Australian Critical Care 34 (2021) 60e66

Contents lists available at ScienceDirect

Australian Critical Care

journal homepage: www.elsevier.com/locate/aucc

Research paper

Prevalence of pressure injuries and the management of support

surfaces (mattresses) in adult intensive care patients: A multicentre
point prevalence study in Australia and New Zealand
Elizabeth Yarad, RN, MN a, *
Anne O'Connor, RN a
Jason Meyer, RN, BN b
Matthew Tinker, RN, MN a
Serena Knowles, RN, PhD b
Yang Li, MBBS, MIPH, HScD b
Naomi E. Hammond, RN, PhD a, b, on behalf of The George Institute for Global Health and
the Australian and New Zealand Intensive Care Society
a
Malcolm Fisher Department of Intensive Care, Royal North Shore Hospital, Sydney, NSW, Australia
b
Critical Care Division, The George Institute for Global Health, Sydney, NSW, Australia

article information a b s t r a c t

Article history: Background: Pressure injuries (PIs) are a patient safety issue that impact patient outcomes. Intensive care
Received 4 October 2019 unit (ICU) patients are at high risk of PIs.
Received in revised form Objectives: To report the prevalence and classification of documented PIs in adult ICU patients, the use of
27 March 2020
pressure injury risk assessment tools, and support surface management as a part of the prevention of PIs.
Accepted 7 April 2020
Methods: This was a prospective, single-day, multicentre, cross-sectional study of patients aged
16
years admitted to adult ICUs in Australia and New Zealand (ANZ), August 2016 as part of the Australian
Keywords:
and New Zealand Intensive Care Society Clinical Trials Group (ANZICS-CTG) Point Prevalence Program.
Active mattress
Reactive mattress
Findings: Data were collected on 671 patients (58% male) in 47 ICUs. The mean [standard deviation] age
Support surfaces and weight were 60.2 years [17.2 years] and 82.1 kg [29.7 kg], respectively, with a severity of illness score
Intensive care unit (Acute Physiology and Chronic Health Evaluation [APACHE] II) of 18.2 [8.4]. PIs were reported in 10% (70/
Pressure injury 671) of patients. Patients with a PI had a mean APACHE II score of 22.5 [standard deviation; 7.7], and
Prevalence 57.1% (40/70) met the criteria for sepsis on the study day. There were 107 PIs documented on the study
Risk assessment day (N ¼ 107) in the 70 patients with nearly half of PIs present on ICU admission (46.7%; 50/107). The
sacrum was the most common location for PIs (28.9%; 31/107) and then the heels (15.9%; 17/107). All
units routinely use a risk of PI assessment tool and were cared for on an active or reactive support
surface. Patients with a PI were more often moved to an active support surface.
Conclusions: The prevalence rate was reported at 10% for PIs for adult intensive care patients on the study
day. More than half of the patients with a PI had signs of sepsis on the study day and a higher severity of
illness, and more were cared for on active support surfaces. Most PIs were located at the sacrum and then
the heels. All clinical sites routinely used a PI risk assessment tool.
Crown Copyright © 2020 Published by Elsevier Ltd on behalf of Australian College of Critical Care Nurses
Ltd. All rights reserved.

1. Background

Pressure injuries (PIs) are recognised as a preventable patient

* Corresponding author. safety issue and used as an indicator of the quality of health care.1,2
E-mail address: [email protected] (E. Yarad). A pressure injury is hypoxia of skin tissue, causing skin, fat, and

https://doi.org/10.1016/j.aucc.2020.04.153
1036-7314/Crown Copyright © 2020 Published by Elsevier Ltd on behalf of Australian College of Critical Care Nurses Ltd. All rights reserved.
 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66 61

muscle breakdown, which can ultimately lead to necrosis related to inflammatory response symptoms (SIRS) (deranged temperature,
direct pressure.3,4 The development of PIs is multifactorial and in- deranged white cell count, tachycardia, or tachypnoea) OR a
cludes patient intrinsic factors impacting skin health such as defined focus of infection without SIRS. The Sequential Organ
chronic illness, nutrition, age, weight, haemodynamic stability, Failure Assessment15 (SOFA) score for six domains (respiratory,
tissue perfusion, and oxygenation, and finally, skin temperature (or coagulation, hepatic, cardiovascular, renal, and central nervous
microclimate).3 Extrinsic patient factors impact skin tolerance and system) was calculated using the most deranged value during the
include excessive skin moisture (incontinence, wound drainage, 24-h study period. Status at day 28 for all included patients was also
perspiration) between the skin and the support surface, immobility collected.
due to therapeutic interventions, or the nature of the intensive care PI-specific data were collected for patients with a PI on the study
unit (ICU) admission and cognitive function.3 Such factors are day. The number of PIs per patient, the location of each PI, and the
omnipresent in the critically ill patient exposing them to a higher grading or severity of injury as per the NSW Health Pressure Injury
risk for developing PIs.2,3 Once developed, a PI impacts the patient's Prevention and Management Policy, Pressure Injury classification
morbidity, length of hospital stay, mobility and negatively impacts system.12 As per Appendix 1, PIs were staged I to V, unstageable, or
patient well-being.4,5 suspected deep tissue9 by clinical staff and reported in the patient
PIs are costly to both patients and the healthcare system. In medical record. Also collected was if the PI was present on
2012e13, PIs were estimated to cost the Australian healthcare admission to the ICU from any other location. For reporting, PIs are
system $983 million annually, representing 1.9% of public hospital grouped based on the broad physical locationdhead and neck
expenditure during that time period.6 The indirect cost of increased location: occiput, ear, nose, mouth, chin, cheek and neck (trache-
length of hospital stay was $820 million annually and was associ- ostomy site); torso: chest, back hip, sacrum, and groin; upper limb:
ated with close to 122,000 PI cases with more than half a million elbow, axilla, arm, and wrist; lower limb: leg, leg stump, ankle, foot,
bed days lost.6 Of these costs, a proportion is attributed to ICU toe, and heel. The support surface (mattress) data were how the
patients; however, these data are limited. An American survey of mattress was selected and if the mattress had been changed since
PIs reported prevalence data on the hospital ward with 19% (2244/ ICU admission (Appendix 1).
11,849) of patients having a PI.7 An Australian study reported the Unit-level data included the use of a PI risk assessment tool or
prevalence of hospital-acquired PIs (HAPIs) in ICU versus non-ICU not, the type of mattress available, and if a policy or guideline aided
wards in Queensland Health hospitals over a 3-y period.8 They mattress selection for ICU patients. Risk assessment tools available
found the ICU patient to be almost four times more likely to develop for selection were the Braden Scale©,16 the Cubbin & Jackson©
a HAPI (11%; 34/296) than in the non-ICU patient (3%; 210/6995).8 pressure area risk calculator,17 the Norton Scale©,18 and the
The Australian Commission on Safety and Quality in Healthcare Waterlow Score©.19 Mattress types11 were listed (Appendix 3 & 4),
included PI prevention and management in the 2011 National including a question if the mattress included microclimate
Safety & Quality Health Service Standards (NSQHSS)2,5 across all management.
public healthcare services.3,9 Standard 8 applies to all patients to Descriptive statistics were used for all clinical and demographic
prevent PIs and manage such injuries should they occur.10 Strate- data. Data are reported as mean and standard deviation for
gies include the use of a risk of PI assessment tool and prevention continuous variables or number and percentage for categorical
strategies such as selection of a supportive surface,11 to prevent and variables. No assumptions were made for missing data. The respi-
manage PIs. The reactive support surface or mattress is designed to ratory, coagulopathy, hepatic, cardiovascular, renal, and central
redistribute pressure over a wide body surface area. Support sur- nervous system components of the SOFA15 were scored from 0 to 4
faces are selected in response to the patient's condition, should not and dichotomised to report organ dysfunction or failure with scores
negate the practice of regular patient repositioning, and should be of 1e2 or 3e4, respectively.
regularly assessed for effectiveness.12 Active (or reactive) support
surfaces alternate the body surface area in contact with the 3. Results
mattress with cycling air pressure changes in air cells.12 Support
surface microclimate management is a surface that allows air to A total of 47 ICUs participated in the study day with 671 ICU
escape from air cells to manage skin heat and humidity, a recom- patients included in the prevalence survey. Of the 47 participating
mendation for the clinical care of critically ill patients.12 ICUs, 37 were in Australia and 10 were in New Zealand (Appendix
We aimed to determine the prevalence of PIs and review the A). All included ICU patient characteristics are reported in Table 1,
adoption of prevention strategies following the implementation of also separated by the presence or not of a documented PI to
NSQHSS Standard 8, specifically risk assessment tools, the use of determine any differences in characteristics between groups.
active and reactive mattresses, the use of mattresses with micro- Included patients had a mean age of 60 years (standard deviation
climate management system, and how mattresses are selected in [SD], 17.18 years), were predominantly male (58%), and were more
ANZ ICUs. commonly admitted to the ICU from the Accident and Emergency
Department (33%) and met the criteria of sepsis on the study day
2. Methods (30%). The mean APACHE II13 score was 18 [SD, 8.38] with a mean
weight of 82 kg [SD, 29.68]. On the study day, 6% of patients in the
All adult (aged
16 years) patients occupying a bed at 10 am in ICU were previously admitted to the ICU at least once during the
participating ICUs on the 23rd August 2016, or back up study day, same hospitalisation, and 87% of patients were discharged alive
the 7th September 2016, were included, and data were collected for from the ICU by 28 days after the study day (Table 1).
a 24-h study period. Routine data were collected for all patients Of the included patients, 10% had one or more documented PIs
including age, sex, weight, the Acute Physiology and Chronic Health on the study day (70/671). Patients with a PI had a mean age of 61
Evaluation (APACHE) II13 score, readmission to the ICU, APACHE III14 years [SD, 16.50], a mean body weight of 76 kg [SD, 20.36], and a
surgical or medical diagnostic code, trauma admission, and the mean APACHE II13 of 22 [SD, 7.67]. More than half of patients with a
source of admission to the ICU. Also, if during the study day, the PI met the criteria for sepsis on the study day (57%). Organ failure
patient met the criteria for acute respiratory distress syndrome or for patients with a PI, as reported by SOFA scoring,15 was cat-
sepsis. For the purposes of this study the criteria for sepsis was egorised as respiratory (31%), coagulopathy (7%), renal (10%), car-
having both a defined focus of infection and two or more systemic diovascular (37%), and central nervous system (24%) failure.
62 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66

Table 1
Patient characteristics on the study day and at day 28.

Patient characteristics All patients (N ¼ 671) Patients without a pressure injury (N ¼ 601) Patients with a pressure injury (N ¼ 70)

Age (years),a m [SD]a 60.2 [17.20] 60.1 [17.26] 61.1 [16.63]

Sex (male), n (%) 391 (58.27) 354 (58.90) 37 (52.86)
Weight (kg), m [SD] 82.1 [29.68] 82.83 [30.50] 75.84 [20.51]
Severity of illness
APACHE II,13 score m [SD] 18.24 [8.38] 17.79 [8.34] 22.09 [7.67]
Readmissions to the ICU, n (%) 41 (6.11) 34 (5.66) 7 (10.00)
Reason for ICU admission, n (%)
Surgical 238 (35.47) 226 (94.96) 12 (5.04)
Medical 433 (64.53) 375 (86.61) 58 (13.39)
Subgroup categories: n (%)
Trauma admission 58 (8.64) 53 (8.82) 5 (7.14)
Criteria met for sepsis on the day 198 (29.51) 158 (26.29) 40 (57.14)
Criteria met for ARDS on the day 20 (2.98) 16 (2.66) 4 (5.71)
Source of admission to the ICU: n (%)
Operating theatre e elective 149 (22.21) 146 (24.29) 3 (4.29)
Operating theatre e emergency 79 (11.77) 72 (11.98) 7 (10.00)
Accident & emergency 224 (33.38) 199 (33.11) 25 (35.71)
Hospital ward 129 (19.23) 109 (18.14) 20 (28.57)
Another ICU/hospital 90 (13.41) 75 (12.48) 15 (21.43)
Organ failure or dysfunction: n (%)
Respiratory
Dysfunction (SOFA15 score, 1e2) 308 (45.90) 283 (47.09) 25 (35.71)
Failure (SOFA15 score, 3e4) 127 (18.93) 105 (17.47) 22 (31.43)
Coagulopathy
Dysfunction (SOFA15 score, 1e2) 174 (25.93) 154 (25.62) 20 (28.57)
Failure (SOFA15 score, 3e4) 25 (3.73) 20 (3.33) 5 (7.14)
Hepatic
Dysfunction (SOFA15 score, 1e2) 108 (16.10) 96 (15.97) 12 (17.14)
Failure (SOFA15 score, 3e4) 25 (3.73) 24 (3.99) 1 (1.43)
Cardiovascular
Dysfunction (SOFA15 score, 1e2) 147 (21.91) 128 (21.30) 19 (27.14)
Failure (SOFA15 score, 3e4) 195 (29.06) 169 (28.12) 26 (37.14)
Renal
Dysfunction (SOFA15 score, 1e2) 152 (22.65) 128 (21.30) 24 (34.29)
Failure (SOFA15 score, 3e4) 59 (8.79) 52 (8.65) 7 (10.00)
Central nervous system
Dysfunction (SOFA15 score, 1e2) 182 (27.12) 158 (26.29) 24 (34.29)
Failure (SOFA15 score, 3e4) 106 (15.80) 94 (15.64) 17 (24.29)
Day 28 outcomes: n (%)
Remain in the ICU 27 (4.02) 22 (3.66) 5 (7.14)
Discharged from the ICU alive 584 (87.03) 531 (88.35) 53 (75.71)
Remain in the hospital ward 116 (17.29) 107 (17.80) 9 (12.86)
Discharged from the hospital alive 433 (64.53) 395 (65.72) 38 (54.29)
Mortality in the ICU or in the ward 94 (15.64) 76 (12.65) 18 (25.71)
ICU length of stay (days), m [SD] 11.98 [22.25] 11.55 [23.10] 15.66 [12.11]
Hospital length of stay (days) m [SD] 20.37 [23.52] 19.71 [24.10] 26.06 [16.65]

ARDS ¼ acute respiratory distress syndrome; ICU ¼ intensive care unit; APACHE ¼ Acute Physiology and Chronic Health Evaluation; SOFA ¼ Sequential Organ Failure
Assessment.
a
Mean (m) and standard deviation [SD]. All other values are n (%).

Patients with a PI were predominantly admitted to the ICU from the

Accident and Emergency Department (36%) and the hospital ward
(29%) (Table 1).
There were a total of 107 documented PIs for 70 patients on the
study day. The majority of patients had one PI (67%), 15 patients
(21%) had two, and eight patients had three or more PIs (Fig. 1).
Nearly half of PIs on the study day were located on the patient's
torso (43%), and over half of these (57%) were present on admission.
The sacrum was included in the torso grouping which was the most
prevalent location of all PIs (29%). The lower limb was the second
most prevalent grouped location (28%) which included the heel
(16%). Other physical locations of PI were the head and neck (22%)
and upper limb (5%). Most PIs were classified as stage I12 (44%),
stage II12 (33%), stage III12 (7%), and stage IV12 (2%). The depth could
not be visualised in 12% of PIs, and 2% were suspected deep-tissue12
injuries. The majority of stage I and II PIs were located on the
sacrum (stage I: n ¼ 11, stage II: n ¼ 14) PIs (Table 2 and Appendix Fig. 1. Pressure injury numbers per patient (N ¼ 70).
2).
 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66 63

Table 2 reported using a unit policy or guideline for selection of a mattress

Characteristics of pressure injuries present on study day (N ¼ 107) and (70%). Equally, both active and reactive support surfaces were
those also present on ICU admission (N ¼ 50).
available in the participating ICUs (51% vs 49%), and less than a third
Characteristic of PI Present on Present also on of the ICUs (30%) had microclimate management12 available in the
study day (N ¼ 107) ICU admission (N ¼ 50) support surface used (Table 4).
Location n (%)
Head and neck group 24 (22.43) 5 (20.83) 4. Discussion
Occiput 1 (0.93) 1 (100.00)
Ear 8 (7.48) 1 (12.50)
Nose 3 (2.80) 1 (33.33) 4.1. Key findings
Mouth 9 (8.41) 1 (11.11)
Chin and cheek 2 (1.87) 1 (50.00) We report on the prevalence of PIs and support surface
Tracheostomy site 1 (0.93) 0 (0.00)
(mattress) management in 47 Australian and New Zealand ICUs in
Torso group 46 (42.99) 26 (56.52)
Chest 3 (2.80) 2 (66.67) 2016. We found 10% PI prevalence in the participating adult ICUs.
Back 6 (5.61) 4 (66.67) Patients with a PI were more likely to meet the criteria of sepsis,
Hip 4 (3.74) 3 (75.00) have a higher severity of illness score, and a longer length of hos-
Sacrum 31 (28.70) 17 (54.84) pital stay. The most common location for PIs was the sacrum fol-
Groin 2 (1.87) 0 (0.00)
Upper limb group 5 (4.67) 1 (20.00)
lowed by the heel. When reviewing the grading or severity of the
Elbow 2 (1.87) 1 (0.00) PIs located at the sacrum, over half are stage II and worse PIs, while
Axilla, arm, wrist 3 (2.80) 0 (0.00) the heels are predominantly stage I PIs. Mattress type used varied,
Lower limb group 30 (28.34) 16 (53.33) with the majority using active or reactive support surfaces. All ICUs
Leg, stump 4 (3.74) 2 (50.00)
reported using a risk assessment tool for PI risk identification in
Foot, ankle, toe 9 (8.41) 4 (44.44)
Heel 17 (15.89) 11 (64.71) routine practice.
Location not recorded 2 (1.87) 2 (100.00)
Stage of PIsa n (%) 4.2. Relationship to previous studies
Stage I 47 (43.93) 19 (40.43)
Stage II 35 (32.71) 8 (22.86)
Stage III 8 (7.48) 8 (100.00) We reported a PI prevalence of 10%. The method in some studies
Stage IV 2 (1.87) 2 (100.00) is to exclude stage I PI from analysis as these are reversible.8,21 For
Unstageable: depth 13 (12.04) 11 (84.62) comparison with Queensland audits of 18 ICUs (2012e2014) with a
unknown prevalence of 11%,8 we separated patients with stage I PIs only
Suspected deep tissue 2 (1.85) 2 (100.00)
(n ¼ 30) from all patients with one or more PIs (N ¼ 70), giving a
ICU ¼ intensive care unit; PI ¼ pressure injury. prevalence rate of 6% on the study day in 2016. Both prevalence
a
Grading of PIs as per The Royal Children's Hospital Melbourne, Clinical
rates (10% with all PIs and 6% with stage I removed) falls within the
Guideline (Nursing), Pressure Injury Prevention and Management 201220.
6e18.5% prevalence found in a global systematic review of acute
care PI publications between 2010-2015.22 Chaboyer et al.21 per-
formed a systematic review and meta-analysis of ICU PI incidence
Over half (54%) of patients on the study day were on an active and prevalence data and reported when studies exclude stage I PIs,
support surface, and the remaining were cared for on a reactive prevalence was 12e16% in six studies of 11,994 patients. We cannot
surface (44%).11 More patients with a PI were on an active support compare this with our reported prevalence of 6% in 671 patients
surface (73%) as opposed to a reactive surface (26%) (Table 3). owing to sample size variation.
All participating ICUs reported the use of a risk assessment tool The impact of the NSQHSS Standard 8 which was implemented
for PIs as a part of patient care. The majority used either the in Australian ICUs in 2012 has been reviewed using a follow-up
Waterlow Score19 (53%) or the Braden Scale©16 (43%). Most ICUs Queensland bedside audit of all hospital patients (ICU and non-ICU)

Table 3
Type of support surface (mattress) and selection on study day.

Support surface characteristics All patients, (N ¼ 671) Patients without a pressure injury (N ¼ 601) Patients with a pressure injury (N ¼ 70)

Mattress type on the study day, n (%)

Active/alternating/dynamic 361 (53.80) 310 (51.58) 51 (72.86)
Reactive/constant low-pressure 297 (44.26) 279 (46.42) 18 (25.71)
Other: Normal ward bed mattress 6 (0.89) 5 (0.83) 1 (1.43)
Unknown 7 (1.04) 7 (1.16) e
How the current mattress was selected n (%)
By pressure injury risk score 134 (19.97) 107 (17.80) 27 (38.57)
Next available 208 (31.0) 196 (32.61) 12 (17.14)
Nurse unit manager 7 (1.04) 7 (1.16) 0 (0.00)
Team leader 60 (8.94) 53 (8.82) 7 (10.00)
Unit policy 203 (30.25) 185 (30.78) 18 (25.71)
Bedside registered nurse 13 (1.94) 12 (2.00) 1 (1.43)
Other ICU mattress 16 (2.38) 16 (2.66) 0 (0.00)
Standard ward bed mattress 3 (0.45) 2 (0.33) 1 1.43)
Clinical condition 3 (0.45) 2 (0.33) 1 (1.43)
Unknown 24 (3.58) 0 (0.00) 1 (1.43)
Was the study day mattress changed since ICU admission n (%)
Yes 117 (17.44) 92 (15.31) 25 (35.71%)
No 512 (76.30) 471 (70.91) 41 (58.57%)
Unknown 42 (6.26) 38 (6.32) 4 (5.71%)

ICU ¼ intensive care unit; PI ¼ pressure injury.

64 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66

Table 4 guidelines of PI risk assessment and management in the partici-

Pressure injury risk assessment tool, mattress type, and policy/guideline per pating ICUs.
participating intensive care unit.
Limitations of our study include the lack of longitudinal data.
Risk assessment tools, mattress type and policies per ICU N ¼ 47 Data were collected over a 24-h period which neither allows for the
PI risk assessment toola n (%) cumulative incidence of pressure injuries over an extended period
Braden©16 20 (42.55) of time nor accounts for seasonal variation. Owing to the study
Cubbin & Jackson©17 0 design, we did not collect detailed information regarding specifics
Norton©18 0
to determine if a PI was present on hospital admission or a HAPI.
Waterlow©19 25 (53.19)
Other tools 3 (6.34) We did not ask for the number of days from ICU admission to when
Mattress type n (%) the PI was first reported or if the PI was identified after hospital
Reactive or constant low-pressure support surface 23 (48.94) admission for those PIs identified in patients who were transferred
Active or alternating or dynamic support surface 24 (51.06)
from another ICU or hospital ward. We did not collect the risk of PI
Mattress microclimate management available 14 (29.79)
Policy/guideline for selection of mattress 33 (70.21)
score for each patient on the study day, the expertise of the clinician
How mattress is selected as stated in policy/guidelinea n (%) reporting risk assessment, or the stage of PI or if the PI was due to
Next available 12 (25.53) any identifiable contributing clinical factors (such as securing of
Team leader 22 (46.81) tubes, clinical condition, administration of inotropes, and nutri-
Nurse unit manager 10 (21.28)
tional status), nor other possible preventative interventions which
Other: Bedside registered nurse 3 (6.38)
Clinical condition 2 (4.26) would provide further understanding of PI prevention. Data were
a
not collected regarding the location of a PI and impact the support
Multiple answers allowed.
surface might have, such that mattress selection will have no
impact on a PI of the mouth or tracheostomy site. This study is
in 2015.2 PI prevalence reduced compared with 2011 audit results designed to document current practice and quality of care to inform
(12%e7.3%), and reduction in hospital-acquired PIs was also re- targeted questions for future research regarding PI prevention in
ported (7.9%e4.1%).2 These findings are encouraging but report all the ICU with a number of strategies and interventions including
hospital-acquired PIs and not ICU-specific PIs. mattress choice.
The Australian Council on Healthcare Standards (ACHS)23
routinely reports clinical indicators relating to the provision of 4.4. Clinical implications
health services from data submitted by contributing health orga-
nisations to watch trends. The 2018 ACHS report23 of hospital-wide This point prevalence study was conducted to determine a
inpatients with one or more PIs showed no variation in rates from snapshot of current clinical care regarding the use of support sur-
2010 to 2017 and the 2017 prevalence of 0.017 inpatients per 100 faces and the prevalence of PIs to inform future research. Partici-
bed days. Incidence and prevalence data regarding ICU-specific PIs pating ANZ ICUs use regular and routine patient risk assessment for
are limited. PI and support surface selection in response to the documented risk
The most common location of PI in our study was the sacrum assessments. There are patient factors that cannot be changed
and heel. A systematic review by Chaboyer et al.21 reports the before the ICU such as the severity of illness, immobilisation due to
routine use of skin inspection to identify PIs, with the most com- injury or critical illness, nutrition, and body habitus. However, pa-
mon location of PI being the sacrum, followed by the heels. These tients with signs and symptoms of sepsis and a high admission
findings are consistent with ours, with Chaboyer et al.21 concluding APACHE II13 score are at the greatest risk of having or developing a
that these locations were likely to be related to the frequent use of PI in the ICU. Recognition of at-risk patients and early PI prevention
the supine position used during clinical interventions in the ICU. strategies are highly recommended.3,5
Chaboyer et al.21 also discuss the need for multiple interventions We found the majority of ear, mouth, and nose PIs were iden-
for PI prevention, of which mattress selection is a part. tified after ICU admission, perhaps related to medical device
In respect to support surface management, the NSQHSS Stan- securing systems which are considered a unique type of PI.4
dard 8 recommends the use of constant low-pressure redistribu- Nursing staff should be mindful of the higher risk for PI with
tion support surfaces,11 and the benefit of one type of surface over medical devices and have regular checks scheduled in their daily
another to prevent PI is yet to be evaluated.21 The choice of support plan. During the ICU stay, clinical staff members can optimise
surface in our study was guideline based in 70% of the contributing nutrition and movement in bed, select an appropriate support
ICUs and primarily chosen by the nursing team leader followed by surface, and use multifaceted PI prevention plans to minimise the
the next mattress available for use. Support surfaces were reactive development of PIs.9,10,21
(or constant low-pressure) or active (alternating or dynamic), and
one in three beds had microclimate management protecting the 4.5. Future directions
patient's skin from trapped heat and moisture and recommended
for the critically ill.12 Patients with a documented PI were more Further research is required to determine how best to prevent
likely to be on an active support surface, and clinical staff changed the development of hospital-acquired PIs. Immediate recognition
mattresses in response to risk assessment. of those most at risk, optimisation of nutrition, increased in bed
movement, and reduced sedation practices are all possible areas
impacting PI development and require further investigation.12 The
4.3. Strengths and limitations data from this study help us understand current practices, which
will aid ongoing guideline development and future research to
Our study has several strengths including that it provides cur- inform the use of pressure-relieving support surfaces to prevent
rent, prospective, multicentre, binational data on pressure injury and manage PIs in the critically ill patient. Further evidence is
and support surfaces in use in adult ICUs. Nearly a quarter of all required regarding the preferred supportive surface and the impact
ICUs in Australia and New Zealand in 2016 contributed data (24%; of the NSQHSS Standard 8 where Australian States have imple-
47/197).24 We have described key aspects of PI prevalence, detailed mented public programs such as financial investment for appro-
descriptions and location of PIs, use of support surfaces, and priate pressure-relieving mattresses in the ICU.2
 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66 65

5. Conclusion results. Yang Li: Analysed the results. Naomi E. Hammond:

Designed the study, Operationalised the study, Analysed the re-
The prevalence rate was reported at 10% for PIs for adult sults, Wrote the manuscript with input and review from all authors.
intensive care patients on the study day. More than half of the
patients with a PI had met the criteria for sepsis on the study day Acknowledgements
and higher severity of illness and experienced a longer ICU and
hospital stay. Routine use of PI risk assessment tools occurs in all The authors thank the intensive care staff members of the 47
ICUs surveyed with half of PIs identified after ICU admission. listed hospitals, site investigators, and research staff members as
Further research is required to provide effective, evidence-based, PI listed in Appendix A.
preventative management of adult patients in ANZ ICUs, targeting
patients at the highest risk of PI. Appendix A. Supplementary data

Ethical approval Supplementary data to this article can be found online at

https://doi.org/10.1016/j.aucc.2020.04.153.
The Australian and New Zealand Intensive Care Society (ANZICS)
Clinical Trials Group (CTG) invited affiliated adult ICUs to partici- Appendix 1
pate in the prospective, observational ANZICS Point Prevalence
Program (PPP).25 Human Research and Ethics Committee approval Pressure injury classification system Appendix 10.4 of the NSW
for all relevant jurisdictions was obtained with a waiver of consent Health Policy Directive: Pressure Injury Prevention and Manage-
approved owing to the observational nature of the study (NSW ment PD2014_007.9
HREC approval reference: LNR/14/RPAH/235). Individual Research https://www1.health.nsw.gov.au/pds/ActivePDSDocuments/
Governance Office approval was obtained by participating sites PD2014_007.pdf.
where required.
Appendix 2. Pressure injury location by pressure injury stage
Funding
Location Stage I Stage II Stage III Stage IV Unstageable Suspected
This study was funded by The George Institute for Global Health deep
and the Australian and New Zealand Intensive Care Foundation Sacrum 11 14 2 1 3 0
with donations provided to included sites. Heel 9 2 1 0 5 0
Mouth 4 4 0 0 1 0
Foot, ankle, toe 4 2 0 0 2 1
Conflict of interest

None declared.

CRediT authorship contribution statement

Elizabeth Yarad: Designed the study, Analysed the results,

Wrote the manuscript with input and review from all authors. Appendix 3
Anne O'Connor: Designed the study. Jason Meyer: Operationalised
the study, Analysed the results. Matthew Tinker: Designed the The Royal Children's Hospital Melbourne, Clinical Guideline
study. Serena Knowles: Operationalised the study, Analysed the (Nursing), Pressure Injury Prevention and Management 201220.
66 E. Yarad et al. / Australian Critical Care 34 (2021) 60e66

Appendix 4 [11] NSQHS standard 8 pressure injury. Definitions sheet 2 V3.0 [Internet]. 2014
Sept 17. cited 2018 Oct 29]. Available from: https://www.health.qld.gov.au/__
data/assets/pdf_file/0029/433478/pip-audit-def.pdf.
Examples of pressure redistribution devices to assist in identi- [12] Australian Wound Management Association. Pan Pacific Clinical Practice
fying the type of mattress/support surface used11. Guideline for the prevention and management of pressure injury. WA:
https://www.health.qld.gov.au/__data/assets/pdf_file/0029/ Cambridge Media Osborne Park; 2012.
[13] Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Apache II: a severity of
433478/pip-audit-def.pdf. disease classification system. Crit Care Med 1985;13:818e29.
[14] Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, et al.
References The Apache III prognostic system: risk prediction of hospital mortality for
critically III hospitalized adults. Chest 1991;100(6):1619e36.
[15] Vincent JL, de Mendonca A, Cantraine F, Moreno R, Takala J, Suter PM, et al.
[1] Thomas DR. Prevention and treatment of pressure ulcers. JAMDA 2006;7: Working Group on sepsis-related problems of the European society of
46e59.
intensive care medicine. The SOFA (Sepsis-related organ failure assessment)
[2] The impact of the National Safety & Quality Health Service Standards. score to describe organ dysfunction/failure. Intensive Care Med 1996;22:
Australian commission on safety and quality in health care [Internet]. 2018 707e10.
May [cited 2018 Nov 29]. Available from: https://www.safetyandquality.gov. [16] Bergstrom N, Braden B, Laquzza A, Holman V. The Braden Scale for predicting
au/wp-content/uploads/2018/05/NSQHS-Impact-Report.pdf. pressure sore risk. Nurs Res 1987;36(4):205e10.
[3] Rolls K, Jones S, Power D, Cook S, Flood K, Hennessy C, et al. Pressure injury
[17] Jackson C. The revised Jackson/Cubbin pressure area risk calculator. Intensive
prevention for critically ill adults: a clinical practice guideline (SHPN ACI Crit Care Nurs 1999;15:169e75.
140009). Agency for Clinical Innovation Version 1.1 [Internet]. 2014 [cited
[18] Norton D, McLaren R, Exton-Smith AN. An investigation of geriatric nursing
2018 Nov 29]. Available from: https://www.aci.health.nsw.gov.au/networks/ problems in hospital [Internet]. 1962 [cited 2019 March 18]. Available from:
icnsw/intensive-care-manual/statewide-guidelines/pip. https://trove.nla.gov.au/version/25658898.
[4] National Pressure Ulcer Advisory Panel. European pressure ulcer advisory
[19] Waterlow J. The Waterlow score [Internet]. 1984 [cited 2019 March 18].
Panel and Pan pacific pressure injury alliance. Prevention and treatment of Available from: http://www.judy-waterlow.co.uk/waterlow_score.htm.
pressure ulcers: clinical practice guideline. Osborne park. Western Australia:
[20] The Royal Children's hospital Melbourne, clinical guideline (nursing), pressure
Cambridge Media; 2014. injury prevention and management [Internet]. 2012 [cited 2019 March 18].
[5] Australian commission on safety and quality in health care. National safety Available from: https://www.rch.org.au/rchcpg/hospital_clinical_guideline_
and quality health service standards [Internet]. 2012 September [cited 2018 index/Pressure_Injury_Prevention_and_Management/.
Nov 29]. Available from: https://www.safetyandquality.gov.au/wp-content/ [21] Chaboyer WP, Thalib L, Harbeck EL, Coyer FM, Blot S, Bull CF, et al. Incidence
uploads/2011/09/NSQHS-Standards-Sept-2012.pdf.
and prevalence of pressure injuries in adult intensive care patients: a sys-
[6] Nguyen KH, Chaboyer WP, Whitty JA. Pressure injury in Australian public tematic review and meta-analysis. Crit Care Med 2018;46(11):e1074e81.
hospitals: a cost-of-illness study. Aust Health Rev 2015;39:329e36. [22] Tubaishat A, Papanikolaou P, Anthony D, Habiballah L. Pressure ulcers prev-
[7] VanGilder C, Amlung S, Harrison P, Meyer S. Results of the 2008-2009 inter- alence in the acute care setting: a systematic review, 2000-2015. Clin Nurs Res
national pressure ulcer prevalence survey and a 3-year, acute care, unit- 27 April 2017:643e59.
specific analysis. Ostomy Wound Manag 2009;55:39e45.
[23] Australian council on healthcare standards (ACHS). Australasian clinical in-
[8] Coyer F, Miles S, Gosley S, Fulbrook P, Sketcher-Baker K, Cook J-L, et al. dicator report: 2010e2017. 19th ed. Sydney, Australia: ACHS; 2018.
Pressure injury prevalence in intensive care versus non-intensive care pa-
[24] The George Institute for Global Health. Point prevalence program [Internet].
tients: a state-wide comparison. Aust Crit Care 2017;30:244e250r. 2019 [cited 2018 Oct 30]. Available from: https://www.georgeinstitute.org/
[9] NSW Health Policy Directive. Pressure injury prevention and management projects/point-prevalence-program.
PD2014_007 [Internet]. 2014 Mar 24 [cited 2018 Jan 30]. Available from: [25] Thompson K, Hammond K, Eastwood G, Festa M, Glass P, Rajbhandari D, et al.
https://www1.health.nsw.gov.au/pds/ActivePDSDocuments/PD2014_007.pdf. The Australian and New Zealand intensive care society clinical Trials Group
[10] National Safety & Quality Health Service Standards. Standard 8 preventing
point prevalence program, 2009-2016. Crit Care Resuscit 2017;19(1):88e93.
and managing pressure injuries [Internet]. 2012 [cited 2018 Jan 30]. Available
from: https://www.safetyandquality.gov.au/wp-content/uploads/2012/10/
Standard8_Oct_2012_WEB.pdf.