)	2.5 (0.5 to 6.5)	p=0.0002	-	Very low	Critical
Mean change in ulcer size (final values)– NPUAP stage 2 ulcers, people in hospital, 7 days follow-up⁵¹
1	Randomised trial	Very serious^e	No serious inconsistency	No serious indirectness	Serious^b	Serious^h	7.3 (s.d 2.4) n= 25	5.3 (s.d 2.1) n=23	-	MD 2 higher (0.73 to 3.27 higher)	Very low	Critical
Mean change in ulcer size (final values) – NPUAP stage 3 and 4 ulcers, people in hospital, 7 days follow-up⁵¹
1	Randomised trial	Very serious^e	No serious inconsistency	No serious indirectness	No serious	Serious^h	37.1 (s.d 8.1) n=17	12.4 (s.d 3.5) n=12	-	MD 24.7 higher (20.37 to 29.03 higher)	Very low	Critical
Mean comfort scores (perception of comfort) (better indicated by lower values) – NPUAP stage 2 to 4 ulcers, people in hospital, 7 days follow-up⁵¹
1	Randomised trial	Very serious^e	No serious inconsistency	No serious indirectness	No serious imprecision	None	4.1 (s.d 1.3) n=20	3.7 (s.d 1.3) n=19	T[37]=0. 91 p>0.05	MD 0.4 higher (0.42 lower to 1.22 higher)	Low	Critical
Mortality - Shea grade 2 ulcers or above, people in a nursing home, mean 36 days follow-up⁵⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very seriousd	None	11/43 (25.6%)	7/41 (17.1%)	RR 1.5 (0.64 to 3.49)	85 more per 1000 (from 61 fewer to 425 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very seriousd	-	-	17.1%	RR 1.5 (0.64 to 3.49)	86 more per 1000 (from 62 fewer to 426 more)	Very low	Important
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Ferrell (1993) terminated at interim analysis as difference much larger than expected. There was unclear sequence generation and blinding, as well as insufficient reporting of incomplete outcome data. There was a higher drop-out than event rate for the ‘proportion completely healed’ outcome.

b: The confidence interval crossed 1 MID point.

c: Mulder (1994) did not provide details of the randomisation method and there was unclear allocation concealment and blinding. It was also unclear from which group drop-outs came from; not all of the pre-specified outcomes were reported and ulcer size was not reported at baseline. There was iInsufficient reporting of incomplete outcome data and a higher drop-out than event rate for' proportion completely healed' outcome.

d: The confidence interval crossed both MID points.

e: Day (1993) did not report clear methods of randomisation, allocation concealment or blinding. There was insufficient reporting of incomplete outcome data and not all of the pre-specified outcomes were analysed. The authors did not report initial ulcer sizes.

f: There was not enough data to put in Revman.

g: The Cochrane review did not conduct meta-analysis as the outcomes were measured in different ways. Ferrell (1993) used tracing of the epithelial border of the ulcer on plastic film and then the are measured using a polar planimeter. The wounds were assessed using the four-point Shea scale and the Sessing scale (similar to Shea scale, but was undergoing development at time of the study), which has 7 verbal descriptions of ulcers including colour, presence of granulation tissue, evidence of infection, drainage, odour and eschar. Mulder (1994) assessed wound surface area by photoplanimetry. Ulcer volume = ulcer length × width × depth (of deepest ulcer point). The pressure ulcers were assessed using the International Association of Enterostomal Therapists staging system. Only stage 3 and 4 ulcers were included in this study.

h: The baseline had a larger difference than the difference between the final values therefore the results should be viewed with caution. There was no log transformation of data.

i: A non-parametric test (Wilcoxon rank-sum) was used but there was no log transformation of data.

Table 45Clinical evidence profile: low-air-loss bed versus low-air-loss overlay

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Low-air-loss bed	Low-air-loss overlay	Relative (95% CI)	Absolute	Quality	Importance
Median change in ulcer area (cm²) – people in acute care, mean 24 day follow-up³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	3.9 cm2	1.9 cm2	p=0.060	-	Very low	Critical
Mean changes in pressure ulcer surface area– people in acute care, mean 24 day follow-up³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	10.2 cm2	3.8 cm2	-	-	Very low	Critical
Proportion of people with pressure ulcers completely healed
-	-	-	-	-	-	-	-	-	-	-	-	-
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of change in size of ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: There was very little data provided (median change in area and range). It was unclear (and unlikely) that the outcome assessment was blind to treatment group. No description of co-interventions except skincare protocol applied to both groups; insufficient reporting of incomplete outcome data; high drop-out.

b: No data were available to analyse in Revman.

Table 46Clinical evidence profile: air-fluidised therapy (AFT) versus standard or conventional therapies

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Air-fluidised bed	Standard care	Relative (95% CI)	Absolute	Quality	Importance
Proportion with 50% reduction in total surface area – Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	9/31 (29%)	8/34 (23.5%)	RR 1.23 (0.54 to 2.8)	54 more per 1000 (from 108 fewer to 424 more)	Very low	Critical
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	23.5%	RR 1.23 (0.54 to 2.8)	54 more per 1000 (from 108 fewer to 423 more)	Very low	Critical
Proportion with improvement in pressure ulcers – Shea stage 3 or 4 ulcers, people at home, 36 weeks follow-up¹⁸⁰
1	Randomised trial	Very serious^c^,ⁱ	No serious inconsistency	No serious indirectness	Serious^d	None	19/22 (86.4%)o	9/13 (69.2%)o	RR 1.25 (0.84 to 1.86)	173 more per 1000 (from 11 fewer to 595 more)	Very low	Critical
1	Randomised trial	Very serious^c^,ⁱ	No serious inconsistency	No serious indirectness	Serious^d	None	-	69.2%	RR 1.25 (0.84 to 1.86)	173 more per 1000 (from 11 fewer to 595 more)	Very low	Critical
Proportion with improvement in pressure ulcers – Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Serious^d	None	22/31 (71%)	16/34 (47.1%)	RR 1.51 (0.99 to 2.3)	240 more per 1000 (from 5 fewer to 612 more)	Low	Critical
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Serious^d	None	-	47.1%	RR 1.51 (0.99 to 2.3)	240 more per 1000 (from 5 fewer to 612 more)	Low	Critical
Proportion with improvement in pressure ulcers – Shea all stages (people in surgery and people at home) – meta-analysed⁶^;¹⁸⁰
2	Randomised trials	Serious^a^,^c	No serious inconsistency	No serious indirectness	Serious^d	None	41/53 (77.4%)o	25/47 (53.2%)o	RR 1.4 (1.04 to 1.88)	213 more per 1000 (from 21 more to 468 more)	Low	Critical
2	Randomised trials	Serious^a^,^c	No serious inconsistency	No serious indirectness	Serious^d	None	-	58.1%	RR 1.4 (1.04 to 1.88)	232 more per 1000 (from 23 more to 511 more)	Low	Critical
Change in mean ulcer area (mm²) – stage 2 or 3 ulcers (not specified which classification system), people in hospital, 15 days follow-up (final values)¹²⁶
1	Randomised trial	Very serious^e	No serious inconsistency	No serious indirectness	Very serious^k	Serious^l	1158mm2	2051mm2	-	p=0.05	Very low	Critical
Change in total surface area (median, range) cm²– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Very serious^f	Serious ^m	-1.2 (-38.0 to +15.5)	+0.5 (-55.1 to +94.7)	-	Difference (median): -1.7cm2 (95% CI -9.2cm2 to -0.6cm2)	Very low	Critical
Reduction in pain^g– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Very serious^a^,^h	No serious inconsistency	No serious indirectness	Serious^d	None	8/13 (61.5%)	4/14 (28.6%)	RR 2.15 (0.85 to 5.48)	329 more per 1000 (from 43 fewer to 1000 more)	Very low	Important
Increase in pain^g– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Very serious^a^,^h	No serious inconsistency	No serious indirectness	Very serious^b	None	0/13 (0%)	3/14 (21.4%)	Peto OR 0.12 (0.01 to 1.31)	183 fewer per 1000 (from 212 fewer to 49 more)	Very low	Important
1	Randomised trial	Very serious^a^,^h	No serious inconsistency	No serious indirectness	Very serious^b	None	0/13 (0%)	21.4%	Peto OR 0.12 (0.01 to 1.31)	182 fewer per 1000 (from 212 fewer to 49 more	Very low	Important
Time in hospital (better indicated by lower values) – Shea stage 3 or 4 ulcers, people at home, 36 weeks follow-up¹⁸⁰
1	Randomised trial	Very serious^c	No serious inconsistency	No serious indirectness	Very serious^b	None	11.5 (s.d 8.8) days n= 47	21.5 (s.d 23.8) days n= 50	-	MD 10 lower (161.64 lower to 141.64 higher)	Very low	Important
Median length of stay in hospital after randomisation– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Very serious^f	None	16 days	15 days	-	-	Very low	Important
Patient satisfaction (better indicated by higher values) – stage 2 or 3 ulcers (not specified which classification system), people in hospital, 15 days follow-up¹²⁶
1	Randomised trial	Very serious^e^,^h	No serious inconsistency	No serious indirectness	No serious imprecision	None	57.5 (s.d 6.1) n= 8	48.6 (s.d 12.3) n=10	-	MD 8.9 higher (0.18 to 17.62 higher)	Low	Critical
Increase in comfort– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Very serious^a^,^h	No serious inconsistency	No serious indirectness	Serious^d	None	8/13 (61.5%)	3/14 (21.4%)	RR 2.87 (0.96 to 8.55)	401 more per 1000 (from 9 fewer to 1000 more)	Very low	Critical
Reduction in comfort– Shea all stages, people in surgery, mean 13 days follow-up⁶
1	Randomised trial	Very serious^a^,^h	No serious inconsistency	No serious indirectness	Very serious^b	None	1/13 (7.7%)	6/14 (42.9%)	RR 0.18 (0.02 to 1.30)	351 fewer per 1000 (from 420 fewer to 129 more)	Very low	Critical
Mortality –Shea all stages, people in surgery and at home⁶^;¹⁸⁰
2	Randomised trials	Serious^a^,^c	No serious inconsistency	No serious indirectness	Very serious^b	None	22/89 (24.7%)	26/88 (29.5%)	RR 0.83 (0.51 to 1.34)	50 fewer per 1000 (from 145 fewer to 100 more)	Very low	Important
2	Randomised trials	Serious^a^,^c	No serious inconsistency	No serious indirectness	Very serious^b	None	-	27.9%	RR 0.83 (0.51 to 1.34)	50 fewer per 1000 (from 145 fewer to 100 more)	Very low	Important
Proportion of patients with pressure ulcers completely healed
-	-	-	-	-	-	-	-	-	-	-	-	-
Time to complete healing
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of change in size of ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Allman (1987) did not report clear allocation concealment and there were baseline differences. The size of ulcer at baseline was not reported.

b: The confidence interval crossed both MID points.

c: Strauss (1991) did not report clear allocation concealment and there was insufficient reporting of incomplete outcome data. The size of ulcer at baselinewas not reported. There was also a high drop-out rate.

d: The confidence interval crossed 1 MID point.

e: Munro (1989) did not report clear allocation concealment and no information regarding sample size calculations, randomisation method, blinding, baseline characteristics or extent of follow-up were reported. No raw data was presented in the paper and there was insufficient reporting of incomplete outcome data.

f: It was not possible to analyse data in Revman.

g: There was a change in pain intensity from baseline (from asking participants to score 0 to 5 on words to describe pain (none, mild, discomforting, distressing, horrible or excruciating)).

h: The participant self-reported outcomes.

i: Improvement was assessed by an independent nurse reviewer's assessment of the participants' pressure ulcere. There was no definition of improvement.

j: Improvement was defined as those pressure ulcers that had healed, much improved, or a little improved. Non-improvement included those that were unchanged, a little worse, or much worse. This was assessed by an investigator and a plastic surgeon independently from photographs.

k: The change scores were given by study but it was not possible to analyse data in Revman as no standard deviations were given.

l: The ulcer size (diameter) at day 1 had a larger difference between the groups than the difference between the ulcer sizes at day 15. There was no log transformation of data.

m: Non-parametric tests were used but there was no log transformation of data.

n: Less than half the participants completed the questionnaire.

o: Strauss used an independent nurse reviewer's assessment of the participants' pressure ulcer, the data was given for both reviewers and then for the purposes of this review, the results were amalgamated for the 35 participants who were assessed.

Table 47Clinical evidence profile: alternating-pressure mattress versus alternating-pressure mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Alternating-pressure mattress	Alternating-pressure mattress	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed – grade 2 and above (grading system not specified), elderly adults, 4-week follow-up – alternating-pressure mattress (Nimbus 1) versus alternating-pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	10/16 (62.5%)	5/14 (35.7%)	RR 1.75 (0.79 to 3.89)	268 more per 1000 (from 75 fewer to 1000 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	-	35.7%	RR 1.75 (0.79 to 3.89)	268 more per 1000 (from 75 fewer to 1000 more)	Very low	Critical
Proportion of pressure ulcers completely healed – grade 2 and above (Torrance classification system), elderly adults, 18 months follow-up – alternating pressure mattress (Nimbus 3) with Aura cushion and 4-hourly turning versus alternating pressure mattress (Pegasus Cairwave Therapy System) with Proactive 2 seating cushion and 8-hourly turning¹⁶¹
1	Randomised trial	Serious^c	No serious inconsistency	No serious indirectness	No serious imprecision	None	65/71 (91.5%)	65/70 (92.9%)	RR 0.99 (0.9 to 1.09)	9 fewer per 1000 (from 93 fewer to 84 more)	Moderate	Critical
1	Randomised trial	Serious^c	No serious inconsistency	No serious indirectness	No serious imprecision	None	-	92.9%	RR 0.99 (0.9 to 1.09)	9 fewer per 1000 (from 93 fewer to 84 more)	Moderate	Critical
Proportion of people with decrease in pressure ulcer size– grade 2 and above (grading system not specified), elderly adults, 4-week follow-up– alternating pressure mattress (Nimbus 1) versus alternating pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	4/16 (25%)	6/14 (42.9%)	RR 0.58 (0.21 to 1.65)	180 fewer per 1000 (from 339 fewer to 279 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	-	42.9%	RR 0.58 (0.21 to 1.65)	180 fewer per 1000 (from 339 fewer to 279 more)	Very low	Critical
Proportion of people with increase in pressure ulcer size– grade 2 and above (grading system not specified), elderly adults, 4-week follow-up– alternating pressure mattress (Nimbus 1) versus alternating pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	2/16 (12.5%)	3/14 (21.4%)	RR 0.58 (0.11 to 3.00)	90 fewer per 1000 (from 191 fewer to 429 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	-	21.4%	RR 0.58 (0.11 to 3.00)	90 fewer per 1000 (from 190 fewer to 428 more)	Very low	Critical
Median rate of reduction in surface area (cm/day) – grade 2 and above (grading system not specified), elderly adults, 4-week follow-up– alternating pressure mattress (Nimbus 1) versus alternating pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^e	Serious^h	0.089cm2/day	0.107cm2/day	Difference 0.018 cm² (95% CI 0.179 to 0.143) p=0.92	-	Very low	Critical
Median absolute reduction in wound surface area per day – grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up – alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f	No serious inconsistency	No serious indirectness	Very serious^e	Serious^h	0.12cm 2 (range 0 to 0.21cm2)	0.08cm2 (range 0.04 to 0.33cm2)	p=0.570	-	Very low	Critical
Median relative reduction in wounds surface area– grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f	No serious inconsistency	No serious indirectness	Very serious^e	Serious^h	2.44% (range 0-7.14%)	1.34% (range 1.11-2.88%)	p=0.570	-	Very low	Critical
Median absolute reduction in wound surface are per day– grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f	No serious inconsistency	No serious indirectness	Very serious^e	Serious^h	0.11cm2 (range 0.04 to 0.41cm2)	0.05cm2 (range 0-0.48cm2)	p=0.570	-	Very low	Critical
Median relative reduction in wounds surface area – grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f	No serious inconsistency	No serious indirectness	Very serious^e	Serious^h	1.57% (range 0.45-5%)	0.99% (range 0-2.54%)	p=0.570	-	Very low	Critical
Mean time in hospital (for those who completed the trial) – grade 2 and above (Torrance classification system), elderly adults in hospital and nursing homes, 18 months follow-up– alternating pressure mattress (Nimbus 3) with Aura cushion and 4-hourly turning versus alternating pressure mattress (Pegasus Cairwave Therapy System) with Proactive 2 seating cushion and 8-hourly turning¹⁶¹
1	Randomised trial	Serious^c	No serious inconsistency	No serious indirectness	Very serious^e	None	21.6 days n=57	21.7 days n=55	-	-	Very low	Important
Comfort (people in hospital)– grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f^,^g	No serious inconsistency	No serious indirectness	Very serious^e	None	5 (very comfortable)	4 (comfortable)	p=0.006	-	Very low	Critical
Comfort (people in nursing homes)– grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– Nimbus 3 versus P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)⁵⁶
1	Randomised trial	Very serious^f^,^g	No serious inconsistency	No serious indirectness	Very serious^e	None	5 (very comfortable	4 (comfortable)	p=0.002	-	Very low	Critical
Comfort – grade 2 and above (grading system not specified), elderly adults, 4-week follow-up– alternating pressure mattress (Nimbus 1) versus alternating pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a^,^g	No serious inconsistency	No serious indirectness	Very serious^e	None	Median 8/10	Median 8/10	-	-	Very low	Critical
Mortality – grade 2 and above (grading system not specified), elderly adults, 4-week follow-up– alternating pressure mattress (Nimbus 1) versus alternating pressure mattress (Pegasus Airwave)⁵⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	6/22 (27.3%)	5/19 (26.3%)	RR 1.04 (0.38 to 2.86)	11 more per 1000 (from 163 fewer to 489 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^d	None	-	26.3%	RR 1.04 (0.38 to 2.86)	11 more per 1000 (from 163 fewer to 489 more)	Very low	Important
Mortality – grade 2 and above (Torrance classification system), elderly adults, 18 months follow-up– alternating pressure mattress (Nimbus 3) with Aura cushion and 4-hourly turning versus alternating pressure mattress (Pegasus Cairwave Therapy System) with Proactive 2 seating cushion and 8-hourly turning¹⁶¹
1	Randomised trial	Serious^c	No serious inconsistency	No serious indirectness	Serious^c	None	16/71 (22.5%)	10/70 (14.3%)	RR 1.58 (0.77 to 3.23)	83 more per 1000 (from 33 fewer to 319 more)	Low	Important
1	Randomised trial	Serious^c	No serious inconsistency	No serious indirectness	Serious^c	None	-	14.3%	RR 1.58 (0.77 to 3.23)	83 more per 1000 (from 33 fewer to 319 more)	Low	Important
Mortality – grade 2 and above (grading system not specified), elderly adults in hospital and nursing homes, 2 week follow-up– alternating pressure mattress (Nimbus 3) versus alternating pressure mattress (P.Biwave, P.Airwave, P. Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro) ⁵⁶
1	Randomised trial	Very serious^f	No serious inconsistency	No serious indirectness	Very serious^d	None	7/17 (41.2%)	3/15 (20%)	RR 2.06 (0.64 to 6.57)	212 more per 1000 (from 72 fewer to 1000 more)	Very low	Important
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Devine (1995) did not report whether there was blinding of outcome assessors and there was baseline differences in the groups (more people incontinent of urine in alternating pressure mattress (Nimbus) group, more people catheterised in alternating pressure mattress (Airwave) group). Baseline pressure ulcer size was not reported and there was drop-out higher than event rate. The study used a very small sample size.

b: The confidence interval crossed 1 MID point.

c: Russell (2000) did not provide details of the randomisation method and there was unclear allocation concealment.

d: The confidence interval crossed both MID points.

e: There was not enough data available to analyse in Revman.

f: Evans (2000) did not report a method of randomisation and there was unclear allocation concealment. A large proportion of participants did not complete follow-up (11/20 in nursing home group and 75% of hospital group). The study used a very small sample size.

g: The participants self-reported outcomes.

h: There was no log transformation of data.

Table 48Clinical evidence profile: alternating-pressure mattress overlay versus alternating-pressure mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Alternating-pressure mattress overlay	Alternating-pressure mattress	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	20/59 (33.9%)	19/54 (35.2%)	RR 0.96 (0.58 to 1.6)	14 fewer per 1000 (from 148 fewer to 211 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	35.2%	RR 0.96 (0.58 to 1.6)	14 fewer per 1000 (from 148 fewer to 211 more)	Very low	Critical
Time to healing (median days)¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^e	None	20 days (12 to not estimable)	20 days (10 to not estimable)	-	p=0.86 log-rank test	Very low	Important
Absolute change in surface area (cm²) - change values (better indicated by higher values) – grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	1 (SD 2.3) n=33	2 (SD 6.1) n=36	-	MD 1 lower (3.14 lower to 1.14 higher)	Very low	Critical
% change in surface area (change values) (better indicated by higher values) – grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-35 (SD 605.5) n=33	34.4 (SD 108.6) n=36	-	MD 69.4 lower (279.01 lower to 140.21 higher)	Very low	Critical
Pressure ulcer improvement– grade 1 or 2 pressure ulcers (EPUAP classification system), elderly adults, follow-up period not specified¹⁶²
1	Randomised trial	Very serious^d	No serious inconsistency	No serious indirectness	No serious	None	56/75 (74.7%)	60/83 (72.3%)	RR 1.03 (0.86 to 1.25)	22 more per 1000 (from 101 fewer to 181 more)	Low	Critical
1	Randomised trial	Very serious^d	No serious inconsistency	No serious indirectness	No serious	None	-	72.3%	RR 1.03 (0.86 to 1.25)	22 more per 1000 (from 101 fewer to 181 more)	Low	Critical
Worsening of pressure ulcers– grade 1 or 2 pressure ulcers (EPUAP classification system), elderly adults, follow-up period not specified¹⁶²
1	Randomised trial	Very serious^d	No serious inconsistency	No serious indirectness	Very serious^b	None	16/75 (21.3%)	22/83 (26.5%)	RR 0.8 (0.46 to 1.41)	53 fewer per 1000 (from 143 fewer to 109 more)	Very low	Critical
1	Randomised trial	Very serious^d	No serious inconsistency	No serious indirectness	Very serious^b	None	-	26.5%	RR 0.8 (0.46 to 1.41)	53 fewer per 1000 (from 143 fewer to 109 more)	Very low	Critical
Time in hospital (mean) – grade 1 or 2 pressure ulcers (EPUAP classification system), elderly adults, follow-up period not specified¹⁶²
1	Randomised trial	Very serious^d	No serious inconsistency	No serious indirectness	Very serious^e	None	22.17 days	20.05 days	-	p=0.23	Very low	Important
Patient acceptability (requested changes for comfort or other device-related reasons) – grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	230/989 (23.3%)	186/982 (18.9%)	RR 1.23 (1.03 to 1.46)	44 more per 1000 (from 6 more to 87 more)	Low	Important
1	Randomised trial	Serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	18.9%	RR 1.23 (1.03 to 1.46)	43 more per 1000 (from 6 more to 87 more)	Low	Important
Proportion of participants with negative comments on mattress motion– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	328/929 (35.3%)	285/891 (32%)	RR 1.1 (0.97 to 1.26)	32 more per 1000 (from 10 fewer to 83 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	32%	RR 1.1 (0.97 to 1.26)	32 more per 1000 (from 10 fewer to 83 more)	Very low	Important
Proportion of participants with positive comments for mattress motion– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious	None	272/929 (29.3%)	263/891 (29.5%)	RR 0.99 (0.86 to 1.14)	3 fewer per 1000 (from 41 fewer to 41 more)	Low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious	None	-	29.5%	RR 0.99 (0.86 to 1.14)	3 fewer per 1000 (from 41 fewer to 41 more)	Low	Important
Proportion of participants commenting negatively on getting into or out of bed– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	124/929 (13.3%)	127/891 (14.3%)	RR 0.94 (0.74 to 1.18)	9 fewer per 1000 (from 37 fewer to 26 more)	Low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None		14.3%	RR 0.94 (0.74 to 1.18)	9 fewer per 1000 (from 37 fewer to 26 more)	Low	Important
Participants commenting negatively on movement in bed– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	290/929 (31.2%)	260/891 (29.2%)	RR 1.07 (0.93 to 1.23)	20 more per 1000 (from 20 fewer to 67 more)	Low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	-	29.2%	RR 1.07 (0.93 to 1.23)	20 more per 1000 (from 20 fewer to 67 more)	Low	Important
Proportion of participants commenting positively on movement in bed– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	25/929 (2.7%)	27/891 (3%)	RR 0.89 (0.52 to 1.52)	3 fewer per 1000 (from 15 fewer to 16 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	3%	RR 0.89 (0.52 to 1.52)	3 fewer per 1000 (from 14 fewer to 16 more)	Very low	Important
Proportion of participants commenting on temperature as hot or warm– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	67/929 (7.2%)	50/891 (5.6%)	RR 1.29 (0.9 to 1.83)	16 more per 1000 (from 6 fewer to 47 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	5.6%	RR 1.29 (0.9 to 1.83)	16 more per 1000 (from 6 fewer to 46 more)	Very low	Important
Proportion of participants commenting on sweaty or sticky temperature– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	32/929 (3.4%)	23/891 (2.6%)	RR 1.33 (0.79 to 2.26)	9 more per 1000 (from 5 fewer to 33 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	2.6%	RR 1.33 (0.79 to 2.26)	9 more per 1000 (from 5 fewer to 33 more)	Very low	Important
Proportion of participants commenting on cold or cool temperature– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	11/929 (1.2%)	11/891 (1.2%)	RR 0.96 (0.42 to 2.2)	0 fewer per 1000 (from 7 fewer to 15 more)	Very	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	1.2%	RR 0.96 (0.42 to 2.2)	0 fewer per 1000 (from 7 fewer to 14 more)	Very	Important
Mattress not working or not working properly– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	16/929 (1.7%)	18/891 (2%)	RR 0.85 (0.44 to 1.66)	3 fewer per 1000 (from 11 fewer to 13 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	2%	RR 0.85 (0.44 to 1.66)	3 fewer per 1000 (from 11 fewer to 13 more)	Very low	Important
Hard to tuck sheet under or sheets come off or gather ormattress cover slips– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	19/929 (2%)	6/891 (0.7%)	RR 3.04 (1.22 to 7.57)	14 more per 1000 (from 1 more to 44 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	0.7%	RR 3.04 (1.22 to 7.57)	14 more per 1000 (from 2 more to 46 more)	Very low	Important
Mattress or bed too high– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	72/929 (7.8%)	48/891 (5.4%)	RR 1.44 (1.01 to 2.05)	24 more per 1000 (from 1 more to 57 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	5.4%	RR 1.44 (1.01 to 2.05)	24 more per 1000 (from 1 more to 57 more)	Very low	Important
Mattress slippy– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	9/929 (1%)	4/891 (0.4%)	RR 2.16 (0.67 to 6.98)	5 more per 1000 (from 1 fewer to 27 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	0.5%	RR 2.16 (0.67 to 6.98)	6 more per 1000 (from 2 fewer to 30 more)	Very low	Important
Mattress too soft or edges soft or slope– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	19/929 (2%)	29/891 (3.3%)	RR 0.63 (0.35 to 1.11)	12 fewer per 1000 (from 21 fewer to 4 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	3.3%	RR 0.63 (0.35 to 1.11)	12 fewer per 1000 (from 21 fewer to 4 more)	Very low	Important
Not able to use backrest– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	4/929 (0.4%)	2/891 (0.2%)	RR 1.92 (0.35 to 10.45)	2 more per 1000 (from 1 fewer to 21 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	0.2%	RR 1.92 (0.35 to 10.45)	2 more per 1000 (from 1 fewer to 19 more)	Very low	Important
Mattress-related fall¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	0/828 (0%)	4/891 (0.4%)	Peto OR 0.14 (0.02 to 1.03)	4 fewer per 1000 (from 4 fewer to 0 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	0.5%	Peto OR 0.14 (0.02 to 1.03)	4 fewer per 1000 (from 5 fewer to 0 more)	Very low	Important
Mattress-related suspected contact dermatitis– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	0/929 (0%)	1/891 (0.1%)	Peto OR 0.13 (0 to 6.54)	1 fewer per 1000 (from 1 fewer to 6 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	0.1%	Peto OR 0.13 (0 to 6.54)	1 fewer per 1000 (from 1 fewer to 6 more)	Very low	Important
Mattress-related climbed over or fell through cot sides– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	2/929 (0.2%)	1/891 (0.1%)	RR 1.92 (0.17 to 21.12)	1 more per 1000 (from 1 fewer to 23 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	0.1%	RR 1.92 (0.17 to 21.12)	1 more per 1000 (from 1 fewer to 20 more)	Very low	Important
Mattress deflation during transfer– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	0/929 (0%)	1/891 (0.1%)	Peto OR 0.13 (0 to 6.54)	1 fewer per 1000 (from 1 fewer to 6 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	0.1%	Peto OR 0.13 (0 to 6.54)	1 fewer per 1000 (from 1 fewer to 6 more)	Very low	Important
Mortality– grade 2 and above (classification system not specified), elderly adults, 30 day follow-up¹³⁶^,¹³⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	20/59 (33.9%)	12/54 (22.2%)	RR 1.53 (0.83 to 2.82)	118 more per 1000 (from 38 fewer to 404 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	None	-	22.2%	RR 1.53 (0.83 to 2.82)	118 more per 1000 (from 38 fewer to 404 more)	Very low	Important
Rate of change in size of pressure ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Nixon (2006) did not report blinding and the drop-out was higher than the event rate. The outcomes of patient acceptability and side effects were for the study as a whole rather than solely those who had pressure ulcers.

b: The confidence interval crossed both MID points.

c: The confidence interval crossed 1 MID point

d: Russell (2003) did not report blinding and there was unclear allocation concealment. There was insufficient reporting of incomplete outcome data.

e: There was not enough data to analyse in Revman.

f: This was a non-validated assessment of outcome.

Table 49Clinical evidence profile: air-filled devices versus alternating pressure mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Air-filled devices	Alternating-pressure mattress	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed – grade 2 ulcer or above (EPUAP classification system), elderly adults, maximum follow-up 42 days¹⁴⁵
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	7/34 (20.6%)	1/26 (3.8%)	RR 5.35 (0.7 to 40.84)	167 more per 1000 (from 12 fewer to 1000 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	3.9%	RR 5.35 (0.7 to 40.84)	170 more per 1000 (from 12 fewer to 1000 more)	Very low	Critical
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of change in size of pressure ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size or volume of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Osterbrink (2005) did not report the randomisation method, whether there was allocation concealment or blinding and there was insufficient reporting of incomplete outcome data. The authors did not report baseline ulcer size.

b: The confidence interval crossed both MID points and there were a limited number of events.

Table 50Clinical evidence profile: alternating-pressure cushion versus dry flotation cushion

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Alternating-pressure cushion	Dry flotation cushion	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with pressure ulcers completely healed – grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	3/14 (21.4%)	5/11 (45.5%)	RR 0.47 (0.14 to 1.56)	241 fewer per 1000 (from 391 fewer to 255 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	None	-	45.5%	RR 0.47 (0.14 to 1.56)	241 fewer per 1000 (from 391 fewer to 255 more)	Very low	Critical
Rate of healing (cm²/day) – grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	Serious^d	0.13 (SD 0.37)	0.27 (SD 0.56)	-	MD 0.14 lower (0.52 lower to 0.24 higher)	Very low	Critical
Rate of healing (cm³/day) – grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	Serious^d	0.56 (SD 0.86)	0.49 (SD 0.86)	-	MD 0.07 higher (0.61 lower to 0.75 higher)	Very low	Critical
% change in area per day– grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^c	Serious^d	2.56 (SD 7.86)	5.71 (SD 5.57)	-	MD 3.15 lower (8.42 lower to 2.12 higher)	Very low	Critical
% change in volume per day– grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b	Serious^d	1.00 (SD 1.83)	0.68 (SD 0.86)	-	MD 0.32 higher (0.76 lower to 1.4 higher)	Very low	Critical
Mortality– grade 2 ulcers or above, elderly adults, mean 51 days follow-up⁴⁴
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b^,^e	None	3/14 (21.4%)	1/11 (9.1%)	RR 2.36 (0.28 to 19.66)	124 more per 1000 (from 65 fewer to 1000 more)	Very low	Important
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Very serious^b^,^e	None	-	9.1%	RR 2.36 (0.28 to 19.66)	124 more per 1000 (from 66 fewer to 1000 more)	Very low	Important
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size or volume of pressure ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Clark (1998): unclear details of randomisation; unblinded observer; grading system of ulcers not specified. High drop-out.

b: Confidence interval crossed both MID points.

c: Confidence interval crossed 1 MID point.

d: No log transformation of data.

e: Limited number of events.

Table 51Clinical evidence profile: profiling bed versus foam mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Profiling bed	Foam mattress	Relative (95% CI)	Absolute	Quality	Importance
Proportion of people with completely healed pressure ulcers – any grade of pressure ulcers, surgical or medical adults, grade 1 occurred, 5-10 days follow-up⁸⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	4/4 (100%)	2/10 (20%)	RR 3.96 (1.28 to 12.24)	592 more per 1000 (from 56 more to 1000 more)	Very low	Critical
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	None	-	20%	RR 3.96 (1.28 to 12.24)	592 more per 1000 (from 56 more to 1000 more)	Very low	Critical
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Rate of change in size of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size or volume of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Keogh (2001) did nto report clear blinding and not all of the study's pre-specified outcomes were reported. Not all participants had pressure ulcers (only 14 had existing pressure ulcers), so there was a small sample size and uneven distribution, with only 4 in the experimental group. Grade 1 pressure ulcers analysed only. The authors did not address incomplete outcome data. There was a high drop out from study and it is not possible to identify how many of those who dropped-out had existing pressure ulcers at start of the trial.

b: There were a limited number of events.

Table 52Clinical evidence profile: constant force mattress versus low-air-loss mattress

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Constant force mattress	LAL mattress	Relative (95% CI)	Absolute	Quality	Importance
Mean % rate of closure per week – grade 3 or 4 ulcers (classification system not specified), long-term or subacute people in hospital from wards specialising in ventilator-dependent or extensive wound care needs, at 8 week follow-up²⁹
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	Serious^b	Serious^c	9 (s.d 4.8) n= 10	5 (s.d 3.7) n= 8	-	MD 4 higher (0.07 to 7.93 higher)	Very low	Critical
Proportion of people with complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Reduction in size and volume of pressure ulcer
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: The randomisation reported was inadequate and there was unclear allocation concealment and blinding. No details of incomplete outcome data, type of analysis, pressure ulcer sizes at baseline or classification of pressure ulcers were given. The study used a very small sample size.

b: The confidence interval crossed 1 MID point.

c: The data was not log transformed.

Table 53Clinical evidence profile: wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushion^b

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other	Wheelchair cushion with cyclic pressure-relief protocol	Standard wheelchair cushion	Relative (95% CI)	Absolute	Quality	Importance
Pressure ulcer closure (cm²)^c – stage 2 or 3 ulcers (classification system not specified), paraplegic or tetraplegic wheelchair users, 30 days follow-up¹⁰⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	78.5 (s.d 74.4) n=22	12.49 (s.d 52.0) n=22	p<0.001	MD 66.01 higher (28.08 to 103.94 higher)	Low	Critical
Pressure ulcer closure rate (cm²/day)^c – stage 2 or 3 ulcers (classification system not specified), paraplegic or tetraplegic wheelchair users, 30 days follow-up¹⁰⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	2.17 (s.d 1.46) n=22	2.3 (s.d 2.04) n=22	p<0.001	MD 1.94 higher (0.89 to 2.99 higher)	Low	Critical
PUSH score improvement^c – stage 2 or 3 ulcers (classification system not specified), paraplegic or tetraplegic wheelchair users, 30 days follow-up¹⁰⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	2.5 (s.d 2.3) n=22	0.7 (s.d 1.1) n=22	p=0.001	MD 1.8 higher (0.73 to 2.87 higher)	Low	Critical
% surface area reduction^c – stage 2 or 3 ulcers (classification system not specified), paraplegic or tetraplegic wheelchair users, 30 days follow-up¹⁰⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	45.0 (s.d 22.0) n=22	10.2 (s.d 34.9) n=22	p<0.001	MD 34.8 higher (17.78 to 51.82 higher)	Low	Critical
% PUSH score improvement^c – stage 2 or 3 ulcers (classification system not specified), paraplegic or tetraplegic wheelchair users, 30 days follow-up¹⁰⁸
1	Randomised trial	Very serious^a	No serious inconsistency	No serious indirectness	No serious imprecision	None	21.9 (s.d 24.6) n=22	5.8 (s.d 9.2) n=22	p=0.003	MD 16.1 higher (5.13 to 27.07 higher)	Low	Critical
Proportion of people with pressure ulcers completely healed
-	-	-	-	-	-	-	-	-	-	-	-	-
Time to complete healing of pressure ulcers
-	-	-	-	-	-	-	-	-	-	-	-	-
Pain (wound-related)
-	-	-	-	-	-	-	-	-	-	-	-	-
Time in hospital or NHS care
-	-	-	-	-	-	-	-	-	-	-	-	-
Patient acceptability
-	-	-	-	-	-	-	-	-	-	-	-	-
Side effects
-	-	-	-	-	-	-	-	-	-	-	-	-
Mortality
-	-	-	-	-	-	-	-	-	-	-	-	-
Health-related quality of life
-	-	-	-	-	-	-	-	-	-	-	-	-

a: Makhsous (2010) did not provide details of sequence generation, allocation concealment or blinding. The study used a small sample size.

b: The study included people who had a spinal cord injury and so would not be able to reposition themselves.

c: Change scores were presented in the paper.

Table 54Economic evidence profile: alternating pressure overlays verses alternating pressure mattress replacements verses high specification foam mattresses

Study	Applicability	Limitations	Other comments	Incremental cost	Incremental effects	Cost effectiveness	Uncertainty
Fleurance 2005⁶¹ (UK)	Partially applicable^a	Potentially serious limitations^b	A decision analytic model which compared 3 alternatives: alternating pressure overlays (AO), alternating pressure mattress replacements (AR), and high-specification foam mattresses (SC).	Superficial ulcers - 4 week horizon AR-AO = -£20 SC-AR = £100 Severe ulcers - 4 week horizon AR-AO = -£11 SC-AR = £56	QALYs Superficial ulcers -4 week horizon AR-AO = 0.00005 SC-AR = -0.00027 Severe ulcers - 4 week horizon AR-AO = 0.00002 SC-AR = -0.00011	Superficial - 4 week horizon AR dominates other options Severe - 4 week horizon AR dominates other options	The probability of AR being the most CE at a threshold of £20,000 was 64% for superficial ulcers at 1 and four weeks, and 61% and 61-62% for severe pressure ulcers at 1 and four weeks respectively.

a: UK NHS setting; cost year 2003.

b: Quality of life data is obtained from healthcare professionals rather than from patients, short time horizon may not capture full economic impact of these devices. Estimates of health effect estimated rather than obtained from the literature, baseline health outcomes not based on randomised data.

Table 55Unit costs

Device	Purchase cost	Rental cost	Source
High specification foam mattresses
Softform premiere	£199.00	NA	Correspondence with Invacare
Harvest Reflect 2 Replacement Mattress	£140.00	NA	Correspondence with Harvest healtcare
Harvest Prime Comfort Plus	£120.00	NA	Correspondence with Harvest healtcare
Pentaflex (4 way turn, acute)	£204.14	NA	Huntleigh
Constant low pressure
Breeze	£3,453.70	£12.85 per day^a	Huntleigh
Alternating pressure
Nimbus 3	£3,565.18	£13.56 per day^a	Huntleigh

a: Minimum of 10 day rental

Table 56Unit costs

Device	Cost	Source
High specification foam mattresses and overlays
Softform incubator pad (high specification foam)	£49.48	NHS supply chain catalogue¹
Softform cot mattress (high specification foam)	£107.63	NHS supply chain catalogue¹
Repose babytherm redistributing overlay (with pump)	£91.55	NHS supply chain catalogue¹
Repose paediatric mattress Overlay (with pump)	£91.55	NHS supply chain catalogue¹
Repose mattress overlay (with pump)	£106.11	NHS supply chain catalogue¹
Softform premiere	£199.00	Correspondence with manufacturer
Dynamic support surfaces
Nimbus paediatric mattress	£13.56 per day rental (purchase price £3,293)	Correspondence with manufacturer
Nimbus 3 mattress	£13.56 per day rental (purchase price £3,565)	Correspondence with manufacturer

Recommendations	18. Use high-specification foam mattresses for adults with a pressure ulcer. If this is not sufficient to redistribute pressure, consider the use of a dynamic support surface. 19. Do not use standard-specification foam mattresses for adults with a pressure ulcer. 20. Consider the seating needs of people who have a pressure ulcer who are sitting for prolonged periods. 21. Consider a high-specification foam or equivalent pressure redistributing cushion for adults who use a wheelchair or who sit for prolonged periods and who have a pressure ulcer.
Relative values of different outcomes	The GDG identified that the proportion of people with pressure ulcers completely healed, time to complete healing, reduction in size and volume and rate of reduction in size and volume of pressure ulcers were the most critical outcomes to inform decision making.
Trade off between clinical benefits and harms	One study showed that a low-air-loss bed was more clinically beneficial than a foam mattress overlay for complete healing. Another study showed no difference between a foam mattress overlay and a low-air-loss bed for grade 3 and 4 pressure ulcers. There was higher mortality in the low-air-loss bed group and there was a clinical benefit of low-air-loss bed for rate of ulcer healing, but no clinical difference for patient acceptability or reduction by 1 grade of pressure ulcer. Air-fluidised therapy beds showed some clinical benefit compared to standard or conventional therapies (these varied and included repositioning and massage, use of a sheepskin, gel pad, heel and elbow protector, alternating pressure mattress, alternating pressure pads, air-filled mattresses, water filled mattresses and pads or high density foam pads) for improvement in pressure ulcers. This study also showed a clinical benefit for the reduction in pain, time in hospital, patient satisfaction, and an increase in comfort. There was no clinical benefit for length of stay in hospital or reduction in mortality. There was no clinical difference between a water mattress overlay and a low-tech mattress for pressure ulcers completely healed but there were less people on the water mattress overlay with pain. There were no clinically beneficial results for alternating-pressure mattress overlays compared to alternating-pressure mattresses except for percentage change in surface area. In this study there were more deaths in the mattress group compared to the mattress overlay group. There was a clinical benefit for air-filled devices compared to alternating pressure mattress for proportion of people with pressure ulcers completely healed. There were varying results for 1 alternating-pressure mattress compared to another alternating-pressure mattress. This however was dependant on the exact device in use. There were no clear outcomes for low-air-loss beds compared to low-air-loss overlays. There was an uncertain clinical benefit for a constant force mattress when compared to a low-air-loss mattress. There was a large clinical benefit for a profiling bed when compared to a foam mattress for complete healing of pressure ulcers. An alternating-pressure cushion was more clinically beneficial than a dry flotation cushion for completely healing grade 2 or above pressure ulcers. There was less mortality in the dry flotation cushion group. A 3-D macroporous overlay compared to a gel overlay showed no clinical difference for pressure ulcers completely healed but a benefit of the 3-D macroporous overlay for reducing mortality, discomfort and suspension due to worsening of pressure ulcers or intolerance, and increasing improvement. A wheelchair cushion equipped with an individualised pressure-relief protocol was more clinically beneficial for the reduction in the size of pressure ulcers. Overall, there is no high quality evidence to suggest a benefit in healing for any particular type of device. However people with pressure ulcers are at risk of developing further pressure ulcers, thus the GDG felt it appropriate to recommend the use of high specification mattresses (which are widely used in current clinical practice) for all people as standard. This recommendation is reflective of preventation strategies for pressure ulcers. The GDG highlighted that standard foam mattresses should not be provided to people who have developed pressure ulcers and developed a recommendation to emphasise this. Wheelchair cushions Evidence from one study showed that a wheelchair cushion equipped with individualised cyclic pressure-relief protocol had more benefit when compared to a standard wheelchair cushion in spinal cord injured patients, however the GDG felt that the intervention was very complicated and combined a number of other interventions and may not be representative of clinical practice. This was a very small study of low quality and the GDG thought that the study may not have been long enough for patients to reach complete healing of ulcers and the loss to follow up may potentially be large. The GDG therefore chose to develop a recommendation similar to that included in the ‘prevention guideline’ to highlight that people who use a wheelchair who have pressure ulcers should be provided with a high specification foam cushion as a minimum. The GDG also felt that people who were likely to be seated for a long period, who have a pressure ulcer, should be provided with a high specification foam cushion and should have their seating requirements carefully considered.
Economic considerations	High specification foam mattresses are considered to be cost-effective for the prevention of pressure ulcers, compared to standard mattresses. It can therefore be inferred that it is cost-effective for people to remain on these mattresses once a pressure ulcer develops. The mattress will assist with on-going prevention, and will also aid pressure ulcer treatment. High specification foam mattresses are therefore considered to be cost-effective for the treatment of pressure ulcers, compared to standard mattresses. The GDG considered 1 cost-utility analysis which compared alternating pressure mattresses and alternating pressure overlays to high specification foam mattresses for the treatment of pressure ulcers. The study found that alternating pressure mattresses dominate alternating pressure overlays and high specification mattresses, with reduced costs and increased QALYs. However this study had potentially serious limitations and was considered to be only partially applicable. The GDG also considered the unit costs of various devices, and acknowledged that dynamic support surfaces are more costly than high specification foam mattresses. The GDG discussed that, although there was no clear clinical evidence, the crucial factor when healing pressure ulcers is pressure redistribution, and that the dynamic support surfaces do reduce pressure better than high specification foam. Efficient pressure redistributionwill improve quality of life and up front unit costs will be offset, or at least mitigated, by the reduction in further treatment costs. Based on the existing economic evidence, discussion of unit costs and potential benefits of dynamic surfaces, the GDG agreed that where high specification foam was not sufficient for the healing of a pressure ulcer, the use of dynamic support surfaces was likely to be cost effective.
Quality of evidence	The evidence reviewed was of varying populations, interventions, comparisons, and outcomes which made a conclusion of which type of device was preferential very difficult. The evidence was of low to very low quality due to study limitations and most outcomes had serious to very serious imprecision. The GDG discussed the evidence limitations. They further agreed that that none of the studies followed up people for a sufficient amount of time, which is likely to have limited the apparent effectiveness of the mattresses.
Other considerations	The GDG agreed that high specification foam mattresses are commonly used in the NHS for the prevention of pressure ulcers, thus the group considered that it would be unethical to discontinue the use of this device, if it had been started prior to the development of the pressure ulcer.

Recommendations	22. Use a high-specification cot or bed mattress or overlay for all neonates, infants, children and young people with a pressure ulcer.
Relative values of different outcomes	The GDG identified that the proportion of people with pressure ulcers completely healed, time to complete healing, reduction in size and volume and rate of reduction in size and volume of pressure ulcers were the most critical outcomes to inform decision making.
Trade-off between clinical benefits and harms	The GDG used 2 statements from the Delphi consensus panel to develop the recommendation, ‘Healthcare professionals should use a high specification cot or bed mattress for all neonates, infants and children who have developed pressure ulcers’ and ‘Healthcare professionals should use a high specification cot or bed overlay for all neonates, infants and children who have developed pressure ulcers’. The former was accepted during Round 1 of the Delphi consensus survey, the latter was not accepted. During Round 1, qualitative comments received on the use of high specification bed mattresses highlighted that the provision of a pressure redistributing device should be tailored to the child and that selection should consider the physical, clinical and environmental situation. Other comments emphasised that the use of any pressure redistributing device should only be used in combination will a repositioning regimen. The GDG subsequently discussed the latter statement on the use of overlays as well as the comments received during Round 1, which focused on the benefits of using an overlay where a mattress is unavailable. In particular, comments noted that the use of an overlay would be preferable to delaying pressure redistribution. However, comments also highlighted that there were potential safety issues in the use of certain overlays, particularly where this raises the height of the child above the bed rails. The statement was therefore amended to highlight that an overlay may be considered where a mattress is unavailable but safety should be considered where this is used. The GDG therefore amended the statement to ‘Healthcare professionals should consider the use of a high specification cot or bed overlay for neonates, infants and children who have developed pressure ulcers, where a high specification mattress is not available, taking into account safety’ for inclusion in Round 2 of the Delphi consensus survey. The amended statement was included in Round 2 of the survey and was agreed by the panel. Qualitative responses gathered in Round 2 emphasised that any overlay used should be high specification and noted that repositioning should still be considered a mainstay of care, regardless of the use of a pressure redistributing device. The GDG discussed both of the agreed statements. The GDG felt that high specification mattresses should be provided as standard to all neonates, infants, children and young people who have developed a pressure ulcer, as the potential benefits in reducing pressure were likely to outweigh any harms. The GDG discussed further the use of high specification overlays. The GDG noted that it was likely that high specification mattresses would be provided to people admitted to secondary care, as these would be given as standard preventative treatment. The GDG therefore felt that the recommendation should provide the option of providing people with pressure ulcers a high specification bed mattress or overlay. The GDG felt, although the Delphi consensus panel had agreed a statement that overlays used only be used in the absence of a mattress, that recommending the provision of either device would help to ensure that pressure was reduced across different settings in which care was provided (for example, in the operating theatre, where care is provided in the home, or in long term residential settings). The GDG felt that the option to provide an overlay to those who had developed a pressure ulcer should remain as pressure reduction was the most urgent consideration in this situation and, if the use of an overlay was to facilitate the reduction of pressure, this would be important where there is a delay in providing a high specification foam mattress. Furthermore, the GDG noted that some populations (for example, neonates in neonatal intensive care units) were often provided with high specification cot or bed overlays as standard.
Economic considerations	There are costs associated with high specification foam cots and mattresses and overlays. The estimated purchase costs are £50-£199 (typical products identified by GDG members), and the devices can be used over a number of years, therefore the expected cost per patient is low. The GDG considered these costs likely to be offset by the benefits of the intervention in terms of improvement in the person's quality of life, and reduction in future treatment costs through improved healing.
Quality of evidence	No RCTs or cohort studies were identified for neonates, infants, children or young people. Formal consensus using a modified Delphi was therefore used to develop the recommendation. To inform the recommendation, the GDG used 2 statements which were included in Round 1 of the Delphi consensus survey and reached 83% and 64% consensus agreement. The latter statement was therefore included in Round 2 of the survey, where it reached 86% consensus agreement. Further details can be found in Appendix N.
Other considerations	The GDG highlighted that healthcare professionals should be aware of potential safety considerations in the use of overlays for neonates, infants, children and young people, where the use of an overlay causes the child to be above the height of the bed rails.

Recommendations	23. If pressure on the affected area cannot be adequately relieved by other means (such as repositioning), consider a dynamic support surface, appropriate to the size and weight of the child or young person with a pressure ulcer, if this can be tolerated.
Relative values of different outcomes	The GDG identified that the proportion of people with pressure ulcers completely healed, time to complete healing, reduction in size and volume and rate of reduction in size and volume of pressure ulcers were the most critical outcomes to inform decision making.
Trade-off between clinical benefits and harms	The GDG used 1 statement from the Delphi consensus panel to develop the recommendation ‘Healthcare professionals should not use dynamic support surface for the treatment of pressure ulcers in neonates, infants and children’. The statement was not accepted during Round 1 of the Delphi consensus survey and was therefore amended for inclusion in Round 2 of the survey. The GDG discussed the use of dynamic support surfaces and the comments received during Round 1, which focused upon considering the appropriateness of a dynamic support surface, taking into account a child's weight, clinical condition and tolerability. The statement was therefore amended to highlight that a dynamic support surface may be considered however, any decision should account for these factors. Additionally, the GDG identified that dynamic support surfaces may be appropriate for both children and young people, depending upon individual factors. Therefore the statement was also amended to include children. The GDG therefore developed the statement ‘Healthcare professionals should consider the use of a dynamic support surface for children and young people who have developed pressure ulcers, where this can be tolerated, if pressure on the affected area cannot be relieved by other means (such as repositioning). The support surface should be appropriate for the size and weight of the child’ which was included in Round 2 of the survey, where it was accepted. The GDG discussed the agreed statement and agreed that a recommendation should be made. Qualitative comments received during round 2 of the survey noted that there were some dynamic support surfaces which would be appropriate for use in infants and children. Comments received from panel members generally felt the use of a dynamic support surface may be appropriate where pressure could not be relieved by any other means, for example repositioning. Comments also highlighted that individual patient factors should be taken into account when assessing the potential benefits that use of a dynamic support surface might bring. The GDG therefore felt that on balance, there were potential benefits of using a dynamic support surface for children and young people who have developed pressure ulcers, where pressure cannot be relieved by other means (for example, where a child cannot be repositioning). Any consideration of using a dynamic support surface should account for the individual patient factors and to minimise any potential harms, the size and weight of the child or young person should be carefully considered and an appropriate dynamic support surface selected.
Economic considerations	Dynamic support surfaces are more costly than high specification devices, for example the Nimbus paediatric mattress which can be rented for £13.56 per day. Pressure redistribution is crucial for the treatment of pressure ulcers, and is greatly facilitated by such dynamic devices. Therefore the GDG agreed that the cost of these devices will likely be offset by improvements in quality of life and reduction in further treatment costs. Note that people with pressure ulcers are considered to be at high risk of developing further pressure ulcers, therefore these devices may also help to prevent further pressure ulcers.
Quality of evidence	No RCTs or cohort studies were identified for neonates, infants, children or young people. Formal consensus using a modified Delphi was therefore used to develop the recommendation. To inform the recommendation, the GDG used 1 statement which was included in Round 1 of the Delphi consensus survey, where it reached 12% consensus. The statement was therefore amended for Round 2, where it was accepted at 95% consensus agreement. Further details can be found in Appendix N.
Other considerations	The GDG emphasised that any dynamic support surface used for a child or young person should be appropriate to the size and weight of the child, to prevent safety issues and ensure optimum effectiveness.

Recommendations	24. Consider using specialist support surfaces (including dynamic support surfaces where appropriate) for neonates, infants, children and young people with pressure ulcers, taking into account their current pressure ulcer risk and mobility. 25. Tailor the support surface to the location and cause of the pressure ulcer for neonates, infants, children and young people.
Relative values of different outcomes	The GDG identified that the proportion of people with pressure ulcers completely healed, time to complete healing, reduction in size and volume and rate of reduction in size and volume of pressure ulcers were the most critical outcomes to inform decision making.
Trade-off between clinical benefits and harms	The GDG used 1 statement from the Delphi consensus panel to develop the recommendation ‘Healthcare professionals should not use a standard foam cot/bed mattress for neonates, children, infants or young people who have previously developed pressure ulcers and should use specialist patient support surfaces as clinically indicated.’ The statement was not accepted during Round 1 of the Delphi consensus survey and was therefore amended for inclusion in Round 2 of the survey. The GDG discussed comments received during Round 1 which highlighted that this would depend upon the reason for initial pressure ulcer development. The GDG agreed that, given pressure ulcers caused by devices were not included in the current guideline, standard foam mattresses should not be used for those who have developed a pressure ulcer previously, given this would mean that they were at risk of subsequent pressure ulcer development. The GDG therefore amended the statement to reflect that these should not be used routinely, however current risk level should be considered when choosing a specialist support surface for this population. The GDG therefore amended the statement for inclusion in Round 2 of the Delphi consensus to ‘Healthcare professionals should not routinely use a standard foam cot/bed mattress for neonates, children, infants or young people who have previously developed pressure ulcers and should consider using specialist support surfaces, taking into account current risk level and mobility.’ The statement was accepted. The GDG discussed the agreed statement and agreed that a recommendation should be made. Qualitative comments received during round 2 of the survey generally felt that high specification foam mattresses may be appropriate for this population however, standard foam would not be appropriate for infants, children and young people who had previously had a pressure ulcer. Some panel members noted that some people who had a pressure ulcer were likely to be a lesser risk as their risk was related to an acute situation. The GDG acknowledged this but did not feel that standard foam mattresses should be used for anyone who had previously developed a pressure ulcer, due to the increased risk of developing another pressure ulcer. The GDG also noted that many people in secondary care would be provided with a high specification foam mattress as standard care. The GDG felt that the benefits of providing an alternative foam mattress over a standard foam mattress outweighed any potential harms. A recommendation was therefore developed to reflect this and to highlight that specialist support surfaces, for example, dynamic support surfaces, would be more appropriate for this population. The GDG highlighted that provision of any support surface should take into account current risk level and mobility, as highlighted by comments from the Delphi consensus panel, who noted that a child's level of pressure ulcer risk may vary, depending on the risk factors. Comments from the Delphi consensus panel also highlighted the importance of ensuring that the support surface chosen was appropriate to the location and cause of the pressure ulcer. For example, the GDG noted that some pressure redistributing devices may increase pressure in at a risk site, whilst decreasing the pressure elsewhere. The GDG therefore felt that it was important to highlight that the choice of pressure redistributing surface should be tailored to the location and cause of the pressure ulcer. It was noted that this was particularly important given the specific sites considered to be at risk in neonates, infants, children and young people, for example, the head and scalp. The GDG therefore chose to develop a recommendation to reflect this.
Economic considerations	Use of high specification and dynamic support surfaces is considered to be cost-effective for individuals with pressure ulcers and for those at significant risk of developing them; therefore standard foam mattresses are not considered to be an efficient use of resources. Furthermore, high specification foam and dynamic surfaces are current best practice and therefore this is not thought to require a substantial increase in resource. Current risk level, mobility, location, and cause of pressure ulcer should be taken into account when selecting the device in order to ensure that an effective device is implemented and clinical and economic benefits are realised as soon as possible.
Quality of evidence	No RCTs or cohort studies were identified for neonates, infants, children or young people. Formal consensus using a modified Delphi was therefore used to develop the recommendation. To inform the recommendation, the GDG used 1 statement which was included in Round 1 of the Delphi consensus survey, where it reached 72% consensus. The statement was therefore amended for Round 2, where it was accepted at 89% consensus agreement. Further details can be found in Appendix N.
Other considerations	There were no other considerations.

6Pressure redistributing devices

6.1. Introduction

6.2. Review question: What are the most clinically and cost effective pressure redistributing devices for the management of pressure ulcers?

6.2.1. Clinical evidence (adults)

6.2.1.1. Low tech constant pressure devices

6.2.1.2. High-tech pressure devices

6.2.1. Economic evidence (adults)

Published literature

Unit costs

6.2.2. Clinical evidence (neonates, infants, children and young people)

6.2.3. Economic evidence (neonates, infants, children and young people)

Published literature

Economic considerations

6.2.4. Evidence statements

6.2.4.1. Clinical (adults)

6.2.4.1.1. Water mattress overlay versus low-tech mattress

6.2.4.1.2. 3-D macroporous overlay versus gel overlay

6.2.4.1.3. Low-air loss-bed versus low-tech foam mattress overlay

6.2.4.1.4. Low-air-loss bed versus low-air-loss overlay

6.2.4.1.5. Air-fluidised therapy (AFT) versus standard/conventional therapies

6.2.4.1.6. Alternating-pressure mattress (Nimbus 1) versus. alternating-pressure mattress (Pegasus Airwave)

6.2.4.1.7. Alternating-pressure mattress (Nimbus 3) with cushion (Aura) and 4-hourly turning versus an alternating-pressure mattress (Pegasus Cairwave Therapy System) with seating cushion (Proactive 2) and 8-hourly turning

6.2.4.1.8. Alternating-pressure mattress (Nimbus 3) versus alternating-pressure mattress (Pegasus Biwave, Pegasus Airwave, Pegasus Cairwave or AlphaXCell) or alternating-pressure mattress overlay (AlphaXCell or Quattro)

6.2.4.1.9. Alternating-pressure mattress overlay versus alternating-pressure mattress

6.2.4.1.10. Air-filled devices versus alternating-pressure mattress

6.2.4.1.11. Alternating-pressure cushion versus dry flotation cushion

6.2.4.1.12. Profiling bed versus foam mattress

6.2.4.1.13. Constant force mattress versus low-air-loss mattress

6.2.4.1.14. Wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushion

6.2.4.2. Economic (adults)

6.2.4.3. Clinical (neonates, infants, children and young people)

6.2.4.4. Economic (neonates, infants, children and young people)

6.3. Recommendations and link to evidence

6.3.1. Adults

6.3.2. Neonates, infants, children and young people

Publication Details

Copyright

Publisher

NLM Citation

Summary of included studies

Table 42Clinical evidence profile: water mattress overlay versus low-tech mattress

Table 43Clinical evidence profile: 3 dimensional macroporous overlay versus gel overlay

Table 44Clinical evidence profile: low-air-loss bed versus low-tech foam mattress overlay

Table 45Clinical evidence profile: low-air-loss bed versus low-air-loss overlay

Table 46Clinical evidence profile: air-fluidised therapy (AFT) versus standard or conventional therapies

Table 47Clinical evidence profile: alternating-pressure mattress versus alternating-pressure mattress

Table 48Clinical evidence profile: alternating-pressure mattress overlay versus alternating-pressure mattress

Table 49Clinical evidence profile: air-filled devices versus alternating pressure mattress

Table 50Clinical evidence profile: alternating-pressure cushion versus dry flotation cushion

Table 51Clinical evidence profile: profiling bed versus foam mattress

Table 52Clinical evidence profile: constant force mattress versus low-air-loss mattress

Table 53Clinical evidence profile: wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushionb

Table 54Economic evidence profile: alternating pressure overlays verses alternating pressure mattress replacements verses high specification foam mattresses

Table 55Unit costs

Table 56Unit costs

Table 53Clinical evidence profile: wheelchair cushion with equipped with individualised cyclic pressure-relief protocol versus standard wheelchair cushion^b