do a cochrane review - o-wm.com · 36 ostomy wound management® april 2013 ar...

36 ostomy wound management® April 2013 www.o-wm.com

Feature

Pressure-redistribution surfaces for Prevention of surgery-related Pressure ulcers: a meta-analysisHai-Yan Huang, RN, MSc; Hong-Lin Chen, MSc; and Xu-Juan Xu, RN, MSc

abstractPressure-redistribution surfaces are generally recommended to prevent pressure ulcers (PUs) in high-risk patients, but their use in surgery-related PU prevention remains controversial. A meta-analysis was conducted to assess the relative preventive impact of pressure-redistribution surfaces versus standard hospital mattresses (usually a hospital-issue, foam-based mattress) on the incidence of surgery-related PUs. Systematic literature searches were performed using the terms pressure ulcer, operation, surgery, mattress, foam, polymer, pad, overlay, surface, and interface. Country, race, language, and publication year of articles was not restricted; randomized or quasi-randomized controlled trials were eligible for analysis. Odds ratio (OR) with 95% confidence intervals (CIs) for surgery-related PU incidence in patients using support surfaces versus standard mattress were calculated by random-effects model. Of the 316 studies identi-fied, 10 involving a total of 1,895 patients were eligible for inclusion in the meta-analysis. Seven studies were random-ized, controlled and three were quasi-randomized controlled trials. Patients who were provided a support surface had a significantly decreased incidence of surgery-related PUs (OR 0.31 [95% CI 0.17–0.59]) compared to patients using a standard mattress. Subgroup analysis showed pressure-redistribution surfaces used intra-operatively did not decrease the incidence of surgery-related PUs (OR 0.59, [95% CI 0.34–1.01]), but PU incidence decreased with postoperative (OR 0.07 [95% CI 0.01–0.49]) as well as with intra-operative and postoperative use (OR 0.20 [95% CI 0.06–0.73]). Funnel plot diagrams suggest a minimal risk of bias. Sensitivity analysis did not materially change the result of the main meta-analysis. Postoperative use of pressure-redistribution surfaces can effectively decrease the incidence of surgery-related PUs, but evidence to substantiate intra-operative use is insufficient. Patients at high risk for surgery-related PUs should be placed on a pressure-redistribution surface during the postoperative period, but intra-operative use can remain pru-dent until more well-designed, adequately powered, urgently needed studies are performed.

Keywords: pressure ulcer, surgery, beds, operating table, meta-analysis

Index: Ostomy Wound Management 2013;59(4):36–48

Potential Conflicts of Interest: none disclosed

Ms. Huang is a registered nurse, Department of Obstetrics and Gynecology; Mr. Chen is a lecturer, School of Nursing; and Ms. Xu is Director of Nursing, Affiliated Hospital of Nantong University, Nantong City, Jiangsu Province, China. Please address correspondence to: Xu-Juan Xu, Affiliated Hospital of Nantong University, Department of Nursing, Xi Si Road 20#, Nantong City, Jiangsu Province 226001 China PR; email: [email protected].

IntroductionThe patient who undergoes a long complicated surgery is

potentially at risk for development of pressure ulcers (PUs). A US survey1 of 1,543 patients from 33 of 50 states indicated the overall incidence of PUs in surgical patients was 8.5% (95% confidence interval [CI]: 6.1% to 10.9%). The situation is similar in other countries, with reported prevalence rates of 14.3% in Sweden2 and 21.2% in Netherlands.3 Surgery-related PUs are associated with adverse patient outcomes and may contribute to patient pain, depression, loss of func-tion and independence, increased incidence of infection and

sepsis, additional surgical interventions, prolonged hospital stays, and increased costs of care.4,5

Prevention usually is considered the most efficient meth-od to address the problem; numerous studies have investi-gated PU prevention strategies. A systematic review6 that included 59 randomized controlled trials (RCTs) indicated using support surfaces, repositioning the patient, optimizing nutritional status, and moisturizing sacral skin are appropri-ate strategies to prevent PUs. Pressure-redistribution surfaces can reduce local pressure and decrease the incidence of PUs. A Cochrane review7 specifically assessed support surfaces DO N

OT DUPLIC

ATE

April 2013 ostomy wound management® 37www.o-wm.com

suPPort surFaCe meta-analysIs

used for PU prevention. The evidence showed people at high risk of developing PUs should use higher-specification foam mattresses rather than standard hospital foam mattresses. It is well accepted that pressure-redistribution surface use is an effective prevention strategy for PU.

However, support surface use to prevent surgery-related PUs remains controversial. Some studies showed a mattress overlay or pressure-relieving overlay placed on an operat-ing table is an effective prevention strategy. In a sequential, randomized controlled trial with 446 general, vascular, and gynecological surgical patients, Nixon et al8 reported a sig-nificant reduction in the odds of developing a PU using a dry viscoelastic polymer pad as compared to the standard oper-ating table mattress (22 out of 205, [11%] versus 43 out of 211 [20%], respectively; P = 0.007). In a experimental study, Schultz et al9 randomly assigned 413 surgical patients to re-ceive “usual perioperative care” or a new mattress overlay and found after six postoperative days, 89 patients (21.5%) developed PUs, and patients with PUs who used the new mattress overlay had statistically fewer new ulcers (34 out of 207, [16.4%] versus 55 out of 206 [26.7%], respectively; P <0.02). Postoperatively used pressure-redistribution sur-faces also can decrease the incidence of surgery-related PUs. In a retrospective study, Jackson et al10 provided air-fluid-ized therapy beds to postoperative cardiovascular patients, and only one out of 27 patients developed a PU, compared to 40 PUs in 25 patients not using the intervention. In a randomized, controlled trial conducted among post-hip fracture patients, Donnelly et al11 reported 31 out of 119 (26%) in the control group developed PUs, compared with eight out of 120 in the pressure-redistributing support sur-face group (7%, P <0.001).

Other studies noted effective prevention trends for pres-sure-redistribution surface use that did not reach statistical significance. In a randomized controlled trial, Russell and Lichtenstein12 reported that two out of 98 (2.0%) persons undergoing cardiovascular surgery using a dynamic pres-sure system developed a PU, compared to seven out of 100 (7.0%) in the conventional management group. (χ2=2.806, P = 0.094). In a retrospective case-control trial, Sewchuk et al13 reported eight out of 100 (8.0%) PUs in cardiac surgery patients using a pressure-redistributing support surface, compared to nine out of 50 (18.0%) using conventional management (χ2=3.317, P = 0.069). Chalian et al14 reported no ulcers among 20 head and neck surgery patients (0.0%) provided pressure-redistributing support surfaces, compared to four out of 19 (21.1%) provided conventional manage-ment (χ2=3.84, P=0.050). One recent randomized, controlled trial15 of 175 patients found adverse results: patients laying on the foam overlay experienced slightly more PUs (17.6%) than patients on the standard OR table without the foam overlay (11.1%); the RCT was terminated per these results.

Because a decision was needed in the authors’ facility regard-ing the use of pressure-redistribution surfaces for preventing

surgery-related PUs, a meta-analysis was conducted to sum-marize the evidence collected on pressure-redistribution sur-faces compared to standard mattresses for surgical patients and to assess the PU incidence of PUs when pressure-redistri-bution surfaces were used for surgery-related PU prevention.

methodsInclusion and exclusion criteria. Studies were eligible for

inclusion in this analysis if they met the following criteria: 1) study type: RCTs and quasi-randomized controlled trials; 2) types of participants: high-risk, surgery related-PU patients (high-risk surgeries include cardiac surgery, general surgery, orthopedic surgery, vascular surgery, or surgeries with an ex-pected operative time of more than 3 hours); 3) types of inter-vention: studies comparing pressure-redistribution surfaces to standard mattress; and 4) types of outcomes: studies that assessed the incidence of surgery-related PUs were included. Excluded were studies involving patients in medical wards, mixed wards, and the community, as well as cohort studies, case-control studies, cross-sectional studies, and studies that compared different kinds of pressure-redistribution surfaces to each other (not to standard mattresses).

Search strategy. A search was conducted using MEDLINE (PubMed, www.ncbi.nlm.nih.gov/pubmed/), and Web of Science (http://webofknowledge.com/) from their inception to May 2012. The search strategy included the terms pressure ulcer, operation, surgery, mattress, foam, polymer, pad, overlay, surface, and interface. The search strategy for MEDLINE and Web of Science are listed in Table 1. There were no search restrictions related to country, race, language, or year of pub-lication. Conference proceedings and scanned references of retrieved articles also were hand-searched to identify any ad-ditional relevant studies.

Selection of eligible studies. First, two reviewers indepen-dently identified randomized or quasi-randomized controlled trials through title or abstract; based on inclusion and exclu-sion criteria, eligible studies were included through abstract

Key Points• Theauthorsanalyzeddatafrom10clinicalstudiesto

evaluate the effectiveness of pressure-redistribution surfaces for the prevention of pressure ulcers (PUs) in surgical patients.

• Resultssuggestpostoperative,butnotintra-opera-tive, use of support surfaces significantly reduce the risk of PUs.

• Theauthorsrecommendroutineuseofpressure-redistribution surfaces during postoperative care and that additional studies to investigate their efficacy during surgery should be conducted.

ostomy wound management 2013;59(4):36–48

DO NOT D

UPLICATE


Feature

or full text. Discrepancies as to study inclusion were resolved by discussion between the two reviewers, and unresolved dis-agreements were referred to a third reviewer.

Quality assessment. The quality of included studies was de-termined using the Cochrane Collaboration’s tool for assessing risk of bias.16 The tool included randomized sequence genera-tion, allocation concealment, blinding of participants and per-sonnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. Items in the risk-of-bias assessment were judged adequate (+), unclear (?), or having po-tential for bias (−) for each study. Two reviewers independently assessed the quality. Disagreements were resolved by discussion and by seeking the opinion of a third reviewer.

Data extraction. Information about methodological char-acteristics (study design, randomization method, allocation concealment, blinding, follow-up, baseline comparability, and other study characteristics) study year, country, surgical population, age, surgical duration, type of pressure-reducing mattress, timing of reporting incidence, pressure-reducing mattress applied time, and PU incidence in two groups was extracted independently by two reviewers. The data were ex-tracted to the pre-designed forms. Any differences of opinion were resolved by discussion and consensus reached by discus-sion with a third reviewer.

Statistical analysis. A meta-analysis was performed using Review Manager (RevMan) software (version 5.0.21; Update Software Ltd, Oxford, Oxon, UK). Statistical heterogeneity was explored by χ2 and inconsistency (I2) statistics; an I2 value of 50% or more represented substantial heterogeneity.17 If no het-erogeneity was found, a fixed-effects model was used for meta-analysis; otherwise, a random effect model based on the Der Simonian and Laird estimator was used.18 Summary OR and 95% CI estimates were calculated by taking a weighted average of individual study results. Overall effects were determined us-ing the Z test. Two-sided P <0.050 was considered statistically significant. Potential publication bias was evaluated by funnel

plot. Asymmetry in funnel plots indicates publication bias in meta-analysis. Subgroup meta-analyses were performed in in-tra-operative and postoperative subgroups. The intraoperative period begins when the patient is transferred to the operating room bed and ends with the transfer of a patient to the postan-esthesia care unit (PACU). The postoperative period begins af-ter the transfer to the PACU and terminates with the resolution of the surgical sequelae. Sensitivity analysis was performed by removal of three quasi-randomized controlled trials.

resultsStudy characteristics. Ten (10) studies met the inclusion

criteria and provided sufficient data for meta-analysis.8-15,19,20 Figure 1 shows the stages used in identifying studies for in-clusion in the review. Study characteristics from the 10 ar-ticles included in meta-analyses are shown in Table 2.

Five studies9,10,13,14,19 were conducted in the US; two8,11 in the UK; and Germany,15 Canada,12 and the Netherlands18 each contributed one study. Seven (7) studies followed an RCT design8,9,11,12,15,19,20; the other three studies were quasi-ran-domized controlled trials.10,13-14 Pressure-reducing mattresses involved in the studies included air- or fluidized-therapy beds,10,13,14,19 pressure-redistributing support surfaces,9,11,20

dry viscoelastic polymer,8 a multicell pulsating dynamic mat-tress,12 and thermo-active viscoelastic foam overlays.15 The control mattresses used for comparison all were standard foam mattresses. In terms of study time frame, five studies used the mattresses intra-operatively,8,9,13,15 three involved postoperative use,10,11,20 and in two studies mattresses were used intra- and postoperatively.12,19

Risk of bias. One RCT was rated as level 1b evidence with low risk of bias, six RCTs as level 1d evidence with high risk of bias, and the remaining three quasi-randomized controlled

table 1. search strategy for Pubmed

1.PressureUlcer[MeshTerms]

2.SurgicalProcedures,Operative[MeshTerms]

3.Mattress[Tw]orFoam[Tw]orPolymer[Tw]orPad[Tw]orOverlay[Tw]orSurface*[Tw]orInterface[Tw]

4. #1 and #2 and #3

The search strategy for Web of Science

1.TS=(PressureSAMEUlcer*)orTS=(PressureSAMESore*)orTS=(BedSAMESore*)orTS=(DecubitusSAMEUlcer*)

2.TS=Surg*orTS=Operat*

3.TS=MattressorTS=FoamorTS=PolymerorTS=PadorTS=OverlayorTS=Surface*orTS=Interface

4. #1 and #2 and #3

Tw = Text Words; Ts = Topics; * = character substitution

Figure 1. Study selection flow diagram.

DO NOT D

UPLICATE


Featureta

ble

2. Cha

ract

eris

tics

of st

udie

s in

clud

ed in

met

a-an

alys

is

PU

inci

den

ce (%

)

Stu

dy,

Yea

rC

ount

ryS

tud

y d

esig

nS

urg

ical

p

op

ula

tio

nA

ge

(yea

rs)

Su

rgic

al

du

rati

on

Pre

ssu

re-

red

istr

ibu

tion

surf

aces

Tim

ing

of

rep

ort

ing

in

cid

ence

Pre

ssu

re-

red

istr

ibu

tio

n su

rfac

es

app

lied

tim

e

Pre

ssu

re-

red

istr

i-b

uti

on

surf

aces

g

rou

p

Sta

n-

dar

d

mat

-tr

ess

gro

up

Don

nelly

et

al, 2

01111

UK

RC

TH

ip f

ract

ure

op

erat

ed

with

hem

i art

hro

pla

sty

dyn

amic

hip

scr

ew o

r ot

her

>65

<2

hour

s 17

6

>2

hour

s 55

Hee

l ele

vatio

n p

lus

pre

ssur

e-re

dis

trib

utin

g su

pp

ort

sur

face

Day

10

–12

Po

stop

erat

ivel

y6.

7%

(8/1

20)

26.1

%

(31/

119

)

Jack

son

et

al, 2

01110

US

Qua

si-

rand

omiz

ed

tria

l

Car

dio

vasc

ular

sur

ger

y p

atie

nts

with

req

uire

d

vaso

pre

sso

rs fo

r at

leas

t 24

ho

urs

or

req

uire

d m

echa

nica

l ve

ntila

tion

for

at le

ast

24 h

our

s p

ost

op

era-

tivel

y

Ran

ge:

41–8

9M

ean

: 6

hour

sA

ir-flu

idiz

ed

ther

apy

bed

sD

urin

gI C

UP

ost

oper

ativ

ely

3.6%

(1

/28

)10

0.0%

(2

5/2

5)

Feuc

htin

ger

et a

l, 20

0615

Ger

man

RC

TC

ard

iac

surg

ery

with

E

xtra

corp

ore

alc

ircul

a-tio

n

Ran

ge:

33–9

2 M

ean

: 68

>1.5

ho

urs

4-cm

the

rmoa

c-tiv

e vi

sco

elas

tic

foam

ove

rlays

Day

1–2

In

tra-

oper

ativ

e17

.6%

(1

5/8

5)

11.1

%

(10

/90

)

Sew

chuk

et

al, 2

00613

US

Qua

si-

rand

omiz

ed

tria

l

Car

dia

c su

rger

yR

ange

: 40

–94

Mea

n:

224

–725

m

inut

es

Flui

d p

ress

ure-

red

ucin

g O

R b

ed

mat

tres

s

Day

1–2

8In

tra-

oper

ativ

e8.

0%

(8/1

00

)18

.0%

(9

/50

)

Cha

lian

et

al,

2001

14

US

Qua

si-

rand

omiz

ed

tria

l

Hea

d a

nd n

eck

surg

er-

ies

Ran

ge:

36–8

1M

ean

: 11

–23.

5 m

inut

es

Flui

d p

ress

ure-

re

duc

ing

OR

m

attr

ess

(RIK

)

Day

1–3

Intr

a-op

erat

ive

0.0%

(0

/20

)21

.1%

(4

/19

)

Rus

sell

et a

l, 20

0012

Can

ada

RC

TC

ard

iova

scul

ar s

urg

ery

Mea

n: 6

5>

3 ho

urs

Mul

ticel

l pul

sat-

ing

dyn

amic

m

attr

ess

Day

1–7

Intr

a-op

erat

ive

and

po

stop

era-

tivel

y

2.0%

(2

/98

)7.

0%

(7/1

00

)

Aro

novi

tch

et a

l, 19

9919

US

RC

TC

ard

ioth

ora

cic,

uro

-lo

gic

, or

vasc

ular

Mea

n: 6

4>

3 ho

urs

Dry

pol

ymer

ov

erla

ys, P

RM

S

afte

r su

rger

y

Day

1–7

Intr

a-op

erat

ive

and

po

stop

era-

tivel

y

1.1%

(1/9

0)

8.8%

(7

/80

)

Sch

ultz

et

al, 1

9999

US

RC

TS

ched

uled

sur

ger

y25

–91

Mea

n: 6

6>

2 ho

urs

Foam

ove

rlays

Day

1–6

Intr

a-op

erat

ive

16.4

%

(34

/207

)26

.7%

(5

5/2

06)

Nix

one

tal

,19

988

UK

RC

TG

ener

al\v

ascu

lar

and

g

ynec

olo

gic

al55

–69

(56%

) 70

+ (4

4%)

>1.5

ho

urs

Dry

vis

coel

astic

p

olym

er p

adD

ay1

Intr

a-op

erat

ive

10.7

%

(22/

205

)20

.4%

(4

3/2

11)

Hof

man

et

al, 1

99420

Net

her-

land

sR

CT

Fem

ora

l-ne

ck f

ract

ure

op

erat

ed w

ith A

O-

scre

ws,

dyn

amic

hip

sc

rew

, gam

ma

nail,

ar-

thro

pla

sty,

bla

de-

pla

te

Mea

n: 8

0N

ot m

en-

tione

dP

ress

ure-

de-

crea

sing

mat

-tr

esse

s

Wee

k 1

Po

stop

erat

ivel

y25

.0%

(5

/20

)63

.7%

(1

4/2

2)

DO NOT D

UPLICATE


Feature

trials were rated level 2b evidence. Overall, a general risk of bias was observed (see Figure 2).

Main meta-analysis. Results of a total of 1,895 patients were reported in the 10 studies and included in the meta-analysis, including 301 patients with PUs. There was substan-tial heterogeneity among the 10 studies (χ2 (9) = 32.31, P = 0.000, I2 = 72%). The summary OR of patients using pres-sure-redistribution surfaces compared with patients used standard mattress was 0.31 (95% CI 0.17–0.59; Z = 3.59, P = 0.0003) (see Figure 3). The overall effect also shows a statis-tically significant prevention effect. The funnel plot showed symmetry, suggesting minimal publication bias in this meta-analysis (see Figure 4).

Subgroup analyses. From the five studies that used a pres-sure-reducing mattress intra-operatively, an OR of 0.59 (95%

CI 0.34–1.01; χ2 (4 ) = 8.84, P = 0.07, I2 = 55%; Z = 1.92, P = 0.05) was calculated and showed the study mattresses did not provide statistically significant prevention. From the three studies that used pressure-reducing mattress postoperatively, an OR of 0.07 (95% CI 0.01-0.49; χ2 (2)= 10.09, P = 0.006, I2 = 80%; Z = 2.68, P = 0.007) was calculated and showed the overall effect of the mattresses in preventing PUs was sta-tistically significant. For pressure-reducing mattresses used intra-operatively and postoperatively in the remaining two studies, the OR was 0.20 (95% CI 0.06–0.73; χ2 (1) = 0.41, P = 0.52, I2 = 0%; Z = 2.45, P = 0.01), showing prevention was statistically significant (see Figure 3).

Sensitivity analysis. The three quasi-randomized con-trolled trials showed the summary OR was 0.42 (95% CI 0.24–0.74; χ2 (6) = 16.82, P = 0.010, I2 = 64%; Z = 3.02, P = 0.003). From the three RCTs that used pressure-reducing mattresses intra-operatively, an OR of 0.69 (95% CI 0.36–1.32; χ2 (2) = 6.62, P = 0.04, I2 = 70%; Z = 1.12, P = 0.26) was calculated. From the two RCTs that used pressure-reducing mattress postoperatively, an OR of 0.20 (95% CI 0.10–0.40; χ2 (1) = 0.01, P = 0.94, I2 = 0%, Z = 4.50, P = 0.000) was cal-culated. None of the quasi-randomized controlled trials were removed from the analysis of pressure-reducing mattresses used intra-operatively and postoperatively.

Sensitivity analysis did not materially change the result of the main meta-analysis, which indicated sufficient evi-dence for postoperative use of pressure-redistribution sur-faces to decrease the incidence of surgery-related PUs but insufficient evidence for pressure-redistribution surface use intra-operatively.

discussionIt has been accepted that PUs are caused by three different

tissue forces: pressure, shear, and friction. These three forces play an important role in the occurrence of surgery-related PUs. Patients are immobile, causing prolonged pressure from the operating table during surgery. Shearing and friction in-jury also can occur as patients are repositioned on, then moved from, the table to transport. Additionally, some patients are not repositioned for hours or even days to accommodate use of balloon pumps or other devices postoperatively.

In this meta-analysis, use of pressure-redistribution surfaces postoperatively was found to effectively decrease the incidence of surgery-related PUs, with OR 0.07 (95% CI; 0.01–0.49; Z = 2.68, P = 0.007) compared with standard mattresses. Two sys-tematic reviews6,7 demonstrated similar results. In this meta-analysis, three included studies have investigated the pressure-reduction capabilities of support surfaces postoperatively. Hofman et al’s20 prospective, randomized, controlled clinical trial with 44 patients tested the Comfortex DeCube mattress (Comfortex, Winona, MN) against a standard hospital mattress (used in the authors’ facility) in 44 patients; at 1 week, 25% of the patients provided the study mattress and 64% of the patients provided the standard mattress had clinically relevant PUs (P =

Figure 2. Included studies risk of bias summary.

DO NOT D

UPLICATE


Feature

0.0043); at 2 weeks, the figures were 24% and 68% (P = 0.0067), respectively. Jackson et al10 reported that in 42 postoperative cardiovascular surgery patients, one out of 27 developed a PU (Stage I) while on the air-fluidized therapy bed, compared with 40 ulcers in 25 patients before the intervention (P = 0.0000). In a randomized, controlled trial with 239 postoperative hip fracture patients, Donnelly et al11 reported a significant reduction in the odds of developing a PU using a pressure-redistributing support surface, compared to standard care (eight out of 120 [7%] ver-sus 31 out of 119, [26%], respectively; P <0.001).

Although the current study did not demonstrate a sig-nificant prevention effect intra-operatively for pressure-re-distribution surfaces, sensitivity analysis also did not mate-rially change the result. Pressure-redistribution surfaces can effectively prevent damage from pressure, shear force, and friction, but particular pathogenesis for surgery-related PUs intra-operatively may play a role. It is possible that during the surgical period, anesthesia agents can depress the autonom-ic nervous system, causing enough vasodilatation to lower blood pressure and subsequently decrease tissue perfusion.21 In cardiac surgery, extracorporeal circulation will lead to va-soconstriction of peripheral blood vessels, reducing the sup-ply of blood to the tissue; hemoglobin concentration also will decrease in tissue perfusion.22 Such circumstances may help explain why a pressure-reducing mattress cannot effectively prevent surgery-related PUs intra-operatively.

Although a pressure-reducing mattress used intra-opera-tively may not have a significant prevention effect, results of ad-ditional research merit attention. In a randomized, controlled trial study, Feuchtinger et al15 found patients placed on an OR table foam overlay experienced slightly more PUs (17.6%) than patients on the standard OR table without the foam overlay (11.1%), adverse results that caused the termination of the RCT. These findings and the results of the current meta-analysis sug-gest intra-operative use of pressure-reducing mattresses for pre-venting surgery-related PUs should be implemented with cau-tion. Because the overall result of the current meta-analysis did not meet statistical significance, more well-designed, adequately powered studies are urgently needed. However, in the postop-erative subgroup and intra-operative/postoperative subgroup, meta-analysis found a significant reduction in the incidence of PUs when pressure-reducing mattresses were used.

Sensitivity analysis did not materially change the result, leading the authors to support routine use of pressure-redis-tribution surfaces during the postoperative period, if only for high-risk patients.

In terms of risk assessment, previous meta-analysis23 showed the Braden Scale was not a good instrument for risk assessment of surgery-related PUs and cannot be used alone for predict-ing PU risk in surgical patients. A national survey1 in the US of 1,128 patients found the most common types of surgery associ-ated with PU were cardiac procedures (n = 331, 29.3%), general/

Figure 3. Meta-analysis of pressure-redistribution surfaces in preventing surgery related pressure ulcers. Square = PU incidence, OR of individual study. Diamond = pooled OR of included studies. An I22>50% indicates heterogeneity.

DO NOT D

UPLICATE


Feature

thoracic procedures (n = 313, 27.7%;), orthopedic procedures (n = 232, 20.6%;), vascular procedures (n = 110, 9.8%;), head and neck (n = 50, 10.0%;), and neurologic (n = 58, 5.2%). The current meta-analysis included four cardiac surgery and two orthopedic surgery patients, persons undergoing high-risk surgical proce-dures. The national survey1 also found PUs were present in 5.8% of patients whose surgery lasted for 3 to 4 hours; the proportion of patients with intra-operative PUs increased as the surgical time exceeded 3 hours. Schouchoff’s review24 indicated long procedure time also is a risk factor for intra-operative PUs. Another prospec-tive comparative study with 286 adult patients undergoing surgi-cal treatment2 has found low American Society of Anesthesiolo-gists (ASA) or New York Heart Association (NYHA) scores, low food intake, and female gender were risk factors for surgery-relat-ed PUs. These considerations, not just Braden Scale score, should help determine patient risk for surgery-related PUs.

limitationsThis type of meta-analysis comes with inherent limitations.

First, the quality of the included studies is relatively poor. Only one RCT was rated as level 1b evidence with low risk of bias. Many studies had important methodological limitations, especially in terms of randomized sequence generation and allocation conceal-ment. Most of the included studies were rated unclear or found “potential for bias”. Second, the results of meta-analysis showed substantial heterogeneity between the studies. The heterogeneity may come from the type of pressure-reducing mattress (air-fluid-ized therapy beds, thermoactive viscoelastic foam overlays, pulsat-ing dynamic mattress, polymer overlays, foam overlays, pressure-decreasing mattresses, or others), time of reporting incidence (Day 1–2, Day 1–3, Day 1–7, Day 1–28, or others), surgical popu-lation, surgical duration (cardiac procedures, orthopedic proce-dures, and general procedures >1.5 hours, >2 hours, >3 hours), and other factors. Such limitations originate from the design of the included studies. Clearly, well-designed, large, multicenter randomized trials with the same purpose are urgently needed.

ConclusionsA meta-analysis of relevant publications shows postoperative

use of pressure- redistribution surfaces can effectively decrease the incidence of surgery-related PUs, while evidence is still not suf-ficient for routine use of these surfaces intra-operatively.

The authors suggest pressure-redistribution surfaces should be used routinely during the postoperative period for high-risk, sur-gery-related PU patients, and intra-operative use should be more judicious, pending the results of more well-designed, adequately powered, urgently needed studies. n

reference1. Aronovitch SA. Intraoperatively acquired pressure ulcer prevalence: a national

study. J Wound Ostomy Continence Nurs. 1999;26(3):130–136.2. LindgrenM,UnossonM,KrantzAM,EkAC.Pressureulcerriskfactorsinpa-

tients undergoing surgery. J Adv Nurs. 2005;50(6):605–612.3. SchoonhovenL,DefloorT,GrypdonckM.Incidenceofpressureulcersdueto

surgery. J Clin Nurs. 2002;11(4):479–487.4. GravesN,BirrellF,WhitbyM.Effectofpressureulcersonlengthofhospitalstay.

Infect Control Hosp Epidemiol.2005;26(3):293–297.5. BennettG,DealeyC,PosnettJ.ThecostofpressureulcersintheUK.Age

Ageing. 2004;33(3):230–235.6. Reddy M, Gill SS, Rochon PA. Preventing pressure ulcers: a systematic review.

JAMA. 2006 23;296(8):974–984.7. McInnes E, Jammali-Blasi A, Bell-Syer SE, Dumville JC, Cullum N. Sup-

port surfaces for pressure ulcer prevention. Cochrane Database Syst Rev. 2011;13;(4):CD001735.

8. NixonJ,McElvennyD,MasonS,BrownJ,BondS.Asequentialrandomisedcontrolled trial comparing a dry visco-elastic polymer pad and standard operat-ing table mattress in the prevention of postoperative pressure sores. Int J Nurs Stud. 1998;35:193–203.

9. SchultzA,BienM,DumondK,BrownK,MyersA.Etiologyandincidenceofpressure ulcers in surgical patients. AORN J. 1999;70(3):434–449.

10.JacksonM,McKenneyT,DrummJ,MerrickB,LeMasterT,VanGilderC.Pres-sure ulcer prevention in high-risk postoperative cardiovascular patients. Crit Care Nurse. 2011;31(4):44–53.

11.DonnellyJ,WinderJ,KernohanWG,StevensonM.AnRCTtodeterminetheeffect of a heel elevation device in pressure ulcer prevention post-hip fracture. J Wound Care. 2011;20(7):309–318.

12. Russell JA, Lichtenstein SL. Randomised controlled trial to determine the safety and efficacy of a multi-cell pulsating dynamic mattress system in the preven-tion of pressure ulcers in patients undergoing cardiovascular surgery. Ostomy Wound Manage. 2000;46(2):46–55.

13.SewchukD,PadulaC,OsborneE.Preventionandearlydetectionofpressureulcers in patients undergoing cardiac surgery. AORN J. 2006;84(1):75–96.

14.ChalianAA,KaganSH.Backsidefirstinheadandnecksurgery?:preventingpressureulcersinextendedlengthsurgeries.Head Neck. 2001;23(1):25–28.

15.FeuchtingerJ,deBieR,DassenT,HalfensRA.4-cmthermoactiveviscoelasticfoam pad on the operating room table to prevent pressure ulcer during cardiac surgery. J Clin Nurs, 2006;15(2):162–167.

16.HigginsJPT,AltmanDG.Assessing riskofbias in includedstudies. In:Co-chrane Handbook for Systematic Reviews of Interventions. Available at: www.cochrane-handbook.org. Accessed March 15, 2013.

17.HigginsJP,ThompsonSG.Quantifyingheterogeneityinameta-analysis.Stat Med. 2002;21(11):1539–1558.

18.DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188.

19.AronovitchSA,WilberM,SlezakS,MartinT,UtterD.Acomparativestudyofanalternating air mattress for the prevention of pressure ulcers in surgical patients. Ostomy Wound Manage. 1999;45(3):34–40.

20.HofmanA,GeelkerkenRH,WilleJ,HammingJJ,HermansJ,BreslauPJ.Pres-sure sores and pressure-decreasing mattresses: controlled clinical trial. Lancet. 1994;5;343(8897):568–571.

21. Walton-Geer PS. Prevention of pressure ulcers in the surgical patient. AORN J. 2009;89(3):538–548.

22.FeuchtingerJ,HalfensRJ,DassenT.Pressureulcerriskfactorsincardiacsur-gery: a review of the research literature. Heart Lung. 2005;34(6):375–385.

23.HeW,LiuP,ChenHL.TheBradenScalecannotbeusedaloneforassessingpressure ulcer risk in surgical patients: a meta-analysis. Ostomy Wound Man-age. 2012;58(2):34–40.

24.SchouchoffB.Pressureulcerdevelopment in theoperating room.Crit Care Nurs Quarter. 2002;25(1):76–82.

Figure 4. Funnel plot showed symmetry suggesting minimal publication bias in this meta-analysis.

DO NOT D

UPLICATE

do a cochrane review - o-wm.com · 36 ostomy wound management® april 2013 ar...

Documents