| 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 | ||
| Proportion of people with pressure ulcers completely healed - Shea grade 2 ulcers or above, people in an elderly nursing home, mean 36 days follow-upg59 | ||||||||||||
| 1 | Randomised trial | Very seriousa | No serious inconsistency | No serious indirectness | Seriousb | 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 |
| - | 46.3% | 139 more per 1000 (from 60 fewer to 444 more) | ||||||||||
| 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-upg123 | ||||||||||||
| 1 | Randomised trial | Very seriousc | No serious inconsistency | No serious indirectness | Very seriousd | 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 |
| - | 16.7% | 5 fewer per 1000 (from 124 fewer to 431 more) | ||||||||||
| 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 seriousa,c | No serious inconsistency | No serious indirectness | Seriousb | 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 |
| - | 31.5% | 79 more per 1000 (from 50 fewer to 271 more) | ||||||||||
| 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 seriousc | No serious inconsistency | No serious indirectness | Very seriousd | 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 |
| - | 27.8% | 44 more per 1000 (from 147 fewer to 517 more) | ||||||||||
| 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 seriousa | No serious inconsistency | No serious indirectness | Very seriousf | 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 seriouse | No serious inconsistency | No serious indirectness | Seriousb | Serioush | 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 seriouse | No serious inconsistency | No serious indirectness | No serious | Serioush | 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 seriouse | 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 seriousa | 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 |
| - | - | 17.1% | 86 more per 1000 (from 62 fewer to 426 more) | |||||||||
| Time to complete healing of pressure ulcers | ||||||||||||
| - | - | - | - | - | - | - | - | - | - | - | - | - |
| Pain (wound-related) | ||||||||||||
| - | - | - | - | - | - | - | - | - | - | - | - | - |
| Time in hospital or NHS care | ||||||||||||
| - | - | - | - | - | - | - | - | - | - | - | - | - |
| Side effects | ||||||||||||
| - | - | - | - | - | - | - | - | - | - | - | - | - |
| Health-related quality of life | ||||||||||||
| - | - | - | - | - | - | - | - | - | - | - | - | - |
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.
The confidence interval crossed 1 MID point.
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.
The confidence interval crossed both MID points.
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.
There was not enough data to put in Revman.
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.
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.
A non-parametric test (Wilcoxon rank-sum) was used but there was no log transformation of data.
From: 6, Pressure redistributing devices
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.